2011 69869 Brazil Low Carbon Case study Technical Synthesis Report Land Use, Land-Use Change, and Forestry Coordination Christophe de Gouvello, The World Bank Britaldo S. Soares Filho, CSR-UFMG André Nassar, ICONE Authors Britaldo S. Soares Filho and Letícia Hissa, UFMG André Nassar, Leila Harfuch, Marcelo Melo Ramalho Moreira, Luciane Chiodi Bachion and Laura Barcellos Antoniazzi, ICONE Luis G. Barioni, Geraldo Martha Junior, Roberto D. Sainz, Bruno J. R. Alves, and Magda A. de Lima, EMBRAPA Osvaldo Martins, Magno Castelo Branco, and Renato Toledo, Iniciativa Verde Manoel Regis Lima Verde Leal, CENEA Fábio Marques, Rodrigo Ferreira, Luiz Goulart, and Thiago Mendes, PLANTAR Christophe de Gouvello, Adriana Moreira, Barbara Farinelli, Jennifer Meihuy WORLD BANK Chang, and Rogerio Pinto, The World Bank Júlio Hato and Sérgio Pacca, USP Saulo Ribeiro Freitas, Karla Maria Longo and Ricardo Almeida de Siqueira (National Institute for Space Research, INPE) 2011 Brazil Low Carbon Case study Technical Synthesis Report Land Use, Land-Use Change, and Forestry Coordination Christophe de Gouvello, The World Bank Britaldo S. Soares Filho, CSR-UFMG André Nassar, ICONE Authors Britaldo S. Soares Filho and Letícia Hissa, UFMG André Nassar, Leila Harfuch, Marcelo Melo Ramalho Moreira, Luciane Chiodi Bachion and Laura Barcellos Antoniazzi, ICONE Luis G. Barioni, Geraldo Martha Junior, Roberto D. Sainz, Bruno J. R. Alves, and Magda A. de Lima, EMBRAPA Osvaldo Martins, Magno Castelo Branco, and Renato Toledo, Iniciativa Verde Manoel Regis Lima Verde Leal, CENEA Fábio Marques, Rodrigo Ferreira, Luiz Goulart, and Thiago Mendes, PLANTAR Christophe de Gouvello, Adriana Moreira, Barbara Farinelli, Jennifer Meihuy Chang, and Rogerio Pinto, The World Bank Júlio Hato and Sérgio Pacca, USP Saulo Ribeiro Freitas, Karla Maria Longo and Ricardo Almeida de Siqueira (National Institute for Space Research, INPE) © 2011 The International Bank for Reconstruction and Development / The World Bank Washington dC 20433 telephone: 202-473-1000 Internet: www.worldbank.org Email: feedback@worldbank.org All rights reserved 5 This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank. the findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the executive directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work and accepts no res- ponsibility whatsoever for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions the material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruc- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry tion and Development / The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete in- formation to the Copyright Clearance Center Inc., 222 rosewood drive, danvers, Ma 01923, Usa; tele- phone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com. all other queries on rights and licenses, including subsidiary rights, should be addressed to the office of the Publisher, the World Bank, 1818 h street, nW, Washington, dC, 20433, Usa; fax: 202-522-2422; email: pubrights@worldbank.org. the energy sector Management assistance Program (esMaP) is a global knowledge and technical assistance program administered by the World Bank that assists low- and middle-income countries to increase know how and institutional capacity to achieve environmentally sustainable energy solutions for poverty reduction and economic growth. For more information on the Low Carbon growth Country studies Program or about esMaP’s climate change work, please visit us at www.esmap.org or write to us at: energy sector Management assistance Program The World Bank 1818 h street, nW Washington, dC 20433 Usa email: esmap@worldbank.org web: www.esmap.org Contents Tables ------------------------------------------------------------------------------ 9 Figures ---------------------------------------------------------------------------- 12 6 Maps ---------------------------------------------------------------------------- 15 acronyms and abbreviations ---------------------------------------------------------------------------- 17 acknowlegments ---------------------------------------------------------------------------- 23 executive summary ---------------------------------------------------------------------------- 24 study Overview ---------------------------------------------------------------------------- 25 1.1 Context of the Low-carbon study------------------------------------------------------------------------ 36 1. Introduction ---------------------------------------------------------------------------- 34 1.2 approach of the LULUCF summary report ------------------------------------------------------------ 37 2.1 emissions from Land Use, Land-use Change, deforestation, agriculture and Livestock ------ 38 2. Reference scenario ---------------------------------------------------------------------------- 38 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2.1.1 effects of Land Use and Land-use Change on emissions -------------------------------------- 38 2.1.1.1 Deforestation ------------------------------------------------------------------------------ 38 2.1.1.2 Agricultural Production -------------------------------------------------------------------------- 39 2.1.1.3 Livestock activities ------------------------------------------------------------------------------ 39 2.1.1.4 Forestry-based Carbon Uptake------------------------------------------------------------------ 40 2.1.2 Land Use and Land-use Change simulation Methodology ---------------------------------------- 40 2.1.2.1 area available for the expansion of Productive activities ---------------------------------- 40 2.1.2.2 economic Land-use, agriculture and Livestock Modeling: BLUM Model ---------------- 43 2.1.2.3 allocation of area for agriculture and Livestock activities--------------------------------- 49 2.1.3 Land-use reference scenario-------------------------------------------------------------------------- 53 2.1.3.1 division into geographic Micro-regions ------------------------------------------------------- 63 2.1.3.2 spatialization of Land-use Change and deforestation: sIMBrasIL Model -------------- 64 2.1.4 Calculation of emissions associated with Land use, Land-use Change and deforestation in the reference scenario ---------------------------------------------------- 68 2.1.4.1 emissions from Livestock ---------------------------------------------------------------------------- 69 2.1.4.1.1 Methodology ------------------------------------------------------------------------------ 70 2.1.4.1.2 reference scenario results -------------------------------------------------------------------- 79 2.1.4.2 agricultural emissions ------------------------------------------------------------------------------ 81 2.1.4.2.1 evaluation of Co2 emissions from Changes in soil C stocks ----------------------------- 81 2.1.4.2.2 greenhouse gas Production from the Use of Fossil energy ------------------------------ 90 2.1.4.2.3 synthesis of emissions from agricultural activities -------------------------------------- 91 2.1.5 emissions from deforestation ------------------------------------------------------------------------ 93 2.2 Carbon Uptake through reforestation ---------------------------------------------------------------- 98 2.2.1 Methodology ------------------------------------------------------------------------------ 99 2.2.1.1 details of the Potential Biomass Model -------------------------------------------------------- 99 2.2.1.2 Carbon Uptake Potential through the restoration of the Legal reserves --------------- 113 2.2.1.3 Carbon Uptake Potential through the restoration of riverside Forests ---------------- 114 2.2.1.4 Carbon Uptake Potential through energy Forest Plantations in the Cerrado and atlantic Forest Biomes ------------------------------------------------- 115 2.2.2 reference scenario for Forest restoration ----------------------------------------------------- 116 2.2.3 non-renewable Charcoal and Planted Forests for renewable Charcoal ------------------ 117 7 2.3 reference-scenario emissions results --------------------------------------------------------------- 120 3.1 Mitigation options in agriculture: Zero tillage ------------------------------------------------------ 122 3. Mitigation and Carbon Uptake Options ----------------------------------------------------------------122 3.1.1 emissions reduction Potential associated with Zero tillage ------------------------------- 125 3.1.2 obstacles Limiting the expansion of Zero tillage --------------------------------------------- 127 3.1.3 Proposals for overcoming obstacles ------------------------------------------------------------ 128 3.2 Carbon Uptake through the Increase of Planted Forests for renewable Charcoal ------------- 129 3.2.1 Carbon Uptake Potential associated with the Increase in renewable Charcoal Production ----------------------------------------------------------------------------- 129 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 3.2.2 obstacles to the expansion of Production Forests for renewable Charcoal -------------- 134 3.2.3 Measures for overcoming obstacles ------------------------------------------------------------ 137 3.3 Carbon Uptake through native Forest recovery----------------------------------------------------- 141 3.3.1 Carbon Uptake Potential resulting from a“Legal scenario� for Forest restoration ----- 141 3.3.2 obstacles to Forest restoration and Ways to overcome them------------------------------ 144 3.3.3 reforestation support Policies ------------------------------------------------------------------- 146 3.4 Mitigation options for Livestock activities ----------------------------------------------------------- 152 3.4.1 Main options Considered for Mitigating emissions from Livestock ----------------------- 152 3.4.2 obstacles and Proposals for overcoming them ----------------------------------------------- 155 3.5 reduction of emissions from deforestation --------------------------------------------------------- 156 4.1 additional needs for Land for Carbon Uptake activities and Biofuel export ------------------- 159 4. Low-Carbon Land-Use scenario in Brazil -------------------------------------------------------------159 4.2 toward a new Pattern of Productivity for the Livestock Industry -------------------------------- 160 4.3 Mitigation Potential of direct emissions from Livestock in the Low-carbon scenario -------- 164 4.4 a new Land-use scenario for the Low-carbon scenario-------------------------------------------- 169 4.5 reduction of deforestation in the Low-carbon scenario ------------------------------------------- 176 4.6 additional Measures for Protecting the Forest from deforestation ------------------------------ 179 4.7 Balance of emissions from land use and land-use change in the Low-carbon scenario ------- 186 4.8 Key Uncertainties for emissions estimates ---------------------------------------------------------- 191 4.9 Benefits related to reducing aerosol emissions resulting from deforestation by Burning 193 4.9.1 Methodology: numerical Modeling with CCatt-BraMs ------------------------------------ 194 4.9.1.1. Calculation of aerosol emissions ------------------------------------------------------------- 196 4.9.1.2 aerosol emissions in the reference and Low-carbon scenarios ------------------------- 199 4.9.2 results ----------------------------------------------------------------------------- 202 4.9.2.1 Precipitation ----------------------------------------------------------------------------- 202 4.9.9.2 temperature ----------------------------------------------------------------------------- 206 4.9.3 summary of the reduction of Impacts on rainfall and temperature Regimes in the Low-carbon Scenario ---------------------------------------------------------------------- 208 5.1 Costs of reducing emissions from deforestation --------------------------------------------------- 218 5. analysis of Transition Costs from the Reference scenario to the Low-carbon scenario ---211 5.1.1 Improving Livestock Productivity --------------------------------------------------------------- 218 5.1.2 Forest Protection ----------------------------------------------------------------------------- 220 8 5.2 Forest recovery: Legal Forest reserves --------------------------------------------------------------- 226 5.3 renewable Charcoal ----------------------------------------------------------------------------- 229 5.4 emissions abatement with Zero tillage -------------------------------------------------------------- 233 6. Conclusion ----------------------------------------------------------------------------241 7. 1 herd optimization scenario ---------------------------------------------------------------------------- 247 7. annex a: analysis of Low-carbon scenarios ----------------------------------------------------------244 7.2 Production Forest scenario ----------------------------------------------------------------------------- 250 7.3 ethanol scenario and Production Forests ------------------------------------------------------------ 251 7.4 Legal scenario (reforestation of the Legal reserve) ------------------------------------------------ 253 7.5 aggregate scenario: herd, Production Forests, ethanol, Forest restoration ------------------- 256 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 8. References ----------------------------------------------------------------------------264 Tables Table 1: Summary of additional land needs in the reference and Low-carbon Scenarios. ----------------30 table 2: Comparison of emissions distribution among sectors in the reference and Low-carbon scenarios, 2008-30------------------------------------------------------------------------------------31 9 table 3: Comparison between total pasture area and area of residual vegetation convertible into far- mland/forests in the regions of the BLUM Model (1000 ha) ---------------------------------------------------43 table 4: Brazil - area allocated and production of products covered by the BLUM Model ----------------46 table 5: data sources -------------------------------------------------------------------------------------------------47 table 6: Macroeconomic projections ------------------------------------------------------------------------------49 table 7: Land competition matrix in Brazil -----------------------------------------------------------------------50 table 8: Projection of areas occupied by production forests ---------------------------------------------------55 table 9: Productive land use (crops, pasture and forests) in the different regions of Brazil ------------- 55 table 10: Land use (1000 ha) in the six regions of the model for the reference scenario -----------------56 Table 11: Dairy cattle herd - Reference Scenario ----------------------------------------------------------------57 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry table 12: Land use for Brazil - reference scenario---------------------------------------------------------------59 table 13: description of the base developed for the implementation of sIMBrasIL ----------------------65 table 14: Categories of animals considered in the analysis of livestock emissions ------------------------72 table 15: Zootechnical coefficients considered for each productive system --------------------------------77 table 16: greenhouse gas emissions per animal and per carcass equivalent in kg in different production systems ---------------------------------------------------------------------------------------------------78 table 17: estimates of area, herd, proportion of the herd in productive systems and emissions for the Reference Scenario ---------------------------------------------------------------------------------------------------80 table 18: areas under different uses and total area in 1990, by state-----------------------------------------82 table 19: soil C stock under native vegetation for each region of the Blum model -------------------------86 table 20: Change factors for soil C stock for each type of land use --------------------------------------------88 table 21: Co2, Ch4, and n2o accumulated emissions from emissions from agriculture from 2010- 2030, expressed in Co2eq for the reference scenario ---------------------------------------------------------- 91 table 22: Land use in Brazil between 1990 and 2005 -----------------------------------------------------------94 table 23risk of extinction of arboreal forest species in Brazil in 2000 ---------------------------------------95 table 24: Points of the different IBge fertility classes --------------------------------------------------------- 104 table 25: entries in the environmental database. ------------------------------------------------------------- 107 table 26: Projection of Co2 uptake in the reference scenario (use of coal and/or non-renewable/re- newable charcoal) - 2010 to 2030 -------------------------------------------------------------------------------- 120 table 27: Methane emissions from convertional planting and zero tillage in irigated rice areas in and a different location, comparison of emissions reductions between the two uses --------------------------123 table 28: Cumulative costs and revenue in the reference and Low-carbon scenarios with the adoption of zero tillage from 2010 to 2030 --------------------------------------------------------------------------------- 125 table 29: greenhouse gases produced in the Low-carbon scenario: adoption of zero tillage in 100% of the agricultural area from 2015 to 2030 ------------------------------------------------------------------------ 125 table 30: Co2 uptake from forest plantations for renewable charcoal scenario 1 ----------------------- 132 table 31: Co2 uptake from forest plantations for renewable charcoal scenario 2 ----------------------- 133 table 32: Measures proposed to surmount obstacles---------------------------------------------------------138 table 33: area needed for reforestation under Brazil’s Legal reserve Law, by state---------------------143 table 34: average productivity of selected crops in different countries, 2008 --------------------------- 157 table 35: Mitigation and carbon uptake options for a Low-carbon scenario and associated needs for 10 additional land -------------------------------------------------------------------------------------------------------160 table 36: Comparison of land-use results for the reference and Low-carbon scenarios (millions of ha) ------------------------------------------------------------------------------------------------------- 170 table 37: snapshot of protected areas in the amazon Biome and arPa participation ------------------180 table 38: resources of the InPe for monitoring the amazon by satellite ---------------------------------- 182 table 39: Implementation of the Public Forest Management systems: Benefits and Losses ----------- 184 table 40: summary of expenditure anticipated for Public Forest Management services in 2009 -----185 table 41: Comparison of cumulative emissions distribution among sectors in the reference and Low- -carbon scenarios, 2010-30. --------------------------------------------------------------------------------------189 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry table 42: emissions factors (g/kg) for different biomes for Co2 and aerosols (particulate matter with a diameter less than 2.5 micrometers - PM2.5) ------------------------------------------------------------------ 197 table 43: total annual aerosol emissions (tons per hectare and per year) throughout the country for the reference (reF) and low carbon (LC) scenarios. also shown are the figures of absolute differences (LC-reF) and differences in percentages (LC-reF (%)) between emissions for the two scenarios -- 200 table 44: Mitigation potential and marginal abatement cost of various alternatives, based on three dis- count rates ------------------------------------------------------------------------------------------------------------ 214 table 45: Comparison of sector benchmark Irrs and break-even carbon prices for various mitigation options ---------------------------------------------------------------------------------------------------------------- 216 table 46: Volume of incentive required (undiscounted) in order to achieve the emissions reductions considered in the Low-carbon scenario from 2010 to 2030 ------------------------------------------------- 217 table 47: Investments and expenditures for prototypical livestock systems (2009-30) --------------- 219 table 48: economic and financial performance of prototypical livestock systems (2009-2030) ----- 219 table 49: Investment and expenses in the reference and Low-carbon scenarios ------------------------ 220 table 50: Comparable economic and financial performance in the livestock sector. -------------------- 220 table 51: Projection of costs for forest protection in areas where deforestation is illegal (in millions of Us$). ------------------------------------------------------------------------------------------------- 222 table 52: Livestock-sector investments and expenses to release land to absorb additional lands needed in the reference and Low-carbon scenarios (2010-30)-------------------------------------------------------226 table 53: Marginal abatement cost ------------------------------------------------------------------------------- 230 table 54: summary of economic parameters for the 2010-2030 period ---------------------------------- 231 table 55: Investments in the additional use of renewable charcoal compared with total mitigation mea- sures in the Brazilian industrial sector, considering the adjusted potential ------------------------------231 table 56: hypotheses of the technical-economic analysis --------------------------------------------------- 232 table 57: discrimination of costs considered in the study --------------------------------------------------- 239 table 58: emissions reduction potential in tons of Co2eq, average abatement cost during the period and price to be paid per ton of C to compensate the implementation of zero tillage --------------------- 240 table 59: relationship of the Low-carbon scenarios developed for this study. --------------------------- 245 table 60: area necessary for the reforestation of the legal reserve by state in Brazil (hectares) ------ 247 table 61: Balance of supply and demand for selected products, herd optimization scenario ---------- 248 table 62: Land use in Brazil, herd optimization scenario (1000 ha) ----------------------------------------249 table 63: regional allocation of pasture areas, reference and herd optimization scenario (thousand hectares). -------------------------------------------------------------------------------------------------------------249 11 table 64: regional distribution of cattle herd, reference scenario and herd optimization scenario (100 head) ------------------------------------------------------------------------------------------------------------250 table 65: regional distribution of the production forest in the reference and production forest scena- rios (thousand hectares) -------------------------------------------------------------------------------------------251 table 66: Land use in Brazil, ethanol scenario (in 1000 hectares) ------------------------------------------252 table 67: regional sugar cane distribution in the reference scenario, the herd optimization scenario and the ethanol scenario (in thousand hectares) --------------------------------------------------------------253 table 68: reforestation needs in order to comply with the Legal reserve in the regions of the model (1000 ha) -------------------------------------------------------------------------------------------------------------254 table 69: Pasture area in the regions of the model in 2009 and 2030 (in 1000 ha), in the reforestation Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Scenario of the LR. --------------------------------------------------------------------------------------------------- 254 table 70: Presentation of quantitative results by state for the atlantic Forest and Cerrado ------------256 table 71: Comparison of land use results in all scenarios for Brazil. ----------------------------------------258 table 72: Comparison of results for pasture area in all scenarios for Brazil and regions. ---------------259 table 73: results from the cattle herd in the reference scenarios and herd optimization scenarios and aggregate (1000 head).---------------------------------------------------------------------------------------------260 table 74: results for land use abd herd for selected products inhte aggregates scenario---------------261 Figures Figure 1: ghg mitigation wedges in the Low-carbon scenario, 2008-30. -----------------------------------32 Figure 2: Calculation of available area for the expansion of productive activities. -------------------------41 12 Figure 3: Land use by class, excluding the Pampa, Caatinga and Pantanal biomes. ------------------------42 Figure 4: Methodological land-use diagram ----------------------------------------------------------------------53 Figure 5: evolution of the demand for land in Brazil by crop in the reference scenario - 2006-30 (million ha). ------------------------------------------------------------------------------------------------------------59 Figure 6: architecture of the LULUCF study, with an emphasis on the components that include the defo- restation factor --------------------------------------------------------------------------------------------------------64 Figure 7: example of the database prepared for simulations of land-use change and cover. -------------65 Figure 8: First part of the spatially explicit model for land-use change and soil cover - land allocation. 67 Figure 9: spatially explicit land-use change and soil cover model - simulation of land-use change. ----68 Figure 10: Information flow in the analytical model ------------------------------------------------------------71 Figure 11: Variation i the pasture area occupied by type of productive system Technical Synthesis Report | Land Use, Land-Use Change, and Forestry in the reference scenario ---------------------------------------------------------------------------------------------80 Figure 12: area of the country occupied by agriculture, pasture, planted forests and complementary area in the form of native vegetation and other uses, from 1990-2030. --------------------------------------83 Figure 13: Fictitious land-use change scheme for three crops (a, B, and C). ---------------------------------85 Figure 14: Co2, n2o and Ch4 emissions from agriculture during the 2010-2030 period, expressed in Co2 equivalents in the reference scenario.-----------------------------------------------------------------------92 Figure 15: deforestation dynamic in the three main biomes in Brazil in the reference scenario (km2/ year) ---------------------------------------------------------------------------------------------------------------------96 Figure 16: emissions from land use in the reference scenario. -----------------------------------------------98 Figure 17: diagram of potential carbon removals by reforestation for the Cerrado and atlantic Forest biomes. ---------------------------------------------------------------------------------------------------------------- 100 Figure 18: Points attributed to the WCMI values in the model. Modified by Iverson et al. (1994).----- 102 Figure 19: Points attributed to the amount of rainfall in the model, according to Iverson et al. (1994)--------------------------------------------------------------------------------- 102 Figure 20: Points attributed to altitude classes in the model, modified by Iverson et al. (1994). ------ 103 Figure 21: Points attributed to the degree of incline of the land, modified by Iverson et al. (1994). -- 104 Figure 22: graph of the box-plot where the distribution of the amounts for altitude, rainfall and months of hydrous deficit may be observed for the embrapa and WorldClim databases. ------------------------ 106 Figure 23: Logistical function of biomass uptake using local biomass potential and age of vegetation as parameters. ---------------------------------------------------------------------------------------------------------- 115 Figure 24: reference scenario for charcoal with a low level of legal restrictions; participation of ther- mo-reduction agents in the Brazilian iron and steel market. ------------------------------------------------ 118 Figure 25: reference scenario for charcoal with a high level of legal restrictions: participation of ther- mo-reduction agents in the Brazilian iron and steel producing market ----------------------------------- 119 Figure 26: Projection of Co2 emissions for the reference scenario (charcoal) -------------------------- 120 Figure 27: reference scenario results, emissions from land use and land-use change, 2009-30. ----- 121 Figure 28: Percentage of reduction of soil and water losses from zero tillage (Zt) compared to conven- tional planting (CP).------------------------------------------------------------------------------------------------- 124 Figure 29: Co2e stock from forest plantations for renewable charcoal in scenario 1 ------------------- 132 Figure 30: Co2e stock in forest plantations for renewable charcoal in scenario 2. ---------------------- 133 Figure 31: Comparison of Co2e stock in scenarios 1 and 2 and the reference scenario.---------------- 134 Figure 32: Carbon uptake potential of forest recovery activities and production forests. -------------- 143 Figure 33: Mitigating measures for the construction of the Low-carbon scenario----------------------- 161 13 Figure 34: Change in pasture area occupied according to the type of productive system (million hectares) --------------------------------------------------------------------------------------------------- 162 Figure 35: Variation in number of head of cattle in productive systems, 2009-30 ----------------------- 163 Figure 36: Projection of Brazilian herd productivity between 2009 and 2030 for the reference and Low- -carbon Scenarios. -------------------------------------------------------------------------------------------------- 164 Figure 37: projection of pasture area in Brazil from 2009 to 2030 (Low Carbon scenario) ------------ 165 Figure 38: Comparison of methane emissions from beef-cattle raising (MtCo2e per year), 2008-30 166 Figure 39: Comparison of methane emissions per unit of meat (kg Co2e per kg), 2008–30. ---------- 167 Figure 40: evolution of Brazil’s demand for land, by crop, 2006-30 (millions of ha). -------------------- 171 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Figure 41: evolution of deforestation in the Low-carbon scenario (curve) (km2 per year). ----------- 178 Figure 42: evolution of deforestation in the Low-carbon scenario (LCs) and reference scenarios (rs) (thousands of ha per year). ---------------------------------------------------------------------------------------- 178 Figure 43: Identification of forest degradation patterns in the amazon within the framework of the de- grad program. source: InPe, 2009. ----------------------------------------------------------------------------- 182 Figure 44: reference scenario results: emissions from land use and land-use change, 2009–3. ------ 187 Figure 45: emissions from land use and land-use change under the new land-use dynamic in the Low- -carbon Scenario. ---------------------------------------------------------------------------------------------------- 188 Figure 46: Comparisons of gross emissions distribution among sectors in the reference and Low-car- bon scenarios, 2008–30. ------------------------------------------------------------------------------------------- 190 Figure 47: transport processes simulated by the CCatt-BraMs, including plume rise, deep and shallow convective transport by cumulus, diffusion in the PBL, dry and wet deposition. -------------- 195 Figure 48: estimate of total annual aerosol emissions in Brazil for the reference and Low-carbon scena- rios (table 43). ------------------------------------------------------------------------------------------------------- 199 Figure 49: average monthly precipitation in the 6 regions analyzed in the reference and Low-carbon scenarios from 2007 to 2008 compared to data obtained from the agência nacional de �guas (natio- nal Water agency - ana), which corresponds to monthly precipitation during the period from 1982 to 2005. the margins of error represent the standard deviation for each month. -------------------------- 203 Figure 50: average monthly precipitation in the 6 regions analyzed in the reference and Low-carbon scenarios from 2007 to 2030 (bar graph left axis). also shown is the difference between the reference and Low-carbon scenarios (line graph right axis). ------------------------------------------------------------ 204 Figure 51: difference in precipitation (mm) between the reference and Low-carbon scenarios conside- ring the years 2007 to 2030 during the February, March and april (a), May, april and June (B), august, september and october (C), and november, december and January (d) trimesters. ------------------- 205 Figure 52: average monthly temperature in the 6 regions analyzed in the reference and Low-carbon scenarios in 2007 and 2008 compared to data obtained from the national Meteorology Institute (Insti- tuto nacional de Meteorologia - InMet), which corresponds to the monthly climatology of temperatu- re during the period from 1977 to 2000. the margins of error represent the standard deviation for each month. ----------------------------------------------------------------------------------------------------------------- 206 Figure 53: difference in temperature (Celsius) between the reference and Low-carbon scenarios for the years 2007-2030 for the February, March and april (a), april, May and June (B), august, september and october (C), and november, december and January (d) trimesters. ---------------------------------- 208 Figure 54: difference in accumulated rainfall between the reference and Low-carbon scenarios, using the 2007-2030 average. the color scale refers to amounts in millimeters of rainfall per year. -------- 209 Figure 55: difference between the reference and low-carbon scenarios in average air temperature, taken between 2007 and 203. the color scale reflects the amount in celsius ----------------------------- 210 Figure 56: Marginal abatement Cost (8-percent social discount rate) and break-even carbon price (considering an Irr of 12%) for deforestation avoidance measures. -------------------------------------- 225 Figure 57: Variations in forest restoration costs by intervention scenario -------------------------------- 228 14 Figure 58: MaC and equilibrium price of carbon for Co2 uptake through legal forest restoration. --- 229 Figure 59: Percentages of distribution of investments by group of measures. --------------------------- 232 Figure 60: Cost of land in the state of são Paulo between 1995 and 2008 ---------------------------------- 234 Figure 61: Variation in crops prices used in the present study ---------------------------------------------- 236 Figure 62: results of the reference scenario (Up) and aggregate Low-carbon scenario (down) ----- 263 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Maps Map 1: Map of the main regions of the land-use model. ---------------------------------------------------------44 Map 2: dynamics of areas where sugar cane and cotton are grown in the reference scenario (2010- 2030) ------------------------------------------------------------------------------------------------------------------ 60 Map 3: dynamics of areas where rice and beans are grown for the reference scenario (2010-2030) 61 15 Map 4: dynamics of areas where corn and soybean are cultivated for the reference scenario (2010- 2030). ----------------------------------------------------------------------------------------------------------------- 62 Map 5: dynamic of planted forest and pasture areas for the reference scenario (2010-2030)-------- 63 Map 6: simplification of the soils map for Brazil with six soil categories ---------------------------------- 86 Map 7: total ghg emissions in Co2 equivalent (millions of tons) by state resulting from agricultural land use --------------------------------------------------------------------------------------------------------------- 93 Map 8: deforestation in the reference scenario (2010-2030) ---------------------------------------------- 96 Map 9: Carbon stock mosaic--------------------------------------------------------------------------------------- 97 Map 10: Boundaries of the Cerrado and the atlantic Forest, extracted from the Map of Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Brazilian Biomes ---------------------------------------------------------------------------------------------------- 108 Map 11: altimetry based on the digital srtM Land Model. original data offer an average altitude with 3� resolution. the model presented was re-modeled for 30� ----------------------------------------- 108 Map 12: declivity in percentages based on the digital land model ------------------------------------------ 109 Map 13: average annual precipitation in millimetres --------------------------------------------------------- 109 Map 14: Length of the growing season indicated by the sum of months with higher precipitation than 50 mm ----------------------------------------------------------------------------------------------------------------- 110 Map 15: average temperature of the hottest month of the year --------------------------------------------- 111 Map 16: soil fertility map for Brazil ------------------------------------------------------------------------------ 111 Map 17: Map fo plant cover in Brazil for 2000 ------------------------------------------------------------------ 112 Map18: Map of Brazilian ecosystems, IBge, representing an estimate of the dsitribution of “original� plant formations with a simplified legend indicating areas of transition ---------------------------------- 113 Map 19: Map of carbon uptake potential through the forest restoration of the Legal reserve in the Cer- rado and atlantic Forest in tCo2/ha ------------------------------------------------------------------------------ 114 Map 20: Map of carbon uptake potential through the restoration of riverside forests in the Cerrado and atlantic Forest biomes in tCo2/ha -------------------------------------------------------------------------------- 114 Map 21: Forest productivity (tCo2/ha/year) for the Cerrado and atlantic Forest biomes ------------- 116 Map 22: Mitigation by crop, 2010 to 2030----------------------------------------------------------------------- 126 Map 23: total emissions from agriculture, 2010 to 2030 ----------------------------------------------------- 127 Map 24: Co2 uptake potential through forest restoration by 2030 and total Co2 uptake potential --- 144 Map 25: number of heads of cattle ------------------------------------------------------------------------------- 168 Map 26: total cumulative emissions from livestock, 2010-2030------------------------------------------- 169 Map 27: dynamics of sugar cane cultivation and cotton in the Low-carbon scenario ------------------- 172 Map 28: dynamics of the rice and bean crops in the Low-carbon scenario -------------------------------- 173 Map 29: dynamics of corn crop and soybean in the Low-carbon scenario -------------------------------- 173 Map 30: dynamics of planted forests and pastures in the Low-carbon scenario (2010 – 2030). yellow = reimaned constant; blue = crop decrement; red = crop increment ---------------------------------------- 174 Map 31: Forest regrowth in the low-carbon scenario --------------------------------------------------------- 175 Map 32: area used for agriculture, pasture, and reforestationby region ---------------------------------- 175 Map 33: Comparison of cumulative deforestation, 2007-30------------------------------------------------- 176 Map 34: total area deforested, 2010-2030 --------------------------------------------------------------------- 177 Map 35: total cumulative emissions from deforestation, 2010-2030 ------------------------------------- 179 Map 36: total cumulative emissions from land use (agriculture, Livestock, deforestation, and refores- 16 tation) 2010-30------------------------------------------------------------------------------------------------------ 188 Map 37: Land-use map for the year 2007 in refernce scenario (1 x1 Km resolution) ------------------- 197 Map 38: schematic map of Brazil showing the different regions in the country and their boundaries for the analysis of results (above). Below, normal performance of the number of emissions sources in the different regions obtained with data from the aVhrr sensor (advanced Very high resolution radio- meter) from 1998 to 2008, present in the satellites of the noaa series (national oceanic and atmos- pheric administration)--------------------------------------------------------------------------------------------- 198 Map 39: Figures (a), (B), (C) and (d) show the locations of deforestation from 2007 to 2030 in the refe- rence (reF) and low carbon (BC) scenarios. regions with forest remnants are also shown during that period (in green). Figures (e) and (F) show the optic depth of average aerosol from 2007 to 2030 in the reference (e) and low carbon (F) scenarios, where the current lines represent the average wind field over Brazil ------------------------------------------------------------------------------------------------------------ 201 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Boxes Box 1: ego dynaMIC (environment for geoprocessing objects) --------------------------------------------66 Box 2: Moving towards a “Legal scenario�: Main areas to Protect ------------------------------------------ 142 Box 3: Uncertainties for economic land-use scenarios ------------------------------------------------------- 192 Box 4: Calculating Marginal abatement Costs ------------------------------------------------------------------ 213 Acronyms and Abbreviations aBraF Brazilian association of Plantation Forest Producers (associação Brasileira de Produtores de Florestas Plantadas) anFaVea national association of Motor Vehicle Manufacturers (associação 17 nacional dos Fabricantes de Veículos automotores) ANEEL national agency for electric energy (agencia nacional de energia elétrica) ANP national agency of Petroleum, natural gas, and Biofuels (agência nacional do Petróleo, gás natural, e Biocombustíveis) ARPA amazon region Protected areas Program BdMg Minas gerais development Bank BEN National Energy Balance BLUM Brazil Land Use Model Technical Synthesis Report | Land Use, Land-Use Change, and Forestry BNDES national Bank of economic and social development (Banco na- cional de desenvolvimento econômico e social) Can national Confederation of agriculture and Livestock CBers China-Brazil earth resources satellites CCC Fuel Consumption account (Conta de Consumo de Combustíveis) CCs socio-environmental Commitment register Cde energy development account (Conta de desenvolvimento ener- gético) CdM Clean development Mechanism CeaF Center for alternative energy strengthening (Centro de energias alternativas de Fortaleza) CeIF Clean energy Investment Framework CePeL research Center for electrical energy (Centro de Pesquisas de energia elétrica) Cer Certified emissions reduction CetesB são Paulo state Waste Management agency (Companhia de tec- nologia de saneamento ambiental) CFL Compact Fluorescent Lamp Cgee Center for strategic Management and studies Ch4 Methane CIde Contribution on Intervention in the economic domain (Contri- buição de Intervenção no domínio econômico) CMn national Monetary Council (Conselho Monetário nacional) Cng Compressed natural gas ConaB national Crop supply agency ConPet national Program for the rationalization of the Use of oil and natural gas derivatives (Programa nacional de racionalização do Uso dos derivados de Petróleo e gás natural) Co2 Carbon dioxide 18 CPteC Center for Weather Forecasts and Climate studies Csr remote sensing Center Ctenerg sector energy Fund of the Ministry of science and technology (Fundo sectorial de Ciência e tecnologia para energia) Ct-Petro oil and natural gas sector Fund of the Ministry of science and technology (Fundo sectorial de Ciência e tecnologia para Petróleo e gás) CU Conservation Unit degrad Mapping of Forest degradation in the Brazilian amazon Technical Synthesis Report | Land Use, Land-Use Change, and Forestry DETER Detection System for Deforestation in Real Time ego environment for geoprocessing objects EIA Energy Information Administration eMBraPa Brazilian agricultural research Corporation (empresa Brasileira de Pesquisa agrícola) EPE energy Planning Company (empresa de Planejamento energéti- co) esCo energy efficiency service Company FaPrI Food and agricultural Policy research Institute FdI Foreign direct Investment Fgee guarantee Fund for electric energy Projects Fgts social security (Fundo de garantia do tempo de serviço) FInaMe agency for the acquisition of Machines and equipment (agência Agricola de Financiamentos para aquisição de Máquinas e equipamentos) FIneM equipment and Machinery Financing (Financiadora de equipa- mentos e Máquinas) FIneP agency for the Funding of Projects and studies (Financiadora de estudos e Projetos) FnP FIneP Consulting & trade (FIneP Consultoria & Comércio) FUnaI national Foundation for Indigenous People gdP gross domestic Product geF global environment Facility ghg greenhouse gas gnP gross national Product gtL gas-to-Liquid hFC hydrofluorocarbon IBaMa Brazilian Institute of environment and renewable natural re- sources (Instituto Brasileiro do Meio ambiente e dos recursos naturais renováveis) 19 IBge Brazilian Institute of geography and statistics (Instituto Brasilei- ro de geografia e estatística) IBP Potential Biomass Index ICMBio Chico Mendes Institute of Biodiversity Conservation (Instituto Chico Mendes de Conservação da Biodiversidade) ICone Institute for International Trade Negotiations IEA International Energy Agency IgP-dI general Price Index-domestic availability (�ndice geral de Preços- -disponibilidade Interna) Technical Synthesis Report | Land Use, Land-Use Change, and Forestry INPE national Institute for space research (Instituto nacional de Pes- quisas espaciais) INT national technological Institute (Instituto nacional de tecnolo- gia) I-o Input-output IPaM amazon Institute for environmental research (Instituto de Pes- quisa ambiental da amazonia) IPCC Intergovernmental Panel on Climate Change IPI Industrial Products tax IRR Internal Rate of Return KfW german development Bank Lng Liquefied natural gas LULUCF Land Use, Land-Use Change, and Forestry MaC Marginal abatement Cost MaCC Marginal abatement Cost Curve MCt Ministry of science and technology (Ministério de Ciência e tec- nologia) MeLP Long-term expansion Model MePs Minimum energy Performance standard MIPe Integrated energy Planning Model MMa Ministry of the environment (Ministério do Meio ambiente) MMe Ministry of Mines and energy (Ministério de Minas e energia) M-ref refining study Model Msr residential energy demand Projection Model Mt Ministry of transport (Ministério dos transportes) N Nitrogen naPCC national action Plan on Climate Change NIPE Interdisciplinary Center for strategic Planning 20 nPV net Present Value (Valor Presente Líquido) nrC national research Council N2o nitrous oxide oeCd organisation for economic Co-operation and development PaC government accelerated growth Plan PAS sustainable amazon Program PFC Perfluorocarbon PLANSAB national sanitation Plan (Plano nacional de saneamento Básico) Technical Synthesis Report | Land Use, Land-Use Change, and Forestry PMe Monthly employment survey PNE national energy Plan (Plano nacional de energia) PNLT national Logistics and transport Plan (Plano nacional de Logisti- ca e transporte) PnMC national Plan on Climate Change (Plano nacional sobre Mudança do Clima) PPA Permanent Preservation Area PPCdaM Plan of action for the Prevention and Control of deforestation in the Legal amazon (Plano de ação para Prevenção e Controle do desmatamento na amazônia Legal) PPP Public-Private Partnership ProaLCooL National Alcohol Program ProBIo Project for the Conservation and sustainable Use of Brazilian Bio- logical Diversity ProCeL national electrical energy Conservation Program (Programa de Combate ao desperdício de energia elétrica) Prodes amazon deforestation Monitoring Program (Programa de Cálculo do desflorestamento da amazônia) ProdUsa Programa de estímulo a Produção agropecuária sustentável) ProesCo support Program for energy efficiency Projects (Programa de apoio a Projetos de eficiência energética) ProInFa Incentive Program for alternative electric energy sources (Pro- grama de Incentivo às Fontes alternativas) ProLaPeC agriculture-Livestock Integration Program (Programa de Integra- ção Lavoura-Pecuária) PronaF national Program for the strengthening of Family agriculture (Programa nacional de Fortalecimento da agricultura Familiar) ProPasto National Program for Recuperation of Degraded Pastures ProPFLora Program for Commercial Planting and recovery of Forests (Pro- 21 grama de Plantio Comercial e recuperação de Florestas) r&d Research and Development REDD Reducing Emissions from Deforestation and Degradation rgr global reversion reserve (reserva global de reversão) RSU Urban solid residues (residuos sólidos Urbanos) SAE secretariat of strategic affairs (secretaria de assuntos estratégicos) sFB Brazilian Forest service sF6 sulfur hexafluoride Technical Synthesis Report | Land Use, Land-Use Change, and Forestry UFMg Federal University of Minas gerais (Universidade Federal de Mi- nas gerais) UFrJ Federal University of rio de Janeiro UnFCCC United nations Framework Convention on Climate Change UnICaMP state University of Campinas USP University of são Paulo WTI West text Intermediate Units of Measure Ce Carbon equivalent Co2e Carbon dioxide equivalent ETE Sewage Treatment Plant gCo2e grams of Carbon dioxide equivalent 22 gt Billions of Tons gt Co2e Billion tons of Carbon dioxide equivalent gW gigawatt gWh gigawatt hour ha hectare kg Kilogram km Kilometer km square Kilometer Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2 kW Kilowatt m Meter m3 Cubic Meters Tg Teragram tg Co2e teragram Carbon dioxide equivalent MW Megawatt MWh Megawatt hour ppm Particles per Million tCo2e tons of Carbon dioxide equivalent TWh terawatt hour Currency Exchange 1 Us dollar (Us$) = 2.20 Brazilian reais (r$) Acknowlegments this report has been prepared by a core team led by Christophe de gouvello (the World Bank), Britaldo s. soares Filho (Federal University of Minas gerias, UFMg), an- dré nassar (Institute for International trade negotiations, ICone), Luis g. Barioni and Bruno J. r. alves (Brazilian agricultural research Corporation, eMBraPa), osvaldo 23 Martins and Magno Castelo Branco (Iniciativa Verde), Manoel regis Lima Verde Leal (Centro de energias alternativas e Meio ambiente, Cenea), Fábio Marques (PLan- tar), Júlio hato and sérgio Pacca (University of são Paulo, UsP), and saulo ribeiro Frei- tas (national Institute for space research, InPe). the team was assisted by Letícia hissa (UFMg), Leila harfuch, Marcelo Melo ra- malho Moreira, Luciane Chiodi Bachion and Laura Barcellos antoniazzi (ICone), geraldo Martha Junior, roberto d. sainz, and Magda a. de Lima (eMBraPa), renato to- ledo (Iniciativa Verde), rodrigo Ferreira, Luiz goulart, and thiago Mendes (PLantar), Karla Maria Longo and ricardo almeida de siqueira (InPe), and Mark Lundell, adriana Moreira, Barbara Farinelli, Jennifer Meihuy Chang, govinda timilsina, garo Batmanian, Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Fowzia hassan, Benoit Bosquet, alexandre Kossoy, Flávio Chaves, Mauro Lopes de aze- vedo, Fernanda Pacheco, Megan hansen, augusto Jucá, and rogerio Pinto (the World Bank). the team benefited greatly from a wide range of consultations with the Ministries of Foreign affairs, environment, and science and technology. several seminars were or- ganized, enabling consultation with representatives of the Ministries of Finance, Plan- ning, agriculture, transport, Mines and energy, Industry and Commerce. The team acknowledges the generous support provided by the Sustainable Devel- opment Network for activities related to climate change and regional support through the energy sector Management assistance Program (esMaP). the team is especially gratefull to Mark Lundell and garo Batmanian for their valu- able contribution throughtout the development of the study. special thanks to adriana Moreira for coordinating the initial compilation of the re- port and to Barbara Farinelli for revising the translation, editing and coordinating the printing process. the team would also like to acknowledge Judy Wolf and helena Jansen for their sup- port in translating and editing the report. Executive Summary This report presents the partial results related to land use, land-use change and the forestry sector from a larger multisectoral low-carbon study for Brazil1. Brazil’s commitment to combat climate change had already begun when the country hosted the United nations Conference on environment and development, also known 24 as the rio earth summit, in June 1992. the resulting United nations Framework Con- vention on Climate Change (UnFCCC) led to the creation of the Kyoto Protocol. today, Brazil remains strongly committed to voluntarily reducing its carbon emissions. on december 1, 2008, President Luiz Inácio Lula da silva launched the national Plan on Climate Change (PnMC), based on work of the Interministerial Committee on Climate Change, in collaboration with the Brazilian Forum on Climate Change and civil society organizations. the PnMC calls for a 70-percent reduction in deforestation by 2017, a particularly noteworthy goal given that Brazil has the world’s second largest block of remaining native forest. on december 29, 2009 the Brazilian government adopted Law 12.187, which institutes the national Climate Change Policy of Brazil and set a volun- tary national greenhouse gas reduction target of 36.1 to 38.9 percent of projected emis- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry sions by 2020. as the world’s largest tropical country, Brazil is unique in its greenhouse gas (ghg) emissions profile. In prior decades, the availability of large volumes of land suitable for crop cultivation and pasture helped to transform agriculture and livestock into key sectors for sustaining the country’s economic growth. In the past decade alone, these two sectors accounted for an average of 25 percent of the national gdP. the steady expansion of crop land and pasture has also required the conversion of more native land, making land-use change the country’s main source of ghg emissions today. at the same time, Brazil has used the abundant natural resources of its vast territory to explore and develop low-carbon renewable energy. yet Brazil used to be one of the largest ghg emitters from deforestation and would probably continue to be so if not for the government’s recent adoption of a series of measures to protect the forest. Although drastically reduced in recent years, deforesta- tion could continue to be a potentially large emission source in the future. at the same time, the country is likely to suffer significantly from the adverse effects of climate change. Some advanced models suggest that much of the eastern part of the Brazilian amazon region could be converted into a savannah-like ecosystem before the end of this century. a phenomenon known as amazon dieback, combined with the shorter-term effects of deforestation by fires, could reduce rainfall in the central-west and northeast regions, resulting in smaller crop yields and less available water for hydropower-based electricity2. Urgent solutions are thus needed to reduce Brazil’s vulnerability to climate change and to enable the implementation of adaptation actions in the country. Like many other developing countries, Brazil faces the dual challenge of encourag- ing development and reducing ghg emissions. President Lula echoed this concern in his introduction to the national Plan, stating that actions to avoid future ghg emissions 1 Brazil Low-carbon Case study, World Bank, June 2010. 2 “assessment of the risk of amazon dieback,� World Bank, 2010. should not adversely affect the development rights of the poor, who have done nothing to generate the problem. efforts to mitigate ghg emissions should not add to the cost of development, but there are strong reasons to shift toward a low-carbon economy. Low- carbon alternatives would offer important development co-benefits, ranging from reduced congestion and air pollution in urban transport areas to better waste manage- ment, job creation and costs savings for industry, and biodiversity conservation. Coun- tries that pursue low-carbon development are more likely to benefit from strategic and 25 competitive advantages, such as the transfer of financial resources through the carbon market, new international financing instruments, and access to emerging global mar- kets for low-carbon products. In the future this may create a competitive advantage for the production of goods and services, due to the lower emission indexes associated with the life cycle of products. Study Overview the overall aim of this study was to support Brazil’s efforts to identify opportunities to reduce its emissions in ways that foster economic development. The primary objec- tive was to provide the Brazilian government with the technical inputs needed to assess Technical Synthesis Report | Land Use, Land-Use Change, and Forestry the potential and conditions for low-carbon development in key emitting sectors. To this end, the World Bank study adopted a programmatic approach in line with the Brazilian government’s long-term development objectives, as follows: (i) anticipate the future evolution of Brazil’s ghg emissions to establish a reference scenario; (ii) identify and quantify lower carbon-intensive options to mitigate emissions, as well as potential options for carbon uptake; (iii) assess the costs of these low-carbon options, identify barriers to their adoption, and explore measures to overcome them; and (iv) build a low-carbon emissions scenario that meets the same development expectations. the team also analyzed the macroeconomic effects of shifting from the reference sce- nario to the low-carbon one and the financing required. To build on the best available knowledge and avoid duplicating efforts, the study team undertook a broad consultative process, meeting with more than 70 recognized Brazilian experts, technicians, and government representatives covering most emit- ting sectors and surveying the copious literature available. This preparatory work informed the selection of four key areas with great potential for low-carbon options: (i) land use, land-use change, and forestry (LULUCF), including deforestation; (ii) trans- port systems; (iii) energy production and use, particularly electricity and oil and gas; and (iv) solid and liquid urban waste.3 In order to estimate the emissions Brazil would generate in these four key areas over the study period, the study team defined a “reference scenario� that is later compared with the projected “Low-carbon scenario�. It is worth noting that the reference scenar- io is based on a different methodology than the one used by the Brazilian government 3 Certain industrial sources of nitrous oxide (n2o), hydrofluorocarbons (hFCs), perfluorocarbons (PFCs), sulfur hexafluoride (sF6), and other non-Kyoto ghg gases are not covered by this study. Without a recent complete inventory, it is not possible to determine precisely the share of other sources in the national ghg balance. however, based on the first Brazil national Communication (1994), it is expected that they would not exceed 5 percent of total Kyoto ghg emissions. not all agricultural activities were taken into account when estimating emissions from that sector; and crops taken into account in the calculation of emissions from agriculture represent around 80% of the total crop area. in its national ghg inventory. In particular, having focused on these four areas, the ref- erence Scenario built by this study does not cover 100 percent of all emission sources of the country and therefore should not be considered a simulation of future national emissions inventories. Reference-scenario results for these main areas show that deforestation remains the key driver of Brazil’s future ghg emissions through 2030. Modeling results indi- cate that, after a slight decrease in 2009–11, deforestation emissions are expected to 26 stabilize at an annual rate of about 400–500 Mt Co2. despite its significant decline over the past four years, deforestation remains Bra- zil’s largest source of carbon emissions, representing about two-fifths of national gross emissions (2008). over the past 15 years, deforestation has contributed to reducing Brazil’s carbon stock by about 6 billion metric tons, the equivalent of two-thirds of an- nual global emissions.4 Without the Brazilian government’s recent forest protection efforts, the current emissions pattern from deforestation would be significantly high- er.5 the drivers of deforestation occur at multiple levels. In the amazon and Cerrado regions, for example, the spatial dynamics of agricultural and livestock expansion, new roads, and immigration determine the pattern of deforestation. on a national or inter- national scale, broader market forces affecting the meat and crop sectors drive defores- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry tation. Agricultural production and livestock activities also produce direct emissions, to- gether accounting for one-fourth of national gross emissions. Agricultural emissions mainly result from the use of fertilizer and mineralization of nitrogen (n) in the soil, cultivation of wetland irrigated rice, the burning of sugar cane, and use of fossil fuel– powered agricultural equipment. Livestock emissions result mainly from the digestive process of beef cattle, which releases methane (Ch4) into the atmosphere. Models and reference-scenario results to estimate future demand for land and LULUCF emissions, the study developed two complementary models: i) Brazilian Land Use Model (BLUM) and (ii) simulate Brazil (sIM Brazil). BLUM is an econometric model that estimates the allocation of land area and measures changes in land use resulting from supply-and-demand dynam- ics for major competing activities.6 sIM Brazil, a geo-referenced spatialization model, estimates future land use over time under various scenarios. sIM Brazil does not alter BLUM data; it finds a place for land-use activities, taking into account such criteria as agricultural aptitude, distance to roads, urban attraction, cost of transport to ports, de- clivity, and distance to converted areas. sIM Brazil works at a definition level of 1 km2, making it possible to generate detailed maps and tables. Under the reference scenario, about 17 million ha of additional land are required to accommodate the expansion of all activities over the 2006–30 period. In Brazil as a 4 From 1970 to 2007, the amazon lost about 18 percent of its original forest cover; over the past 15 years, the Cerrado lost 20 percent of its original area, while the atlantic Forest, which had been largely deforested earlier, lost 8 percent. 5 after peaking at 27,000 km² in 2004, deforestation rates have declined significantly, dropping to 11,200 km² in 2007, the second lowest historical rate recorded by the Prodes deforestation observation program (InPe 2008). 6 these include six key crops (soybean, corn, cotton, rice, bean, and sugar cane), pasture, and production forests; the model also projects the demand for various kinds of meat and corresponding needs for hay and corn. whole, the total area allocated for productive uses, estimated at 257 million ha in 2008, is expected to grow 7 percent—to about 276 million ha—in 2030; 24 percent of that growth is expected to occur in the amazon region. In 2030, as in 2008, pastures are expected to occupy most of this area (205 million ha in 2008 and 207 million in 2030). growth of this total amount over time makes it necessary to convert native vegetation for productive use, which mainly occurs in frontier regions, the amazon region, and in Maranhão, Piauí, tocantins, and Bahia on a smaller scale. 27 To estimate the corresponding balance of annual emissions and carbon uptake over the next 20-year period, these and related models calculated land use and land-use change for each 1-km2 plot at several levels.7 Results showed that land-use change via deforestation accounts for the largest share of annual LULUCF emissions—up to 533 Mt Co2e by 2030. direct annual emissions from land use only (agriculture and livestock) increase over the period at an average annual rate of 346 Mt Co2e. Carbon uptake offsets less than 1 percent of gross LULUCF emissions, sequestering 29 Mt Co2e in 2010, down to 20 Mt Co2e in 2030. over the 20-year period, LULUCF gross emissions increase one-fourth, reaching 916 Mt Co2e by 2030. the net balance between land use, land-use change, and carbon uptake results in increased emissions, which reachs about 895 Mt Co2e annually by 20308. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Low-carbon Options for Emissions Mitigation and Carbon Uptake avoiding deforestation offers by far the greatest opportunity for ghg mitigation in Brazil. Under the resulting Low-carbon scenario, avoided emissions from deforesta- tion would amount to about 6.2 gt Co2e over the 2010–30 period, or more than 295 Mt Co2e per year. Brazil has developed forest-protection policies and projects to counter the progres- sion of pressure at the frontier and has experience in economic activities compatible with forest sustainability. Shifting to a Low-carbon Scenario that ensures the growth of agriculture and the meat industry—both important to the Brazilian economy—would also require acting on the primary cause of deforestation: the demand for more land for agriculture and livestock. this study proposed a dual strategy to drastically reduce deforestation: (i) eliminate the structural causes of deforestation and (ii) protect the forest from illegal attempts to cut it. Eliminating the structural causes of deforestation would require a dramatic increase in productivity per hectare. Increasing livestock productivity could free up large quantities of pasture. this option is technically feasible since Brazil’s livestock productivity is generally low, and existing feedlots and crop-livestock systems could be scaled up. Moreover, the use of more intensive production systems could trigger higher economic returns and a net gain for the sector economy (Chapter 7). the potential to release and recover degraded pasture is enough to accommodate the most ambitious growth scenario. 7 Micro-region, state, and country. 8 When calculating national carbon inventories, some countries consider the contribution of natural regrowth towards carbon uptake; therefore, although this study does not compute this contribution in the carbon balance of LULUCF activities, it would be fair to add that information for comparison purposes. If the carbon uptake from the natural regrowth of degraded forests were to be included, then the potential uptake would increase by 109Mt Co2 per year, thus reducing net emissions. The combination of reducing pasture area and protecting forests can lead to a sharp decline in deforestation emissions. this was demonstrated in 2004–07, when new forest-protection efforts, combined with a slight contraction in the livestock sector and resultant pasture area,9 led to a 60-percent reduction in deforestation (from 27,000 km² to 11,200 km²). such a rapid reduction resulted from deforestation and its asso- ciated emissions being related to the marginal expansion of land for agriculture and livestock activities,10 without which there would be no need to convert additional na- 28 tive vegetation and incidentally generate ghg emissions. If the effort to reduce pasture area and protect forests were neglected, emissions from deforestation would resume immediately. To protect against illegal cutting, the forest should be further protected against fraudulent interests. the Brazilian government has made considerable efforts in this area, particularly under the 2004 Plan of action for the Prevention and Control of deforestation in the Legal amazon (PPCdaM). Model-based projections indicate that, under the new land-use dynamic, defores- tation would be reduced by more than two-thirds (68 percent) in 2030, compared to projected levels in the reference scenario. In the atlantic Forest, the reduction would be about 90 percent, while the amazon region and Cerrado would see reductions of 68 percent and 64 percent, respectively. accordingly, in 2030, annual emissions from de- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry forestation would be reduced nearly 63 percent (from about 530 Mt Co2 to 190 Mt Co2) compared to the projected reference scenario. In the amazon, the level of deforesta- tion would fall quickly to about 17 percent of the historic annual average of 19,500 km2 observed in the recent past, thus complying with the nPCC goal of reducing deforesta- tion in the amazon region by 72 percent by the year 201711. The study also proposed ways to reduce direct emissions from agricultural pro- duction and livestock activities. For agriculture, the study proposed an accelerated dissemination of zero-tillage cultivation. Compared to conventional farming systems, zero-tillage involves far fewer operations and can thus reduce emissions caused by altering soil carbon stock and using equipment powered by fossil fuels. Done effec- tively, zero-tillage cultivation can help control soil temperature, improve soil structure, increase soil water-storage capacity, reduce soil loss, and enhance the nutrient reten- tion of plants. For these reasons, expansion of zero-tillage cultivation is accelerated in the Low-carbon scenario, reaching 100 percent by 2015 and delivering 356 Mt Co2e of avoided emissions over the 2010–30 period. To lower direct emissions from beef-cattle farming, the study proposed shifting to more intensive meat-production systems, as mentioned above. It also proposed genet- ic improvement options to reduce Ch4, including improved forage for herbivores and genetically superior bulls, which have a shorter life cycle. The study projects that the combination of improved forage and bulls, along with increased productivity, would re- duce direct livestock emissions from 272 to 240 Mt Co2 per year by 2030; that is, main- taining them close to 2008 levels. 9 the 2005–07 period witnessed the first decline in herd size (from 207 million to 201 million head), following a decade-long increase, together with a slight contraction in pasture area (from 210 million to 207 million ha). 10 Unlike other sectors, whose energy-based emissions are usually proportional to the full size of the sector activity, emissions from deforestation are related only to the marginal expansion of agriculture and livestock activities. 11 over the 1996–2005 period, the historical rate of deforestation in the amazon region was 1.95 million ha per year, according to the PnMC. the study also explored two major carbon uptake options: (i) recovery of native forests and (ii) production forests for the iron and steel industry. For forest recovery, the Low-carbon Scenario considered compliance with legal actions for mandatory re- constitution, in accordance with the laws of riparian forests and legal reserves.12 In this sense, the Low-carbon scenario engendered a “Legal scenario�. Using these defined areas for reforestation, the study modeled their potential for Co2 removal.13 Results showed that the legal scenario has high carbon-uptake potential: a cumulative total 29 of 2.9 gt Co2e over the 20-year period or about 140 Mt Co2e per year on average14. For production forests, the Reference Scenario assumed that the thermo-reduction pro- cess would be based on coke (66 percent), non-renewable charcoal (24 percent), and renewable charcoal (10 percent), based on estimates that reflect the current situation. two Low-carbon scenarios were developed. the first only reflects the maintenance of the current participation of charcoal in iron and steel production (approximately 34 percent), but with a completely renewable origin. the second – considered in the overall calculations of this report – was more daring, so that the hypothesis of competi- tion for the use of land for planted forests would be taken into consideration in a rather conservative way, assuming a total substitution of non-renewable charcoal by 2017 and the use of renewable charcoal for up to 46 percent of total iron and steel ballast Technical Synthesis Report | Land Use, Land-Use Change, and Forestry production by 2030. With this, the volume of greenhouse gas uptake or “sequestration� would be between 500 and 700 MtCo2 in 2030, or from 321 to 517 MtCo2 more than in the Reference Scenario. A New Land-use Dynamic Building a Low-carbon Scenario for land use involves more than adding emissions reductions associated with mitigation opportunities; it must also avoid the potential for carbon leakage. For example, increasing forest recovery leads to carbon uptake, but it also reduces the land area otherwise available for expanding agriculture and live- stock activities. this, in turn, could cause an excess in demand for land use, which could generate deforestation, inducing a lower net balance of carbon uptake. To avoid carbon leakage, ways must be found to reduce the global demand for land for other activities, while maintaining the same level of product supply found in the Reference Scenario. In the Low-carbon Scenario, the amount of additional land required for emissions mitigation and carbon uptake totals more than 53 million ha. of that amount, more than 44 million ha—over twice the land expansion projected under the reference sce- nario—is for forest recovery. together with the additional land required under the ref- erence scenario, the total volume of additional land required is more than 70 million ha, more than twice the total amount of land planted with soybean (21.3 million ha) and sugar cane (8.2 million ha) in 2008 or more than twice the area of soybean projected for 2030 in the reference scenario (30.6 million ha) (table 1). 12 In areas with optimal conditions, forest recovery can remove 100 tC per ha on average in the amazon region. (saatchi, 2007). In the reference scenario, its contribution is limited in terms of quantity. 13 the study model used meteorological and climatic variables (e.g., rainfall, dry season, and temperature) and edaphic (soil and topography) variables to estimate potential biomass. 14 If the carbon uptake from the natural regrowth of degraded forests were to be included, then the potential uptake would increase by 112Mt Co2 per year on average. Table 1: Summary of additional land needs in the reference and Low-carbon Scenarios Scenario additional land needed (2006–30) Reference Scenario: additional expansion of agriculture and livestock production to meet volume of land required for the the needs anticipated in 2030: expansion of agriculture and • 16.8 million ha livestock activities 30 elimination of non-renewable charcoal in 2017 and the participation of 46 percent of renewable charcoal for iron and steel production in 2030: • 2.7 million ha expansion of sugar cane to increase gasoline substitution Low-carbon Scenario: addition- with ethanol to 80 percent in the domestic market and al volume of land required for supply 10 percent of estimated global demand to achieve mitigation measures an average worldwide gasoline mixture of 20 percent etha- nol by 2030: • 6.4 million ha Technical Synthesis Report | Land Use, Land-Use Change, and Forestry restoration of the environmental liability of “legal re- serves� of forests, calculated at 44.3 million ha in 2030: • 44.3 million ha Total 70.4 million additional hectares To increase livestock productivity to the level needed to release the required volume of pasture, the Low-carbon scenario considered three options: (i) promote recovery of degraded pasture, (ii) stimulate the adoption of productive systems with feedlots for finishing, and (iii) encourage the adoption of crop-livestock systems. the increased carrying capacity that results from the recovery of degraded areas, combined with more intensive integrated crop-livestock systems and feedlots for finishing are reflect- ed in an accentuated reduction in the demand for land, projected at about 138 million ha in the Low-carbon scenario, versus 207 million ha in the reference scenario, for the year 2030. the difference would be enough to absorb the demand for additional land associated with both expanded agriculture and livestock activities in the reference scenario and expanded mitigation and carbon uptake in the Low-carbon scenario. A National Low-carbon Scenario the Low-carbon scenario constructed for Brazil in the global, multi-sectoral study is an aggregate of the Low-carbon Scenarios developed for each of the four sectors considered in the study. In each sector, the most significant opportunities to mitigate and sequester ghgs were analyzed, while less promising or fully exploited options in the Reference Scenario were not considered further. In short, this national Low-carbon Scenario is derived from a bottom-up, technology-driven simulation for single subsec- tors (e.g., zero tillage with straw or reduction of deforestation), based on in-depth tech- nical and economic assessments of feasible options in the Brazilian context, and sector- level optimization for two of the four main sectors (land use and transport). This national Low-carbon Scenario has been built in a coordinated way to ensure full consistency among the four main sectors considered. To ensure transparency, the methods and results were presented and discussed on various occasions with a range of government representatives.15 But this Low-carbon Scenario is not presumed to have explored all possible mitigation options or represent a preferred recommended mix. this scenario, which simulates the combined result of all the options retained under this study, should be considered modular—as a menu of options—and not pre- scriptive, especially since the political economy between sectors or regions may differ 31 significantly, making certain mitigation options that at first appear more expensive easier to select than others that initially appear more attractive economically. this Low-carbon scenario represents a 37-percent reduction in gross ghg emis- sions compared to the reference scenario over the 2010–30 period. the total cumula- tive emissions reduction over the period amounts to more than 11.1 gt Co2e, equal to approximately 37 percent of the cumulative emissions observed under the reference scenario. Projected gross emissions in 2030 are 40 percent lower under the Low-car- bon scenario (1,023 Mt Co2e per year) compared to the reference scenario (1,718Mt Co2e per year) and 20 percent lower than in 2008 (1,288 Mt Co2e per year) (table 2, Figure 1). In addition, forest plantations and recovery of legal reserves will sequester the equivalent of 16 percent of reference-scenario emissions in 2030 (213 Mt Co2e per Technical Synthesis Report | Land Use, Land-Use Change, and Forestry year) 16. Table 2: Comparison of emissions distribution among sectors in the reference and Low- carbon Scenarios - 2008-30 Reference 2008 Reference 2030 Low-carbon 2030 sector Energy 232 18 458 27 297 29 Mt CO2e % Mt CO2e % Mt CO2e % Transport 149 12 245 14 174 17 Waste 62 5 99 6 18 2 Deforestation 536 42 533 31 196 19 Livestock 237 18 272 16 249 24 Agriculture 72 6 111 6 89 9 total gross emissions 1,288 100 1,718 100 1,023 100 Carbon uptake -29 -2 -21 -1 -213 -21 Total net emissions 1,259 98 1,697 99 810 79 The two areas where the proposed Low-carbon Scenario succeeds most in reduc- ing net emissions are reducing deforestation and increasing carbon uptake. The main drivers are (i) reduction of total land area needed, via significant gains in livestock pro- ductivity, to accommodate expanded agriculture and meat production and (ii) restora- 15 three seminars were held over the past several years (september 14–16, 2007; april 30, 2008; and March 19, 2009) to present and discuss the study methodology, intermediate results, and near-final results with representatives of 10 ministries. sectoral teams also interacted on various occasions with technical-experts and public-agency representatives. 16 If the carbon uptake from the natural regrowth of degraded forests were to be included, then the potential uptake would increase by 112MtCo2 per year on average, thus reducing the net emissions. tion of legal forest reserves and production forests for producing charcoal for the steel industry. By 2017, the proposed Low-carbon scenario would reduce deforestation by more than 80 percent compared to the 1996–2005 average, thereby ensuring compli- ance with the Brazilian government’s december 2008 commitment. 32 Figure 1: GHG mitigation wedges in the Low-carbon Scenario, 2008-30 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Low-carbon Scenario, 2008-30 It is more difficult to reduce emissions in the energy and transport sectors, since they are already low by international standards, due mainly to hydroelectricity for power generation and bioethanol as a fuel substitute for gasoline in the current energy matrix. as a result, these sectors’ relative share of national emissions increases more in the Low-carbon Scenario than in the Reference Scenario. Meeting the Challenge of the Low-carbon Scenario Implementing the proposed Low-carbon Scenario requires tackling a variety of challenges in each of the four areas considered. The combined strategy of releasing pasture and protecting forests to reduce deforestation to 83 percent of historically observed levels involves five major challenges. First, productive livestock systems are far more capital-intensive, both at the investment stage and in terms of working capital. having farmers shift to these systems would require offering them a large volume of attractive financing far beyond current lending levels. thus, a large volume of financial incentives, along with more flexible lending criteria, would be needed to make such financing viable for both farmers and the banking system. a first attempt to estimate the volume of incentives required indicates an order of magnitude of Us$1.6 billion per year, or Us$34 billion during the period. second, these systems require higher quali- fications than traditional extensive farming, which is used to move on to new areas as soon as pasture productivity has degraded, eventually converting more native vegeta- tion into pasture. therefore, the financing effort should be followed by the intensive development of extension services. a third challenge is preventing a rebound effect: the higher profitability of needing less land to produce the same volume of meat might trigger an incentive to produce more meat and eventually convert more native forest into pasture. Such a risk is espe- cially high in areas where new roads have been opened or paved. Therefore, the incen- tive provided should be selective, especially in the amazon region, and should be given only when it is clearly established, based on valid and geo-referenced land ownership title, that the project will include neither conversion of native vegetation nor areas con- verted in recent years (e.g., less than 5 years). Fourth, several attractive options in the Low-carbon scenario to mitigate emissions 33 or increase carbon uptake amplify the requirement of freeing up pasture to prevent carbon leakage. For example, while replanting the forest to comply with the Legal re- serve Law would remove a large amount of carbon dioxide (Co2) from the atmosphere, this area would no longer be available for other activities. The equivalent additional amount of pasture would need to be freed up; otherwise, a portion of production would have to be reduced or more native forest would eventually be destroyed elsewhere. A more flexible legal obligation regarding forest reserves would make the goal of accom- modating all agriculture, livestock and forestry activities without deforestation less difficult, but it might also mean less carbon uptake. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Final Remarks Brazil harbors considerable opportunities for ghg emissions mitigation and carbon uptake, positioning the country as one of the key players to tackle the challenge posed by global climate change. This study has demonstrated that a series of mitigation and carbon uptake measures are technically feasible and that promising efforts are already under way. yet, implementing these proposed measures would require large volumes of investment and incentives, which may exceed a strictly national response and re- quire international financial support. Moreover, market mechanisms would not be suf- ficient for Brazil to harvest the full range of opportunities to mitigate ghg emissions. Public policies and planning would play a pivotal role, with land competition and forest protection management at the center. 1 Introduction the urgent need to combat global climate change has been firmly established. an overwhelming body of scientific evidence, including the Fourth assessment report of the Intergovernmental Panel on Climate Change (IPCC 2007) and a recent review on the economics of climate change led by nicholas stern (stern 2007), underscore the se- vere risks to the natural world and global economy. According to Stern, how we decide 34 to live over the next 20–30 years—how we treat forests, generate and use energy, and organize transport—will determine whether the risks of global climate change can re- main manageable (stern 2009). Managing Risk: Target Levels Failure to hold greenhouse gas (ghg) concentrations below certain levels would imply a great risk to our planet. Recent studies have put forward various target levels, all of which would need emissions to peak soon. the IPCC (2007) concluded that stabi- lizing ghg concentrations at 550 particles per million (ppm)—the level at which it may be possible to hold the rise in global mean temperature under 3°C above pre-industrial Technical Synthesis Report | Land Use, Land-Use Change, and Forestry levels—would require concentrations to peak no later than 2030 and then fall drasti- cally by 2050; in this scenario, the IPCC estimates global emissions would need to be reduced to about 29 gt Co2e by 2030. another recent study, conducted by the United nations Framework Con- vention on Climate Change (UnFCCC), projects that emissions will reach 61.5 gt Co 2 e by 2030. In this scenario, annual emissions from annex I (indus- trialized) countries would go from 21 gt Co 2 e to just 22.1 gt Co 2 e by 2030, while the bulk of global emissions—50–70 percent of the emissions-mitigation poten- tial—would come from non-annex I (developing) countries. despite the range of un- certainty, developing countries clearly have a vital role to play in shaping international policies and actions to cut emissions back to the required scale. It is difficult to imagine an effective solution to stabilizing ghg concentrations at the The Brazilian Context: Key Role of Forests and Other Sectors required scale without Brazil playing a prominent role. the amazon rainforest, which covers more than half the country, is a reservoir of about 100 billion tons of carbon, se- questering more than 10 times the amount of carbon emitted globally each year. given Brazil’s large forested areas—second only to those of Indonesia—it is perhaps not sur- prising that most of the world’s emissions from deforestation emanate from these two countries. at the same time, Brazil is likely to suffer from the adverse effects of climate change. some advanced models suggest that much of the eastern part of the Brazilian amazon region could be converted into a savannah-like ecosystem before the end of the century. this phenomenon, known as amazon dieback, combined with the shorter-term ef- fects of deforestation by fires, could reduce rainfall in the Central-West and northeast regions, resulting in smaller crop yields and less water available for hydropower-based electricity. as the world’s largest tropical country, Brazil is unique in its ghg emissions pro- file. In prior decades, the availability of large volumes of land suitable for cultivating crops and pasture helped to transform agriculture and livestock into key sectors for sustaining the country’s economic growth. In the past decade alone, these two sectors accounted for an average of 25 percent of national gdP. the steady expansion of crop lands and pasture has also required the conversion of more native land, making land- use change the country’s main source of ghg emissions today. at the same time, Brazil 35 has used the abundant natural resources of its large territory to explore and develop renewable energy, having built numerous large hydropower plants and scaled up bio- ethanol production as a gasoline substitute, which accounts for the low carbon inten- sity of its energy matrix. apart from land use, land-use change, and forestry (LULUCF), Brazil ac- counts for only 2.3 percent of global ghg emissions; but until a few years ago, that percentage used to rise another 3 percent when considering LULUCF. Indeed, the LULUCF sector is pivotal, accounting for about two-thirds of Brazil’s gross Co2e emissions (2008), with two-thirds of that amount represented by deforestation alone. By contrast, Brazil’s energy sector has a per-capita carbon intensity of only 1.9 tCo2 per year—about half the global average and less than one-fifth the average Technical Synthesis Report | Land Use, Land-Use Change, and Forestry of oeCd countries. Were it not for Brazil’s previous large investments in renewable energy, the country’s current energy matrix would be far more carbon intensive, with presumably twice the amount of energy-sector emissions and national emissions 17 percent higher. Four sectors are key contributors to Brazil’s ghg emissions. First and foremost is LULUCF, which covers the forestry dimensions described above. In addition, there are three other major emitting sectors: (i) energy, (ii) transport, and (iii) waste manage- ment. In 2008, the respective emissions contributions of these three sectors were 18, 14, and 5 percent. While waste management’s contribution was low in 2008, it has in- creased more than 60 percent over the past two decades. A National Commitment to Combat Climate Change Climate change has long been a vital part of Brazil’s national agenda. In June 1992, Brazil hosted the United nations Conference on environment and development, known as the rio earth summit, which resulted in an agreement on the UnFCCC and, in turn, the Kyoto Protocol. since then, Brazil has played an active role in the international dialogue on climate change. In 2007, the Brazilian government created the secretariat for Climate Change within its Ministry of environment. the following year, President Luiz Inácio Lula da silva launched the national Plan on Climate Change (PnMC), which put the issue at the forefront of the national agenda. on december 29, 2009 the Brazilian Parliament adopted Law 12.187, which institutes the national Climate Change Policy of Brazil and set a voluntary national greenhouse gas reduction target of between 36.1 percent and 38.9 percent of projected emissions by 2020. Like other developing countries, Brazil faces the dual challenge of encour- aging development while reducing ghg emissions. President Lula echoed this concern in his introduction to the PnMC, stating that actions to avoid fu- ture ghg emissions should not adversely affect the development rights of the poor, who have done nothing to cause the problem. recognizing the need for a low-carbon pathway to growth, Brazil has chosen to benefit from the Clean development Mechanism (CdM), an innovative financial mechanism origi- nally proposed by Brazil, which is defined in article 12 of the Kyoto Protocol. to date, Brazil has initiated more than 300 projects under the CdM. 36 1.1 Context of the Low-carbon Study Objective and Approach of the Low-carbon Study to support Brazil’s integrated effort to reduce its ghg emissions and promote long- term economic development, this study seeks to construct a transparent and internally consistent Low-carbon scenario that could be used by the Brazilian government as a tool for evaluating the elements necessary for building a low-carbon path towards growth. this study on Brazilian emissions is one of the five case studies focusing on specific countries that contribute to the preparation of the Clean energy Investment Framework (CeIF). Technical Synthesis Report | Land Use, Land-Use Change, and Forestry the study emphasized two main aspects: first, it used information from the litera- ture and from existing studies as much as possible to effectively leverage the wealth of information. second, the process emphasized a consultative and iterative approach that involved extensive discussions and exchanges of information with specialists in the field and representatives from the Brazilian government. the team researched the literature exhaustively, and, in a thorough consultative process, met with over 70 acclaimed Brazilian specialists, technicians and government representatives. the con- sultative process, combined with the Bank’s comprehensive knowledge of Brazilian institutions, enabled the team to create partnerships with centres of excellence that are recognized for their national and international expertise in the sectors concerned. General Approach of the Methodology used in the Study the study team analyzed the existing opportunities in each of the four sec- tors identified as the main ghg emitters: land use, land-use change, and forestry (LULUCF); energy; transport; and waste. this summary report only presents the part on land use, land-use change, and forestry. In the complete study, the team created a reference scenario for all four sectors until 2030 based on cur- rent projections and available models for each sector. For the energy and trans- port sectors, the team used existing long-term national and urban plans. how- ever, due to a lack of similar plans for LULUCF and waste management, new models and equations were developed consistent with macroeconomic and de- mographic projections for the energy and transport sectors until the year 2030 . For the LULUCF sector, the team used two complementary models: (i) Land Use Model for Brazil (BLUM); and (ii) sIM Brazil, a geo-referenced spatialization model for allocat- ing land use for specific locations and years, developed by the remote sensing Centre (Csr) of the Federal University of Minas gerais (UFMg). For the waste management sector, the team worked with the environmental agency of são Paulo state (CetesB) to develop sets of equations for modeling waste disposal. The study then evaluated the mitigation and carbon uptake options, assessing all the relevant sub-sectors for each sector; determined the viability of the options investi- gated; and finally, constructed Low-carbon scenarios for each sector. 1.2 Approach of the LULUCF Summary Report 37 the report summarizes specific studies on emissions resulting from deforestation, agriculture, livestock, and ethanol and charcoal production for the iron and steel manu- facturing sector. to develop the LULUCF low-carbon scenario, land use and land-use change were projected in a way that was consistent with the projected liquid and solid biofuels; developed geo-spatial models for soil use; made projections of deforestation, adapting existing modeling exercises; and emissions projections. the next step was to analyze the mitigation and uptake options based on an analy- sis of options for reducing pressure from deforestation and protecting the forests, for mitigating emissions from agriculture and livestock, and for carbon sequestration. An economic analysis was also conducted to reduce the costs of the options proposed. For Technical Synthesis Report | Land Use, Land-Use Change, and Forestry these analyses, the team adapted the “cunha� concept developed by Pacala and socolow (2004), which increases the scale of a particular area or technology to ensure signifi- cant reductions in the ghg emissions that can be deduced from the reference scenario. due to the systemic nature of the LULUCF sector, the team concluded that only using the “cunha� approach was not sufficient. For this sector, they analyzed the country’s potential for carbon uptake on a large scale and for avoiding ghg emissions in other countries through greater ethanol export. For some sub-sectors, including deforesta- tion and land use, the team needed to make new projections whose results would be significantly different from the reference scenario, although the same premises would be used (for example: demand, inflation, and fuel price forecasts). an evaluation of the viability of the options identified was then conducted, including barriers that limit or prevent the implementation of the options analyzed and mea- sures for overcoming them, and environmental and economic benefits. Lastly, a Low-carbon Scenario was developed based on the projection of new land use and land-use change (including the additional extension of land necessary for mitigation and uptake options), on the estimate of the reduction of deforestation, and on projections of emissions reductions. Due to the limited resources, this study did not have a fifth phase, which is still necessary to evaluate the sustainability of the Low- carbon Scenario, including its macroeconomic impact. this report is divided into five parts: an executive summary and an introduction on the low carbon study and the main questions pertaining to land-use related ghg emissions; a chapter on the LULUCF reference scenario; a chapter on the LULUCF Low- carbon scenario; and an analysis of costs for transitioning from the reference scenario to the Low-carbon Scenario proposed. 2. Reference Scenario Brazil’s forests represent an enormous carbon stock. the amazon, a reservoir of about 47 billion tons of carbon, permanently sequesters more than five times the amount emitted globally each year. at the same time, in 2010, Brazil was the world’s second largest emitter of carbon dioxide (Co2) resulting from deforestation—often driven by the need to convert land for agricultural production and livestock pasture. 38 not surprisingly, the land use, land-use change, and forestry (LULUCF) sector ac- counts for more than two-thirds of Brazil’s gross Co2e emissions. of this amount, ap- proximately two-thirds are the result of deforestation, with the remainder being from agricultural production and livestock activities. Conversion of forest land for other land uses results in ghg emissions from the soil, while the digestive process of rumi- nants results in methane (Ch4) emissions. a key challenge for the sector is to identify opportunities to curb the net balance of ghg emissions from deforestation while fos- tering economic growth. this chapter describes the background and development of the LULUCF reference Scenario. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2.1 Emissions from Land Use, Land-use Change, Defores- tation, Agriculture and Livestock 2.1.1 Effects of Land Use and Land-use Change on Emissions There are three major ways that land use and land-use change contribute to carbon emissions: (i) conversion of forest land for other land uses (agriculture, grassland, settlements, etc.), (ii) agricultural production, and (iii) livestock activities. In addition, carbon uptake via reforestation activities affects net ghg levels. 2.1.1.1 Deforestation according to the results of this study, deforestation was responsible for 40 percent of Brazil’s gross emissions in 2008. When forest biomass is destroyed, mainly by fire and decomposition, carbon is emitted into the atmosphere. Brazil has been converting forested areas at a rapid pace (approximately 420,000 km² over the past 20 years). the amazon lost approximately 18 percent of its origi- nal forest cover between 1970 and 2007, the Cerrado lost about 20 percent of its original area between 1990 and 2005, and the atlantic Forest lost approximately 8 percent over the same period (InPe 2009). Between 1990 and 2005, Brazil’s car- bon stock was reduced by 6 billion metric tons, largely as a result of deforestation , an amount that is the equivalent of one year of global emissions, with all sources com- bined. since peaking at 27,772 km² during the 2004-2005 period, Brazil’s de- forestation rates have declined sharply to 11,200 km² in 2007, the second lowest historical annual rate estimated by the deforestation as- sessment program (Prodes) since the year 1988, according to InPe (2008) . this trend continued the following years, a partial reflection of the higher valued Brazilian currency, the real (r$), compared to the U.s. dollar (Us$), which has made export-based production less profitable. Implementation of the Plan of action for the Prevention and Control of deforestation in the Legal amazon (PPCdaM) improved the 39 enforcement of environmental laws through an increase in monitoring capacity, and more rigorous conservation policies for the amazon rainforest, have contributed to this reduction. While the spatial dynamics of livestock and agricultural expansion in the amazon determine the pattern of deforestation at the regional level, deforestation is also af- fected by more wide-ranging dynamics. National and international market forces drive the development of Brazil’s meat and crop sectors. depending on price trends, an array of agricultural and livestock activities compete for land. Many geographical studies have shown that the resulting spatial dynamics are on a national scale. over the past three decades, soybean cultivation has expanded over 1,500 km from south to north (de gouvello, 1999). Technical Synthesis Report | Land Use, Land-Use Change, and Forestry A recent geo-statistical analysis shows that livestock-related activities are the pri- mary reason for the conversion of forest areas, followed by the expansion of agricultur- al production and other phenomena, including migration, opening of paved roads, and land speculation as the main drivers of deforestation (soares-Filho et al., 2009). 2.1.1.2 Agricultural Production ghg emissions from agricultural production are caused mainly by changes in soil carbon stocks, and to a lesser extent by fertilizers and residues, cultivation of wetland irrigated rice, burning of agricultural residues, and use of fossil fuels for agricultural operations. According to the results of this study, direct emissions from agriculture ac- counted for about 6 percent of gross national emissions in 2008. Variations in soil car- bon stock correspond to the loss of organic matter in the soil as a result of a particular land use. 2.1.1.3 Livestock Activities the main source of livestock emissions in Brazil is methane (Ch4) from the diges- tive process of ruminants. According to the results of this study, direct emissions from livestock activities accounted for about 18 percent of gross national emissions in 2008. Livestock emissions are related predominantly to beef-cattle farming. According to the Initial national Communication to the United nations Framework Convention to Cli- mate Change, methane emissions from the beef-cattle subsector were responsible for over four-fifths of the total amount of enteric emissions caused by Brazilian livestock in 1994. thus, this study emphasized emissions from, and mitigation alternatives for this subsector. 2.1.1.4 Forestry-based Carbon Uptake apart from ghg emissions sources associated with land use and land-use change, trees remove Co2 from the atmosphere and store it in the trunk, branches, leaves, flow- ers and fruits, thus generating negative emissions. In Brazil, carbon uptake occurs mainly in the natural re-growth of degraded and production forests. According to the results of this study, it was estimated that forestry-based carbon removal offsets about 4 percent of national gross emissions in 2008. 40 2.1.2 Land Use and Land-use Change Simulation Methodology exploring options for mitigating deforestation emissions first requires projecting future deforestation, which, in turn, requires simulating future land use and land-use change. to establish the reference scenario, the study developed two models: i) Brazil- ian Land Use Model (BLUM) (Box 1) and (ii) simulate Brazil (sIM Brazil) (Box 2). these complementary models were used sequentially. BLUM projected land use and land-use change through 2030. sIM Brazil then allocated this land use and land-use change to specific locations and years, but it was necessary to first determine the area available for the expansion of agriculture and livestock activities. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2.1.2.1 Area Available for the Expansion of Productive Activities Land use and occupation in Brazil were characterized using a combination of re- mote sensing equipment and secondary data, measuring area allocated for pasture and area available for the expansion of farmland, and estimating the area of each municipal- ity with environmental liability. the characterization of the area allocated for pasture was essential as it represents the stock of land already converted for productive pur- poses that could be used for farms and forests if these sectors expand. as the produc- tivity of pastures in Brazil is very low, the intensification of pasture area is one of the most important ways to make the expansion of farmland and production forests viable without affecting the agricultural frontier. the area available for productive use (farm, livestock and production forests) was defined, assuming that there would not be any additional deforestation, in other words, considering only the area available for pas- tures that could be converted for other uses (or more intensive use), considering those pastures in areas with impediments not suitable for farmland (Figure 2). Figure 2: Calculation of available area for the expansion of productive activities 41 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry the total amount of area suitable for the expansion of agriculture and production Source: UFMG forests that does not need to be cleared is 126 million hectares. If one excludes the ama- zon biome and other forests, this total is 89 million hectares. these amounts (Figure 3) apply to pasture in areas without impediment that are suitable for agriculture and forests. Three pieces of information from this analysis are essential for the economic land-use model: total pasture area; available area for the expansion of agriculture and production forests; and calculation of the area that needs to be reforested for the legal scenario. Figure 3: Land use by class, excluding the Pampa, Caatinga and Pantanal biomes 42 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry For the land-use projection model, the pasture area to be converted for agriculture Source: UFMG corresponds to a restricted maximum area to be occupied by agriculture and forests in the projections. This ensures that projections for each micro-region do not result in an expansion of farmland and forests beyond the amount of pasture available. thus, infor- mation on the area available is relevant for projections for the Low-carbon Scenario, which is based on the assumption that any agricultural and forest expansion cannot cause deforestation, and would need to be accommodated in pastures that are suitable for these activities. The convertible pasture area was also important for obtaining data projected for micro and macro-regions. each micro-region was restricted by the maxi- mum area allowed for agriculture and forests, which was defined by the area of pasture to be converted. however, information on convertible pastures was not used in reference scenario projections, as it did not reflect land allocation for the expansion of farms and forests only in the pasture area. In other words, the Reference Scenario considered that any additional projected demand for land would lead to a conversion of residual vegetation by micro-region in areas without impediment (table 3). In this scenario, projections of total expansion area (the sum of the farm areas, pastures and production forests) cannot exceed the area of residual vegetation as presented below. additional demand for land in the Reference Scenario was much less than the available land with residual vegetation. Table 3: Comparison between total pasture area and area of residual vegetation convert- ible into farmland/forests in the regions of the BLUM model (1000 ha) Total Convertible Pasture Residual VegetationConvert- BLUM Region South 18,146 5,681 6,721 Pasture for Farms/Forests ible into Farms/Forests 43 Southeast 44,053 30,335 16,415 Central-West Cerrado 51,200 42,553 30,114 northern amazon 52,551 39,079 167,017 northeast Coast 10,801 0 0 MaPIto and Bahia 32,138 8,365 40,319 Brazil 208,889 126,014 260,586 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2.1.2.2 Economic Land-use, Agriculture and Livestock Modeling: BLUM Model as part of an institutional partnership for the Low Carbon study, the Institute for International trade negotiations (ICone) developed a land-use projection model for Brazil – the BLUM (Brazilian Land Use Model) together with the Food and agriculture Policy research Institute (FaPrI) of the Center for agricultural and rural development (Card) at the University of Iowa. For the analyses, the model divides the country into six main regions based on their homogeneity in the production and marketing of agri- culture and livestock, as well as the division of biomes (Map 1): (1) south – Paraná, san- ta Catarina, rio grande do sul; (2) southeast – são Paulo, rio de Janeiro, espírito santo, Minas gerais; (3) Central-Western Cerrado – southern Mato grosso, goiás and part of Mato grosso within the Cerrado and Pantanal biome; (4) northern amazon – Part of Mato grosso in the amazon biome, amazonas, Pará, acre, amapá, rondônia, roraima; (5) MaPIto and Bahia – Maranhão, Piauí, tocantins, Bahia and (6) northeast Coast – alagoas, Ceará, Paraíba, Pernambuco, rio grande do norte, sergipe. earlier projections obtained in the six regions were divided into micro-regions of the IBge. this division is necessary for calculating the balance of ghg emissions of the livestock sector and for the spatialization of the results. Map 1: Map of the Main Regions of the Land-use Model 44 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry the BLUM is divided into two modules that are interlinked to make land-use projec- Source: Based on data from IBGE. Artwork: ICONE tions: (1) supply and demand and (2) land allocation. the first module is based on par- tial balances of the supply and demand of selected products for each year. Demand con- sists of three components: domestic demand, net exports (exports minus imports) and final stock (only the demand for milk, eggs and meats does not include the final stock variable). supply is made up of two components: production and initial stock (this is also only for grains and sugar cane and its derivatives17). The quantities supplied and demanded are calculated simultaneously based on the microeconomic principle of market balance, according to which the supply and de- mand of each product are equal. This balance occurs when there is a price that leads to the convergence between the supply and demand during the same period of time. The model uses Microsoft excel 5.0 as a platform for its operations and the price is adjusted 17 In the case of sugar cane, only the stocks of its derivatives were considered: sugar and ethanol. annually depending upon the excess demand for each product. the process continues until a balance is achieved and the excess is zero. The demand for each product is estimated nationally based on econometric equa- tions. the explicative variables of the domestic demand were generally: income per capita, population, price of the product in Brazil and trends, among others, with these variables considered differently for every product. For the demand for beef, for ex- ample, the domestic prices of competing meat, such as chicken and pork were also con- 45 sidered in the consumer’s decision. For net exports, global economic growth, domestic prices in US dollars and, in some cases, domestic production and the international fuel price were considered explicative variables in the equations. In sum, the model is based on the following central hypotheses: • The equilibrium price is obtained when the supply and demand are equal for a specific year and product. In this way, prices, demand and supply are endog- enous to the model. The shocks given to the model in the Low-carbon Scenario are introduced exogenically via supply or demand. In the case of ethanol, for ex- ample, as will be discussed later, a shock is given to exports, and new balances in the market are observed for all products. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry • The land allocated for each activity and year is the result of the market equilib- rium. For operations, the area of a farm in a given region and for a specific period is a function of the expected profitability, which, in turn is calculated based on productivity, the projected cost for that year and the price of the previous year. • The model works with prices for producers and consumers following the same tendency over time. This means that the change in demand in relation to price variations is based on prices estimated by producers. • The model assumes perfect availability of resources for investments and work- ing capital, which means that it is not impacted by a credit crisis for the supply and demand. For the results used in this project, given that 2009 was a credit cri- sis year, some specific adjustments were made for the 2009 production with the aim of reproducing the expectations for this year more precisely. • the farms’ regional productivity and total recuperable sugar factor (açúcar total recuperável - atr) are projected as tendencies over time. the model is still not ready to capture climatic impacts or different levels of fertilizer use in productivity. • Prices are established nationally and transmitted to regions using price trans- mission coefficients estimated by regressions. although it is not the object of this study, the impact of transport infrastructure improvements on regional production may also be evaluated. • Production costs were divided into three categories: fixed, variable and trans- port. however, the costs of animal products only include variable costs. • The Reference Scenario projects the evolution of the cattle herd, assuming that there will be no significant productive gains over time through improvements in zootechnical indices such as age at the time of slaughter and rate of animal repositioning. however, since the model has no endogenous zootechnological indices, the equations of profits and the repositioning of bullocks are altered to adjust their responses to price variations. Land-use projections for agriculture and livestock were made using the BLUM Land-Use Projection Model, which is an econometric model that operates at two levels: national supply and demand for every final product, and regional supply and area allo- cated for agricultural products, meaning that a set of parameters was estimated based on a temporal data base. The parameters are generally: price elasticity of demand and 46 income elasticity, price elasticity of supply, and cross elasticity. one of the decisions related to methodology was the choice and selection of prod- ucts covered by the BLUM Model, bearing in mind that it would be impossible to de- termine supply and demand frameworks for all of Brazil’s agricultural products, and that there is a concentration of land use for some products. Calculations for designing the land-use model take into consideration the total demand for products; demand for grains; domestic demand for cotton, rice and beans; domestic demand for corn; domestic demand for soybean, soybean meal and oil; net cotton exports, soybean meal and oil; net exports of corn, rice and beans; demand for ethanol and sugar; demand for beef, pork, chicken and eggs; demand for milk and dairy products; national supply and production of each product; production of corn, soybean, soybean meal and oil, cotton, Technical Synthesis Report | Land Use, Land-Use Change, and Forestry rice, beans, sugar and ethanol; allocation of planted area; allocation of area with grains, corn, soybean, cotton, rice and beans; beef supply; projections for the herd, slaughter and average weight of beef cattle at time of slaughter; pasture area; supply of pork; and supply of chicken and eggs. table 4 presents a series of land-use data for products cov- ered by the BLUM model and table 5 summarizes the sources of data and information used by ICone in the land-use model. Table 4: Brazil - Area allocated and production of products covered by the BLUM model 2006 2007 2008 2006 2007 2008 area allocated (ha) Production (1,000 ton) 844 1,080 1,066 2,724 3,899 4,108 3,018 2,967 2,881 11,722 11,316 12,108 Cotton 2,694 3,052 2,857 1,893 2,106 1,991 Rice 1,529 1,035 1,143 1,578 1,234 1,523 Beans – 1st crop 9,632 9,421 9,656 31,332 36,311 39,922 Beans – 2nd crop 3,332 4,634 5,052 11,183 15,059 18,664 Corn – 1st crop 22,749 20,687 21,334 55,026 58,392 60,052 Corn 2nd crop winter 6,179 6,964 8,235 457,246 549,905 687,758 soybean 5,269 5,455 5,874 n.a. n.a. n.a. sugar cane 208,889 206,323 205,381 n.a. n.a. n.a. Production forest 264,136 261,618 263,479 n.a. n.a. n.a. Pasture Total Source: IBGE; CONAB; UFMG/ICONE/EMBRAPA. Note: n.a. = not applicable Table 5: Data sources Instituto Brasileiro de geogra- Beef cattle herd, swine source data utilized Reference fia e estatística – IBge (Brazil- herd, slaughter of fowl, www.ibge.gov.br ian Institute of geography and swine and beef cattle. statistics) Population estimate. 47 Companhia nacional do abas- Planted area, area har- tecimento - ConaB (national vested, prices, costs, sup- www.conab.gov.br Commodities supply Corpora- ply (balance of supply and tion) demand). Empresa Brasileira de Pesqui- sas agropecuárias – eMBraPa Prices and production of www.embrapa.gov.br (Brazilian agricultural resear- swine and fowl. ch Corporation) Ministério do desenvolvi- mento, Indústria e Comércio International commercial www.mdic.gov.br Technical Synthesis Report | Land Use, Land-Use Change, and Forestry – MdIC (Ministry of develop- data. ment, Industry and trade) Instituto de Pesquisa Econômi- Macroeconomic data from ca aplicada – IPea (Institute for Brazil and the agriculture www.ipeadata.gov.br applied economic research) and livestock sector. Centro de estudos avançados em economia aplicada – Ce- Accompanies prices and www.cepea.esalq.usp.br Pea (Center for advanced costs. studies in applied economics) Banco Nacional do Desenvolvi- Credit data and invest- mento econômico e social – ments in the sugar alcohol www.bndes.gov.br Bndes (national Bank for eco- sector. nomic development) Costs and productivity of www.agroconsult.com.br Agroconsult sugar cane farms and fac- (Fábio Meneghin) tories. stratification of beef cattle www.scotconsultoria.com.br herd, beef cattle slaughter, scot Consultoria (Maurício de Palma noguei- prices and profitability of ra) livestock. União da Indústria de Cana-de- açúcar – UnICa (Brazilian sug- Sugar and alcohol market. www.unica.com.br ar Cane Industry association) associação nacional dos Fabri- Annual vehicle sales per cados de Veículos automotores fuel type, scrappage cost www.anfavea.com.br – anFaVea (national auto- curve. makers association) agência nacional do Petróleo, gás natural e Biocombustíveis Prices of gasoline, diesel, – anP (national Petroleum, www.anp.gov.br energy market. natural gas and Biofuel agen- cy) Food and agricultural Policy International microeco- www.fapri.org research Institute - FaPrI nomic data and modeling. 48 Instituto rio grandense do ar- Rice market. www.irga.rs.gov.br roz – Irga associação Brasileira das In- dústrias de Milho – abimilho Corn market. www.abimilho.org.br (Brazilian Corn Industries as- sociation) associação Brasileira das In- dústrias de Óleos Vegetais – Soybean market. www.abiove.gov.br aBIoVe (Brazilian Plant oil Industry association) Technical Synthesis Report | Land Use, Land-Use Change, and Forestry sindicato nacional das Indús- trias de alimentação animal Animal feed market. www.sindiracoes.org.br – sindirações (national Union for animal Feed Industries) Leite Brasil (Brazil dairy) Data on dairy cattle herd. www.leitebrasil.org.br União Brasileira de avicultura – UBa (Brazilian aviculture Data on aviculture. www.uba.org.br Union) associação Brasileira da Indús- tria Produtora e exportado- ra de Carne suína – aBIPeCs Data on swine herd. www.abipecs.org.br (Brazilian association for the Port Production and export) Pne data – export and Empresa de Pesquisa Energé- consumption of ethanol, tica – ePe (energy research demand for biodiesel from www.epe.gov.br Company) soybean, gnP Brazil and global gnP and fuel price. Macroeconomic projections: Land-use projections are based on a macroeconomic Source: ICONE scenario, which shows the trends of the global gnP, Brazilian gnP, Brazilian popula- tion, inflation, exchange rate and fuel price for a 22-year period: 2009 to 2030. For the reference scenario, gnP projections, fuel price, and exchange rate refer to the B1 sce- nario, “surfing the Marola� of the national energy Plan 2030 (Pne-2030). according to Pne projections, Brazil should grow 3.7 percent between 2009 and 2020, and 4.5 per- cent from 2021 to 2030. the overall gnP growth rate will be 3 percent per year for the entire period projected. For projecting the country’s population, data from the Brazil- ian Institute of statistics (IBge) were used. the macroeconomic scenario considered is important, as it entails components of demand and cost equations. table 6 summarizes the macroeconomic scenario used for 2006, 2008, 2018, and 2030. Table 6: Macroeconomic projections 49 5.39% 3.53% 3.70% 4.50% Variable Unit 2006 2008 2018 2030 4.07% 2.48% 3.00% 3.00% gnP Brazil % per year 186.77 191.87 214.94 236.74 global gnP % per year 67.00 63.50 53.07 42.67 Population Brazil Million 2.17 1.66 3.35 4.77 Fuel Price Us$/barrel 4.72 6.02 3.36 2.46 nominal exchange Rate R$/Us$ Inflation Rate % per year Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Source: National Energy Plan (PNE 2030) and ICONE 2.1.2.3 Allocation of Area for Agriculture and Livestock Activities The land allocation module for the different products and regions is a component used to estimate the production of every product (grain and sugar cane) in each region, thus determining one of the components of the Brazilian supply of each product. equa- tions for the allocation of area are estimated for grain and sugar cane for each region and, considering that productivity per hectare was estimated as a tendency, the pro- duction of each of these products is the result of the increase in the productivity and area of each region. Brazilian production of each product is the result of the sum of the regional production of that product. Beef production is calculated based on estimates of the number of animals slaughtered and the estimated average weight of the carcass- es. The allocation of area in each region in the case of grains and sugar cane was esti- mated considering the regional profitability of each crop and the competing crops as explanatory variables (which are negatively related). this means that the regions that present the greatest returns expected for each product will have a greater allocation of land for that product. In addition, area allocation equations for most grains and sugar cane consider the actual area estimated during the previous period as the explanatory variable, thus avoiding major variations in the estimated areas. Equations for the allocation of pasture area were structured differently for grains and sugar cane. the size of the area allocated for pasture in each region is obtained based on the areas used for other crops (and not expected profitability) and the es- timated development of the herd. This option was chosen after different attempts to estimate pasture land were made. As there are different technological levels and pro- duction systems, allocating the area based on the profitability of the livestock did not produce satisfactory results. It was thus decided to estimate the pasture areas as de- scribed above. In addition, since there is no historical model for pasture areas in Brazil, ICone put together a historical series based on the annual series of beef cattle herds by region (IBge, 2008b) and thus the area was projected based on the variables men- tioned above, given that this is also an innovative model in that respect. the area allocated for production forests constitutes exogenous projections for the land-use model, based on projections from the iron and steel and paper and cellulose sectors of the grupo Plantar and the national energy Plan – neP/2030 (Brazil, 2007). to specify the explanatory variables of the land-use equations for each product and 50 region, which determine land allocation by product in each of the six regions, a land competition matrix was developed, as described in table 7. this matrix was defined based on agricultural aptitude criteria (eMBraPa, 2008a; eMBraPa, 2008b; nIPe/ Cgee, 2005) and the trends of the planted areas observed between 1997 and 2008 (ConaB, 2008; IBge 2008a). the area of each crop corresponds to its own expected profitability and the expected profitability of competing crops (through cross-price elasticity). It should be emphasized that the historical returns also show activities that “take� and “give� land in land competition models. Livestock, for example, is historically an activity that “gives� land in all regions, which means that it can be assumed that farm areas compete with pasture areas, but not vice versa. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Table 7: Land competition matrix in Brazil Product (dependent variable) sugar Corn Cotton soybean Corn Rice Beans Pasture cane Sugar Sugar Soybean Soybean Soybean Cane Cane Corn Rice Corn Corn Soybean Rice Bean south sugar Cane Corn Sugar Sugar Soybean Soybean Soybean Cane Cane Corn Corn Bean Corn Soybean Rice south- sugar Cane Region and Competing Product east Corn Cotton Cotton Soy- Soybean Soybean Soybean Soybean Sugar Sugar bean- Corn Cotton Corn Corn Cane Cane Central- Corn Rice Bean Corn Soybean West Cer- sugar Cane rado Corn Soy- Soybean northern Soybean Soybean Soybean Corn Soybean bean Corn amazon Corn Corn Rice Corn Rice Bean Soybean Bean Corn Corn sugar Cane north- Rice east Soybean Corn Cotton Corn Soybean Coast Corn Cotton Soybean Soybean Soy- Corn Rice Corn Cotton MaPIto Bean bean Rice Bean Bean and Ba- hia Source: ICONE It is important to note that the regions are independent and equations for the alloca- tion of area for grains, sugar cane and pasture are different in each region. however, as total production of each product should equal total demand, if a crop’s area is reduced in one region, the price of the product derived from this crop tends to increase, and the other region then makes up for this effect by expanding the area allocated for the crop. the model’s rationale was based on the principle that equilibrium prices of supply and demand would determine crop profitability in each region and consequently the area 51 allocated for each crop (in the case of grain and sugar cane). the co-linearity observed between the historic series of profitability of some crops, such as soybean and corn for example, which should also be highlighted, made it nec- essary to exclude one of these variables or to create a new variable in the land-use equations for certain regions. specifically, for regional projection equations for areas planted with sugar cane, the profitability of soybean as an explanatory variable had to be excluded. on the other hand, for estimating pasture area, it was decided to add up ar- eas of soybean and corn, creating a new explanatory variable for the land-use module. Since equations for the allocation of area for grain and sugar cane were based on the crops’ expected profitability, the greatest profitability compared to other crops and pastures will lead to the expansion of the area allocated. For example, consider a Technical Synthesis Report | Land Use, Land-Use Change, and Forestry hypothetical situation where the most profitable crop in a specific region is sugar cane, followed by soybean, maize and lastly beef cattle, with an increase in demand for all of these products. even with competition between the crops due to their profitability, there will be an increase in the areas allocated for crops, with the exception of pasture area. This is due to the greater potential to increase the productivity of beef cattle com- pared to other products, which is done by decreasing pasture size and maintaining – or increasing – herd size. there is thus no completely proportional compensation for area between grain and sugar cane if there is pasture that can be converted for agricultural purposes. In other words, the area estimated for a given crop can be reduced in one re- gion and increased in another, but this has to do with profitability, and is not the result of a compensation process between the different regions. In the case of areas allocated for pasture, the expansion of the area for grains and sugar cane inevitably leads to a reduction in pasture area if an increase in the size of the cattle herd doesn’t preclude it. however, history has shown that pasture area does not normally increase in regions with strong competition for land from grains and sugar cane. In fact, the opposite holds true in productive regions that are not traditionally used for the abovementioned products. Thus, if the demand for beef increases and if there are regions with stable or decreasing herds, implying a reduction in pasture area, the size of the herd will inevitably increase in regions along the agricultural frontier, leading to an increase in pasture area. It should be noted that a reduction in pasture area in certain regions will only lead to an increase in pasture area on the agricultural frontier if there is an increase in the size of the beef cattle herd. the BLUM treats increases in on-farm productivity as a tendency that reflects past gains. For a more substantial gain in productivity compared to the reference scenario, it would be necessary to consider technological changes in the model (such as exog- enous technological shocks), which implies changes to the structure of the entire farm in terms of production costs. This is a very strong hypothesis, as productivity levels of Brazilian and international farms are comparable, and profits will have already been incorporated in the future as a linear tendency in time. Moreover, the large volume of pasture area in Brazil, some of which is relatively un- productive, shows potential for agricultural expansion. thus, livestock production may be increased by using better zootechnical indices and improving pasture quality, while reducing the area used. Since this is the most realistic technological change for farms, expansion should occur in pasture areas. In other words, according to microeconomic theory, considering, in simple terms, a technological tree with capital and land factors, 52 and that livestock uses a relatively large amount of land and little capital, a slight varia- tion (increase) of the capital factor would cause a more than proportionate change (reduction) to the land-use factor18. Furthermore, by improving livestock productivity, we will be working on the existing Production Possibilities Frontier (PPF/FPP), which is different with grains, where it would be necessary to use technological innovations that are not yet available, thereby displacing the PPF. thus the preference for focusing on beef cattle in Brazil for most of the gains in productivity would be justified. Considering that the BLUM estimates the allocation of area for the six large regions described above, individual crop substitution levels for short periods of time are not captured. the model’s objective is to estimate the allocation of area as a function of the competition between crops and pasture area. Projections of area allocation thus mea- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry sure land-use change resulting from the supply and demand dynamic for all the prod- ucts that compete for land. the relationship of cause and effect due to the expansion of one crop over another over time and implications from substituting between grains, sugar cane, and pastures can be measured a posteriori using the results of the model if certain assumptions are made. given the fact that the equations for the allocation of area in one region are independent from other regions, a set of assumptions that relate land-use changes in traditional regions to those on the frontier must be established, which is the only way to measure the indirect effect of land-use change. however, it is important to emphasize that to analyze the expansion of the agricul- tural frontier and measure the indirect effect of a specific crop, two different aspects must be considered. First of all, there is an increase in the area on the frontier due to the loss of crop area in other regions, which could be considered the actual indirect effect of the crops. secondly, the frontier expands to some extent as a result of pasture develop- ment due to the insufficient increase in beef cattle production despite increases in the demand for meat. In addition, the model’s results could be used to measure the indirect effect of allo- cating pasture area. This will only occur if the beef cattle herd is redistributed between regions as a result of the expansion of other agricultural activities, and after discount- ing increases in productivity due to the intensification of beef-cattle grazing. If this re- distribution does not occur, there will be no indirect effect. An important result from the model has to do with the total area used for agrosil- vipastoral activities (considering the products selected in this analysis). If this area increases over time, areas with native vegetation will be converted into productive areas. this surplus in area allocation is the result of a combination of two factors: (a) an 18 For this affirmation, the existence of the basic concept of decreasing marginal yield should be recognized. increase in the size of the cattle herd in regions on the agricultural frontier (northern amazon, MaPIto, and Bahia), with a simultaneous reduction in traditional agrosilvi- cultural areas, which could be interpreted as an indirect effect, and (b) the expansion of crops on the agricultural frontier, which is a direct effect. Besides land competition, there are interactions between the sectors analyzed, as well as between products and sub-products. For example, between the meat and grain sectors, the demand for feed from the meat, milk, and egg supply (corn and soybean 53 meal, basically) is one of the components of the domestic demand for corn and soy- bean. In the case of the soybean complex, soybean meal and oil are components of the domestic demand for soybeans, which is determined by the crushing margin. Similarly, ethanol and sugar are components of the demand for sugar cane. The methodological diagram below (Figure 4) summarizes the dynamic of the land-use model developed for this study. Figure 4: Methodological land-use diagram Technical Synthesis Report | Land Use, Land-Use Change, and Forestry as mentioned earlier, results obtained in the model (in the six main regions) were Source: ICONE divided into IBge micro-regions. the criteria used were based on the history of the planted area for each product selected, considering data on the limits of land available for the expansion of productive activities. 2.1.3 Land-use Reference Scenario the reference scenario was developed based on the BLUM land-use projection model, using the Brazilian agriculture and livestock expansion pattern observed in the past. thus, with this scenario, there are no exogenous shocks for any variable consid- ered in the model. The Reference Scenario serves as a basis for comparison with alter- native scenarios that consider the expansion pattern of livestock, agriculture, energy and transport sectors with lower levels of ghg emissions (Low-carbon scenario). It should be emphasized that the demand for ethanol and biodiesel, as well as liquid ethanol exports are exogenous to the model. the consumption scenario for these types of energy was extracted from the national energy Plan 2030 (Pne 2030), produced by the energy Planning Company (ePe) in 2006 and concluded in april 2007 (Brazil, 2007). given the availability of the most recent data during the development of the cur- rent project, the amounts projected in the Pne were updated until the 2008 harvest, after which the variation projected in the Pne 2030 was adopted. ethanol stocks and 54 production are endogenous to the model (see section 3.2). according to the Pne 2030, internal ethanol consumption increases significantly during this period, going from 12.8 billion liters in 2006 to 59.2 billion in 2030. exports should already reach a maximum of 15.8 billion liters in 2020, and drop to 13 billion liters in 2030. In the case of biodiesel, according to the Pne 2030 scenario, and based on 2007 observations, diesel consumption will increase approximately 228 percent, going from 42,784 thousand tons in 2007 to 97,876 thousand tons in 2030. In addition, the minimum percentage of the biodiesel mixture in diesel oil will go from 2 percent in 2008 to 12 percent in 2035. soybean participation in biodiesel production should drop from 88 percent in 2008 to 35 percent in 2035. the result of this scenario is the produc- tion of 802.9 thousand tons of soybean biodiesel in 2008, which will increase to 4,133 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry thousand tons in 2030. Land allocation for planted forests is determined exogenically from the model and represents a restriction on the growth of the other crops. Total area occupied by forests is based on PNE projections, which state the total area occupied by eucalyptus, pine and other tropical woods for all of Brazil in 2010, 2015, 2020, and 2030. to calculate the size of the area from year to year, the constant increase between the periods of time was monitored. Based on the historical series of area occupied by planted forests in each re- gion (1997-2007), the Brazilian projection was divided by region. the participation of each region in Brazil counted as much as the history of the growth of such participation. total area occupied by planted forests in Brazil would go from 5.2 million hectares in 2006 to 8.45 million in 2030, a 60 percent increase. In terms of regional dynamics, the area that stands out is the southern region, which would more than double during that time, reaching 3.7 million hectares in 2030, surpassing the southeast and becoming the largest region. another noteworthy locality is MaPIto and Bahia, whose area would grow 124 percent, reaching 1.5 million hectares in 2030 (table 8). Table 8: Projection of areas occupied by production forests (million ha) South 1,670 1,914 3,712 Regions of the BLUM Model 2006 2008 2030 Southeast 2,452 2,669 2,493 Central-West Cerrado 319 374 533 55 northern amazon 140 149 167 northeast Coast - - - MaPIto and Bahia 688 768 1,545 Brazil 5,269 5,874 8,450 areas considered available for agricultural expansion in the reference scenario Source: PNE, ICONE were those that could be converted into pasture and areas with residual vegetation. only pastures and residual vegetation without impediments according to the UFMg classification were considered, meaning no legal impediments (CUs and tIs), accentu- ated slopes or unsuitable soils. however, the legal impediments of the Permanent Pres- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry ervation areas (PPa) and Legal reserves (Lr) were not taken into consideration. Results of general land-use projections were produced for each large region of the BLUM model for agriculture, pastures and production forests in the reference sce- nario. Although supply and demand projections are not presented here, it is important to emphasize that they are part of the model’s exit data and are determining factors in overall land allocation in Brazil for each activity. as shown in table 9, the projected de- mand for land in Brazil for the year 2018 for the products analyzed will be 263.2 million hectares, meaning that there will be an increase of 1.7 percent in relation to the 259.3 million hectares used for the same products in 2006. this increment is even greater for 2030, with an increase of 6.5 percent in total agricultural area compared to 2006, mov- ing as high as 276.1 million hectares. thus, between 2006 and 2030, there will be a 16.9 million hectare expansion in the area occupied by livestock and agriculture as a result of the conversion of native vegetation. the northern amazon presents the greatest growth for that period, at 24 percent. Table 9: Productive land use (crops, pasture and forests) in the different regions of Brazil (1000 ha) Brazil 259,275 257,297 263,222 276,126 Region 2006 2008 2018 2030 South 34,173 33,561 33,614 34,238 Southeast 54,845 53,517 53,747 53,960 Central-West Cerrado 61,756 61,087 61,843 62,994 northern amazon 56,639 57,695 61,826 70,405 northeast Coast 14,567 14,622 14,913 15,233 MaPIto and Bahia 37,295 36,815 37,678 39,296 Source: ICONE although agriculture and livestock have expanded considerably in absolute terms, this increase could be considered negligible in annual terms. In other words, 16.9 million hectares of deforestation in 24 years means an average annual amount of 700 thousand hectares, well below the average deforestation observed in the Legal amazon alone over the past 10 years, which was about 2 million hectares. the decrease in pasture area in 2030 is accompanied by an increment in the beef cattle herd by 13.9 percent during the same period, indicating that there will be a 14.9 56 percent productivity increase in the sector, moving from 0.99 to 1.13 heads per hectare. Much of this increase in the herd is in the northern amazon, where the total increase will be 20.7 head accompanied by an increment in pasture area of 12.1 million hectares (table 10). Table 10: Land use (1000 ha) in the six regions of the model for the Reference Scenario north-ama- northeast MaPITO and south southeast Central-West zonia Coast Bahia Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 13 5 90 66 402 672 60 71 30 38 249 546 2006 2030 2006 2030 2006 2030 2006 2030 2006 2030 2006 2030 1,241 1,447 122 108 273 227 521 526 62 82 800 842 Cotton 536 341 332 211 59 20 305 584 Rice 282 288 349 355 136 134 174 73 1,289 1,166 763 551 Beans 1a 3,706 3,584 2,134 2,303 680 1,010 621 597 1,406 1,611 1,084 1,187 Beans 2a 967 1,598 290 234 1,363 2,708 341 669 371 398 Corn 1a 8,377 11,474 1,716 1,944 8,322 10,167 2,461 4,076 1,872 2,939 Corn 2a soybean 483 1,292 3,944 7,056 501 1,594 113 110 979 1,214 160 1,435 sugar Cane 1,670 2,831 2,452 2,707 319 910 140 327 0 310 688 1,365 Produc- tion 18,146 13,264 44,053 39,565 51,200 48,395 52,551 64,624 10,801 10,812 32,138 30,399 Forest Pasture The increase in the productive area along the agricultural frontier may be the result Source: ICONE of two different but related phenomena. Firstly, there is a significant increase in the size of the herd on the frontier due to its stabilization in traditional regions and the increase in the demand for meat. In the northern amazon, MaPIto and Bahia, an increase in the size of the herd is expected between 2006 and 2030, of 44 and 13 percent, respec- tively. this may be considered an indirect effect of crop expansion, which occupies pasture areas in central-southern Brazil. In addition, an impact on the frontiers due to the increase observed in crop cultivation is more accentuated in proportional terms in MaPIto and Bahia than in the northern amazon, where the expansion of pasture area is much greater. In MaPIto and Bahia, an increase of 1 and 1.2 million hectares of soy- bean and sugar cane between 2006 and 2030 is projected, respectively. In the northern amazon, soybean increased 1.6 million hectares during the same period, while pasture areas increased 12 million hectares (table 10). While beef-cattle raising is increasing, reducing pasture area in the Central West Cerrado by 2.8 million hectares between 2006 and 2030, soybean and sugar cane will be taking over more area in this region, about 1.8 and 1.1 million hectares, respectively. this indicates that much of the expansion of crop areas will occur on pasture land. the 1.3 million hectare increase in the area for second harvest corn, which will take place between 2006 and 2030, is also noteworthy. although it doesn’t affect land competi- tion, this increase is very important, as it implies an increment in total corn production 57 and thus less need for land for first harvest corn (table 10). The same may be observed in the Southeast, although in different proportions. In this region, there will be a greater increase in the area used for sugar cane, from 3.9 million hectares in 2006 to 7.1 million hectares in 2030. on the other hand, variations in areas used for other crops are not as noticeable, supporting the hypothesis that the region is reaching the limit of expansion in the area used for agriculture. Pasture areas in the southeast will be reduced 4.5 million hectares, with a drop in the beef cattle herd to 2.9 million head, but still with a growth in livestock productivity in the region. the Southeast is the second most important region for dairy production. The dairy cattle herd is practically stable (tables 10 and 11), with a 6.7 million ton increase in milk production. Meat production in the region also increased 221 thousand tons during Technical Synthesis Report | Land Use, Land-Use Change, and Forestry that period. Thus, the reduction in the herd does not mean a loss in production capacity thanks to higher levels of technology. Table 11: Dairy cattle herd (1000 head) – Reference Scenario Region 2006 2008 2018 2030 20,942.81 22,813.01 24,471.55 27,732.54 3,406.60 4,102.30 5,458.26 6,466.67 Brazil 7,186.67 7,091.95 6,865.94 6,997.04 south 3,078.42 3,347.27 3,805.55 4,530.30 southeast 2,636.85 3,428.19 3,828.04 4,524.86 Central-West Cerrado 1,749.15 1,876.15 1,882.13 2,255.77 northern amazon 2,885.12 2,967.16 2,631.64 2,957.90 northeast Coast MaPITO and Bahia Pastures in the southern region will decrease 4.9 million hectares between 2006 Source: ICONE and 2030, while the size of the herd will remain practically constant. a substantial growth in the soybean crop and production forests is expected: 3.1 and 1.1 million hect- ares during the same period respectively (table 10). thus, the south will maintain its considerable participation in soybean production and production forests in the coun- try, with a rather constant total area for agrosilvipastoral activities. the northeast Coast showed slight land-use variations during the period consid- ered. Production forests showed the greatest increase in area – 310,000 hectares be- tween 2006 and 2030; followed by sugar cane – 235,000 hectares; and corn – 204,000 hectares (table 10). this implies that the region is also at the limit of its land occupa- tion, principally due to the edapho-climatic restrictions that impede the productive use of much of its area. It is important to understand how to interpret the phenomenon of herd stabiliza- tion in the frontier regions. Projections for the Reference Scenario basically replicate the general trends observed from 1996-2008, the period for which data on the herd, agriculture and production forests were obtained. What has been observed during this time is that the main determining factor for the expansion of pasture areas in the north- 58 ern amazon region is the increase in the size of the beef cattle herd. this expansion has been rather constant from 1996 to 2006. however from 2006 until 2008 the size of the herd decreased in all regions as a result of an upsurge in the slaughter rate due to the increase in meat exports without a commensurate gain in efficiency in zootechnical in- dices, such as the rate of repositioning and lower slaughter age. With no significant im- provements in zootechnical indexes, the low prices for the animals that were observed the first half of the 2000s motivated ranchers to sell beef cattle for slaughter, reducing the potential to reposition the herd, and driving prices further down. thus, pasture expansion observed in some regions in the reference scenario is mainly due to the effects of herd expansion, while direct competition with agriculture on pastures is a less critical factor. Although results clearly indicate that agriculture is Technical Synthesis Report | Land Use, Land-Use Change, and Forestry shifting to pasture areas, this does not mean that the latter have to move towards the frontier to compensate. Pastures expand on the frontier because of the commensurate increase in the demand for meat, and opportunity costs for herd expansion are lower in this region, thus resulting in an increase in pasture area. This becomes more clear when results per region are observed: the increase in pasture land in the northern amazon is greater than the loss of pasture land in other regions, which is in turn the result of com- petition between agriculture and production forests. Variations in the demand for land for other crops (corn, second-crop winter corn, first and second harvest beans, rice, and cotton) will not be as acute as for soybean, sug- ar cane and livestock. according to the results presented in table 12 and Figure 5, the area for first harvest corn should increase 660,000 hectares between 2006 and 2030. however, the area for second harvest corn will increase 2.3 million hectares in the same period, mostly in the Central-West Cerrado and south. Table 12: Land use (1000 ha) for Brazil - Reference Scenario Cotton 844 1,066 1,320 1,399 Products 2006 2008 2018 2030 Rice 3,018 2,881 2,898 3,231 Beans – 1 harvest st 2,694 2,857 2,380 2,394 59 Beans – 2nd harvest 1,529 1,143 1,281 1,328 Corn – 1 harvest st 9,632 9,656 9,663 10,292 Corn – 2 harvest nd 3,332 5,052 5,402 5,608 Soybean 22,749 21,334 26,023 30,601 Sugar cane 6,179 8,235 10,594 12,700 Production forests 5,269 5,887 7,740 8,450 Total agriculture summer 50,386 51,903 60,814 69,793 Pasture 208,889 205,381 203,003 207,060 Agricultural Area + Pasture 259,275 257,284 263,817 276,853 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Source: ICONE Figure 5: Evolution of the demand for land in Brazil by crop in theReference Scenario - 2006-30 (million ha) the increase observed in the area planted with first and second harvest corn may be explained for the most part by the increase in the demand for feed, due to a boost in pork and chicken production, which increased 79 percent and 66 percent, respectively, during the period analyzed. the area with first harvest beans decreased 0.3 million hectares, and there was a decrease of 0.2 million hectares for the second bean crop. however, bean pro- duction in Brazil should increase 3.5 to 4.9 million tons between 2006 and 2030, due to the increase in productivity expected during that time (table 12). It is important to emphasize that the BLUM land-use model projects productivity dynamics and trends over time. Increases in productivity based on past patterns were examined for the reference scenario. Productivity was found to increase an average of 0.69 and 2.10 percent per year for the crops considered. according to the Pne’s exog- enous estimates, production forests will occupy an area of 8.5 million hectares in 2030, 60 representing an increase of a little over 3 million hectares compared to 2006. Most of this growth occurs in the southern region – approximately 1.2 million hectares (table 12), repeating the expansion tendencies observed in the past. To interpret the results, maps were used representing increments and decrements for each crop modeled during the period studied: sugar cane, cotton, rice, beans, silvi- culture, corn, soybean and pastures. there was a considerable expansion of sugar cane principally in northeastern Paraná, goiás, central western são Paulo (where growth increased), the Minas gerais triangle, Central tocantins, Mato grosso do sul, and the northeast Coast (where growth increased, like in são Paulo). other areas of expansion were also found in the states of Bahia, santa Catarina, rio grande do sul, rio de Janeiro, espírito santo, Piauí, Maranhão, and Mato grosso. In other words, sugar cane is ex- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry panding in all the states where agriculture is present (Map 2). Cotton shows an elevated spatial dynamic (Map 2), and areas where it will be culti- vated on a constant basis between 2010 and 2030 are concentrated in southeast Mato grosso. extensive areas of crop expansion can be seen in southwest Bahia and south- east Mato grosso. Cotton is decreasing in western Bahia and in the state of Mato grosso. the demand for land for cotton is actually fluctuating rather than constant. When demand decreases, areas originally used for cotton are used for pasture and probably other crops during the simulation. Map 2: Dynamics of areas where sugar cane (left) and cotton (right) are grown in the Refer- ence Scenario (2010-2030). Yellow = crop permanence; blue = decrement; red = incr ement rice-growing areas (Map 3) are widely dispersed and less concentrated. Cultivation occurs on a constant basis principally in parts of rio grande do sul, Maranhão, santa Catarina, Mato grosso, Piauí and Pará, and is also expanding in these states (in areas close to where it is already being grown), as well as in Bahia (where it is practically non-existent). there are also areas where it is declining, such as in a small part of rio grande do sul, Mato grosso, tocantins, Maranhão, and Piauí. the spatial dynamics are due to fluctuations in demand in the micro-regions, as national demand has generally 61 remained rather constant. Map 3: Dynamics of areas where rice (left) and beans (right) are grown for the reference scenario (2010-2030). Yellow = crop perman Technical Synthesis Report | Land Use, Land-Use Change, and Forestry ence; blue = crop decrement; red = increment areas where bean growing (Map 3) was more stable during the study period are concentrated in the northeast (Ceará, rio grande do norte, Paraíba, Pernambuco, sergipe and alagoas), where there are also small areas of crop increment. In the states of Maranhão, Piauí, tocantins, and Bahia, there are areas where the crop has both increased and decreased, with expansion occurring mainly in the state of tocantins. the state of Paraná in particular has areas where cultivation has both decreased and increased. Corn (Map 4) is widely cultivated throughout the country and in most of the states where it has remained stable or increased, except for the state of Mato grosso, where there was also an area where the original crop decreased. soybean (Map 4) represents one of the most widespread crops in terms of area in Brazil, mostly in the south, Central south, Minas gerais triangle and in parts of the states of Bahia, Piauí, and Maranhão given that its tendency in the Reference Scenario is to increase in demand. Thus, there are practically no areas where the crop is decreas- ing. areas of expansion predominate, some of which indicate crop intensification in regions bordering the amazon. Map 4: Dynamics of areas where corn (left) and soybean (right) are cultivated for the Reference Scenario (2010-2030). Yellow = crop permanence; blue = crop decrement; red = crop increment 62 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Silviculture, which is practiced originally in the southern, southeast and north- east states, as well as in Pará and amapá, remains relatively constant between 2010 and 2030. spots indicating where cultivation has expanded are sparsely distributed throughout the territory, and appear to recede in central-northern são Paulo state (Map 5). Pasture areas (Map 5) are an important aspect in the simulation model when they are part of the three possible transitions (native vegetation > pasture; pasture > crops; or crops > pasture). they appear to have expanded relatively little in central-southern Brazil, due to direct competition with agriculture and a constant demand for pasture. areas where pasture areas are more stable include the states of Minas gerais (except for the Minas triangle region), Bahia (except for the western region), Ceará, rio de Janeiro, part of rio grande do sul, a large part of Mato grosso do sul and in the states of sergipe, alagoas, Pernambuco, rio grande do norte, and Paraíba. on the other hand, pasture areas have also expanded considerably in the amazon, mainly due to defores- tation. Map 5: Dynamic of planted forest (left) and pasture (right) areas for the Reference Sce- nario (2010-2030). Yellow = pasture permanence; blue=pasture decrement; red =pasture increme 63 nt Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2.1.3.1 Division into Geographic Micro-regions Results obtained for the allocation of area for the reference and Low-carbon Sce- narios in each of the six main regions were spatialized according to the IBge level of micro-regions in order to identify the more dynamic regions that partially determine future areas of expansion of livestock activities. the criteria used were the growth his- tory of each crop and the area available for agricultural use. In the case of sugar cane, since logistics are the greatest restriction for this sector (limiting the distance between factories and canebrakes), the locations of working factories, those under construction and those being planned determine the spatialization of sugar cane production by mi- cro-region over time. The history of the area with planted forests was obtained based on an estimate of forest production data from IBge (IBge, 2008c). the area available for agriculture was estimated by the UFMg and included convertible pastures, or those in areas without any impediment (either legal and/or with steep slopes and unsuitable soils19). areas of residual vegetation without impediment were also considered for livestock, but only in the case of the Reference Scenario. a criterion was developed for prioritizing the different uses, which varied depend- ing on the region, but always with sugar cane coming first – due to its more precise loca- tion between the micro-regions – followed by the other crops in different order, then production forests, and lastly, pastures. Prioritization between crops was defined, tak- ing into consideration the ranking of the region’s most important crops in terms of area planted over the past 10 years. 19 Conversion Units (CU) and Indigenous Lands (IL) are among the legal impediments. Land-use restrictions focus on the rugosity and types of soil. 2.1.3.2 Spatialization of Land-use Change and Deforestation: SIM- BRASIL Model To evaluate emissions resulting from deforestation, a national land use and land-use change spatialization model was constructed (sIMBrasIL), with its accountability for Co2 emissions. For the reference scenario, with regard to deforestation, it was consid- ered that a cycle that enabled the continuity of expansion tendencies and dynamics for 64 agricultural crops and other land uses in Brazil, independent of their consequences on deforestation, with no restrictions, would surpass the legal limits (generating environ- mental liability). this scenario was used as a basis to construct the low-carbon scenar- io, which included this concern, with the aim of decreasing deforestation by reducing the demand for land for agriculture and livestock, which by definition could not gener- ate deforestation beyond the legal limit. Reducing deforestation and recuperating envi- ronmental liability through forest restoration will lead to a reduction in ghg emissions in the low-carbon scenario compared to the reference scenario. The development of the spatially explicit model for land-use change, deforestation and carbon emissions resulting from such conversions evolved in three stages as described in Figure 6 and took into consideration its compatibility with the BLUM model. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Figure 6: Architecture of the LULUCF study, with an emphasis on the components that include the deforestation factor a spatially explicit model for land-use change and soil cover was developed (sIM- Source: UFMG BrasIL – available at www.crs.ufmg.br/simbrasil) after calculating the areas avail- able for livestock expansion and planted forests. the objective of this model is to spa- tialize projects for agricultural expansion and demand for land at the micro-regional level using a 1 km2 cell, with a model by ICone for Brazil as a whole (BLUM Model) for the two scenarios in question: the reference and Low-carbon scenarios. the first part of this model was developed based on material from the data base (Figure 7), which in- cludes variables presented in table 13. Figure 7: Example of the data base prepared for simulations of land-use change and cover 65 Table 13: Description of data developed for the implementation of SIMBRASIL Technical Synthesis Report | Land Use, Land-Use Change, and Forestry hydrography Principal Permanent rivers (MMa) Variable description Declivity shuttle radar topography Mission – nasa calibrated by eMBraPa Elevation shuttle radar topography Mission – nasa calibrated by eMBraPa Suitable areas obtained from the previous stage of the study Plant Cover and obtained from the previous stage of the study Land Use Infrastructure Ports, Waterways, railways (Ministry of transport) highways divided into two classes: Paved and Unpaved (Ministry of transport; Csr) Protected Areas Include the two uses of Federal and state CUs: total protection and sus- tainable use of Indigenous Lands and Military areas (MMa, IBaMa, Csr) Population data on urban population per municipal seat (IBge, demographic Census - 2000) all of the aforementioned data were rasterized at a spatial resolution of 1 km2, the equivalent of matrixes of 4500x4500 cells. In addition to spatial data, the model incor- porates land-use projection tables provided by BLUM for a basket of crops (sugar cane, soybean, maize, cotton, rice and beans), production forests and pastures. however, for each micro-region, there is one land distribution vector passed per year for each of the specified uses and the model seeks to allocate this distribution, the basis being the agricultural aptitude of the land for each crop modeled and production cost factors es- timated by infrastructure and consumer market proxies. the model was implemented based on the ego dynamic platform (Box 1), which was designed to operate on an annual basis on two spatial levels: IBge micro-regions and 1 km² raster cells. the spatially explicit simulation model was implemented on the ego dy- Box 1: egO dYnaMIC (environment for geoprocessing Objects) namic platform (dynamic – environment for geoprocessing objects), an inte- grated software program that consists of an environmental modelling platform. through this platform, which was developed by researchers from the Federal University of Minas gerais in 1998 (soares-Filho et al., 2009) it was possible to develop a diverse range of spatio-temporal models that require analytical opera- 66 tions and/or complex dynamic operations, such as network iterations, feedback, multi-scale approaches, map algebra and the application of a series of algorith- mic complexes for the analysis and simulation of phenomena in time and space. Written in C++ language and Java, the software has a library of operators called functors, which may be understood as a process that acts on a set of entry data, to which a finite number of operations is applied, producing a new set of data as a final product (rodrigues et al., 2007). they are currently implement- ing the most common spatial analysis operations in the geographic Information system (gIs) and a series of others that are created specifically by spatial simula- tions, including transition functions and calibration and validation methods. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry the advantage of the software (its current version 1.2.3. is available on www. csr.ufmg/dinamica) is its flexibility, as it enables the user to construct models by linking up functors, which, once in order, establish a data flow chart. through the graphic interface of the ego dynamic, it is possible to create models by simply connecting the operators through their entrances and exits (ports), which repre- sent connections with specific types of data such as maps, tables, matrixes, math- ematical expressions and constants. the models created appear as a diagram, whose execution follows a data flow chain. three main models were developed through the software that was relevant to the study in question: calculation of available land for expansion, simulation of land-use change and carbon emis- sions from land use and land-use change. the execution of the spatially explicit simulation model for land use and land-use change involved the development of two sub-models. the purpose of the first (Figure 8) is to produce a map based on land use, allocating lands according to the BLUM crop classes (sugar cane, soybean, maize, cotton, rice, beans), plus production forests and pasture. Inputs for the model include the land-use map, the micro-regions map and tables on the demand for land for each crop produced by ICone for BLUM. First, the model identifies the anthropized usable area according to the original land-use map, or potential agricultural expansion areas. then it calculates spatial favorability maps for this expansion, integrating data on agricultural aptitude (assad and Pinto, 2008) and other criteria such as: distance, roads, urban appeal, transfer cost to ports, declivity, and distance and area converted. Figure 8: First part of the spatially explicit model for land-use change and soil cover - land allocation 67 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry The model then calculates the rates of transition for crops and reforestation, divid- ing the areas projected per year by the appropriate pasture area. In case the pasture area is not sufficient, the model converts the area with native vegetation and wild grasses into pasture areas. If there is a decrease in the quantity of an agricultural crop in a micro-region, the model returns this area to the pasture stock. Thus, the model always uses the class of pasture as a temporary stock for transitions between forests, pastures and crops. The spatial allocation uses an automatic cellular mechanism to create marks on the landscape (soares-Filho et al., 2002). In the event that the model is not suc- cessful over time in allocating areas projected by BLUM in a specific micro-region, the residual is passed on to the neighboring micro-regions for an iterative process by area in a later phase, thus generating the best estimate possible in terms of areas with BLUM projections. In the second part, the updated land-use map from 2006 is the main input. other in- puts include the map of micro-regions, maps of transition probabilities calculated ear- lier by complementary models20 and the tables prepared by the BLUM. the simulation model is shown in Figure 9. 20 there are four transition probability maps: Probability of conversion to crops (set of 6 maps + 1 for planted forest), Probability of conversion to pasture, Probability of return, and Probability of regrowth. Figure 9: Spatially explicit land-use change and soil cover model -simulation of land-use change 68 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry The simulation model for the land-use change in question interacts online with the model that projects amazon deforestation (soares-Filho et al., 2008), which uses the demand for land for livestock and agriculture modeled by the BLUM as an input. the model aims to incorporate the indirect causes of deforestation as well as the direct con- version of the demand for land for agriculture and pasture. For the study, the deforestation projection model uses three variables that are es- tablished as constant values for the evaluation of the two scenarios: regional migration rates, and protected areas and infrastructure (i.e. paved and unpaved roads); besides two other variables that model the pressure of agricultural expansion: the rate of ex- pansion of areas occupied by crops and the growth rate of the beef cattle herd, accord- ing to BLUM projections for the two scenarios in question. Inputs include the map of micro-regions and tables on the extension of protected areas, on original forested ar- eas, on crops and on the number of head of cattle in the herd, provided by BLUM, as well as tables on average road density per micro-region. 2.1.4 Calculation of Emissions Associated with Land use, Land-use Change and Deforestation in the Reference Scenario having defined the economic and territorial magnitude of agricultural and livestock activities and their locations, it is possible to calculate the greenhouse gas emissions associated with these activities. This section presents the sequence of emissions from livestock and agriculture. Emissions due to deforestation and uptake associated with the forest area are presented in the next sections. 2.1.4.1 Emissions from Livestock Brazil is one of the world’s main beef suppliers. according to the Usda (2008), the country was responsible for 15.5 percent of global beef production (59.3 million tons of carcass equivalent) and 24.9 percent of exports (7.7 million tons of carcass equivalent) in 2008. almost all of Brazilian production comes from herds in pasture systems. the size of 69 the national beef cattle herd – about 200 million head – and the equally sizable pasture area occupied by livestock (over 170 million ha) has caused some concern about the potential impact of Brazilian beef cattle on the environment. In addition, the lower the animals’ performance, and the longer the time before slaughter, the higher the amount of methane emissions produced per ton of meat. In the case of Brazil, emissions are predominantly related to the beef cattle sub- sector. according to Brazil’s initial national statement at the United nations Frame- work Convention on Climate Change (MCt, 2004), and the first Brazilian inventory of anthropic greenhouse gas emissions (MCt, 2006), total methane emissions resulting from enteric fermentation nationwide were estimated to be 8.8 Mt for the year 1990, and those resulting from animal waste management systems, estimated to be 0.3 Mt, Technical Synthesis Report | Land Use, Land-Use Change, and Forestry amounted to 9.1 Mt. In 1994, methane emissions from livestock were estimated to be 9.8 Mt, with 9.4 Mt attributed to enteric fermentation and 0.4 Mt to animal waste management systems. only annual emissions from enteric fermentation represent 92 percent of the total amount of methane emissions from the agriculture and livestock sector. In 1994, the beef cattle category was responsible for 81 percent of methane emissions from livestock in Brazil, the dairy cattle category contributed 13 percent and other animal categories contributed 6 percent of emissions. the approximate 7 percent increase in emissions in the sector during the period 1990-1994 was predomi- nantly due to the increase in the size of the beef cattle herd. Brazilian beef cattle are mainly raised in extensive pastoral systems with a carrying capacity rate and performance much lower than its potential (IBge, 1998). these facts imply the possibility of reducing emissions by using more technological production systems (e.g. feedlot systems, crop-livestock integration and feedlots) which gener- ate an increase in the animals’ performance, and consequently improved Ch4 and N2o emissions per product unit. In addition, the adoption of more intensive production systems will result in a reduction in the demand for land for cattle, making the occupa- tion of the land by other livestock activities possible without the necessity of opening up new areas. It is also important to note that the restoration of low productivity areas (degraded) may represent a considerable drain of Co2 through the increase in carbon stocks in the organic material in the soil, as productive pastures tend to present higher levels of C in the soil (guo; gifford, 2002; Cerri et al., 2003). one projection of the direct emissions from Brazilian cattle was made by Barioni et al. (2007). the estimate considers a moderate increase in productivity between 2007 and 2025 and points to an increase in efficiency sufficient to counterbalance the higher production necessary to meet the demand. greenhouse gases most commonly associated with livestock activities are carbon dioxide (Co2), methane (Ch4), and nitrous oxide (n2o) (IPCC, 2006). however, Co2 emissions from animal respiration are generally unknown due to the assumption that this carbon is derived from photosynthesis and therefore only represents a return to the atmosphere. however, Ch4 produced by enteric and manure fermentation, and N2o released by the nitrification/denitrification of the nitrogen excreted are sources of net emissions. These molecules assume greater importance due to their higher potential for global warming, which is 21 and 310 times more than Co2 for Ch4 and N2o, respec- tively. 70 anaerobic fermentation is known to produce Ch4, and is a necessary process for eliminating excess hydrogen (Van soest, 1994). the reduction of carbon to produce Ch4 is done by a sub-population of microorganisms called methanogenic bacteria. The growth and abundance of these bacteria in the ruminal environment are aided by the presence of slowly degrading fiber, and are inhibited by the presence of starch and its final product, lactic acid, with the concomitant drop in ph. generally speaking, the more fibrous the food (for a specific level of ingestion), the greater the quantity of Ch4 emitted, and the higher the protein content (and the greater the nitrogen excretion), the higher the nitrous oxide emissions. the more food ingested, the higher the daily Ch4 and N2o emissions for a specific diet. oils and fats also considerably reduce methane production. thus, diet can also influence Ch4 production, with the proportion of energy Technical Synthesis Report | Land Use, Land-Use Change, and Forestry in the diet lost being inversely proportionate to its quality, particularly its energy den- sity. ruminants excrete nitrogen in their feces and urine. Both types of excretion include an endogenous component (from the animal itself), and a food component (as the n from the diet is not used by the animal). n excretion may generally be estimated as the difference between n consumption and its retention in body tissue. For example, a 300 kg animal consuming 7 kg of dry matter/day containing 10 percent of crude protein (n x 6.25) consumes 0.112 kg n/day. If this animal gains 0.9kg/day, with the gain contain- ing 20 percent protein, retention would be 0.029 kg n/day, in other words, the animal would be excreting 0.083 kg n/day. the proportion of n in the feces that is converted into N2o varies depending on the manure management system and environmental con- ditions. In liquid manure management systems (in lagoons, for example), fermentation is anaerobic and Ch4 and N2o production, which is considerable, increases with the am- bient temperature. In the solid management system, and even more with the animal’s direct deposit in pasture areas, aerobic degradation occurs, with much lower produc- tion of these gases. In this case, which better reflects the Brazilian situation, between 0.1 and 0.3 percent of n from the dung is converted into n2o (Loyon et al., 2008). 2.1.4.1.1 Methodology The methodology used for modeling greenhouse gas emissions followed the prem- ise that production, and consequently the herd, are limited by the national demand for meat. In this approach, mitigation alternatives that bring about an increase in herd pro- ductivity result in a reduction in the number of animals for a specific demand that has been determined exogenously. In the present study, the estimate was provided by the ICone consulting firm. Methane and nitrous oxide emissions from ruminants are basically a function of the quantity of food ingested and the quality of the diet. however, an increase in ingestion generally results in an increase in the animal’s performance, leading to a consistent re- duction in methane emissions per production unit, shortening the animals’ life cycle or reducing the number of matrices necessary for producing animals for slaughter. In tropical pastures, which predominate in Brazil, most of the animals’ productivity is associated with greater food consumption and better food quality. These conditions are related to improved production systems. given current productivity levels, Brazil has an unequaled opportunity to increase herd productivity through the adoption of 71 different productive systems. thus, analyses for the projection and identification of emissions mitigation alternatives study prototype farms with a variety of productive systems that reflect different levels of land-use intensification and animal productiv- ity. Characteristic indexes of productivity, herd composition, investment structure and cash flows are attributed to each of these productive systems. To determine the composition of the national herd and the proportion of produc- tive systems, an approach was adopted using exogenous estimates provided by ICone, related to the level of balance between meat production and demand. These data were used to project the size of the herd necessary to meet the demand based on the level of productivity projected (as a function of the composition of productive systems). esti- mates of available pasture areas produced by an economic competition land-use model Technical Synthesis Report | Land Use, Land-Use Change, and Forestry were used as entry data to project the productivity level per area (balanced production / pasture area). In this approach, mitigation alternatives that lead to an increase in herd productivity result in the reduction in the number of animals and not in an increase in total production. Figure 10 presents a diagram of the approach used. Figure 10: Information flow in the analytical model to estimate the quantity of food ingested and methane emissions, the animal’s weight, physiological stage, breed and performance (weight gain, birth rates and milk produc- tion) must be determined. since the animals’ characteristics are heterogeneous in the herd, it is advisable to divide the herd into categories and to calculate emissions, as well as ingestion and emissions (IPCC, 2006) for each category. Categorizing the herd en- ables distinct diets to be attributed to each group, a common on-farm practice, and fa- cilitates herd productivity calculations, which are necessary for estimating the number 72 of animals and composition of the herd. Considering the predominance of beef cattle and the use of dairy cattle for slaughter later on, average milk production was adjusted to include the national dairy cattle herd. In the analysis model developed for this study, the herd was divided into nine cat- egories of animals for calculating methane and nitrous oxide emissions and to deter- mine production costs for each productive system. Table 14: Categories of animals considered in the analysis of livestock emissions Minimum age mínima Maximum age máxima Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Category Variable F0 (In) (Ix) M0 Cows F1 0 12 Bulls F2 12 24 heifers up to 1 year old F3 24 36 heifers 1 to 2 years old Bullocks (up to 1 year old) M1 0 12 heifers 2 to 3 years old M2 12 24 M3 24 36 Calves 1 to 2 years old M4 36 - Calves 2 to 3 years old Calves over 3 years old herd composition, in other words, the percentage of the number of animals in each category, is calculated based on zootechnical indices from the simulated productive system. The percentage of animals in the herd in a dynamic equilibrium is calculated by equations 1-7, based on the zootechnical performance indices attributed to the pro- ductive system. The number of births is calculated using equations 1 and 2, based on the number of cows, and the birth and mortality rates of growing heifers. Where n is the number of births, a is the birth rate (adimensional), ωp is the mortal- ity rate until the first delivery (adimensional), ωi is the mortality of each category for heifers and bullocks (adimensional), and; t(Fi) is the amount of time the animals remain in that category until the first delivery (months). The amount of time they remain in the category is calculated differently for males and females as shown in equations 3 and 4. Where Ix is the maximum age (months) of the animals in each category and βp and 73 βa are the ages from the first delivery until the time of slaughter (months) defined by the system, respectively. to calculate the number of heifers and bullocks, the proportion of 50 percent is as- sumed for males and females and the constant mortality rate ω0 during the year (equa- tion 5). the quantities of animals in the growth categories (Fi, Mi) are therefore calculated based on the quantity of animals in the immediately lower age category, of the same Technical Synthesis Report | Land Use, Land-Use Change, and Forestry gender and mortality rate as that category (equations 6 and 7). Lastly, the number of bulls is calculated based on the bull/cow rate (q) defined by the system in question, according to equation 8. the number of females and males slaughtered (aF and aM, respectively) is calcu- lated according to equations 9 and 10. Carcass production (PC) is then calculated for equation 11. Where CeF and CeM are carcass weight at time of slaughter for males and females, respectively, obtained based on trimestrial research on the slaughter from IBge35and projected by ICone. Calculation of CH4 and N2O Emissions three levels of models were applied for each scenario according to IPCC recommen- dations (2006). the Level 1 model follows the recommendations exactly, applying the factor of 56 kg Ch4/year for each animal after weaning, plus 1 kg Ch4/year for manure; as well as the factor of 0.36 kg n excreted/day, together with its rate of conversion into N2o (0.2%; Loyon et al., 2008). For Level 2, the distribution of the animals in the differ- ent herd categories is calculated as follows: cows, heifers, 1-2 year old bullocks, 2-3 74 year old bullocks, 1-2 year old calves, young bulls, 2-3 year old young bulls, young bulls over 3 years old, and mature bulls, as described earlier. thus, the performance of each category in each production system is defined based on published data (FnP, 2008). Based on weight and weight gain, liquid energy requirements for maintenance (eLm) are: Where PV = live weight (kg), Cfi = a coefficient that varies according to the animal category, MJ day-1 kg-1 (0.386 for lactating cows, 0.370 for bulls, and 0.322 for other categories). Moreover, there are equations for adjusting the eLm for activity and move- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry ment, with no increase for animals in cow barns; 17 percent for animals in rich pas- tures, small corrals, with flat topography; and 36 percent for animals on deficient pas- ture area, large corrals, with sloping topography. net energy requirements for the gain (eLg) are calculated based on weight and weight gain: (13) Where PV = adult live weight (kg), gPd = daily weight gain (kg/day), and C = coeffi- cient with values of 0.8 for females, 1.0 for neutred and 1.2 for non-neutred males (nrC, 2000). For lactating cows, net energy requirements for lactation (eLI) are calculated based on the production of milk and its fat content (nrC, 1989): Where L is milk production (kg/day) and g is the proportion of fat in the milk (%). For gestating cows, the net energy requirements for gestation (eLp) are calculated based on the demand for/of maintenance: the compositions (digestibility and gross protein content) of the diets for each cat- egory are defined according to commonly observed amounts. the digestible energy values (ed, %) are converted into eLm and eLg: 75 Ingestion of dry matter (IMs, kg/day) for growing animals and for finishing is calcu- lated based on live weight and net energy for diet maintenance and activity: For adult animals, the IMs is calculated according to: Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Based on this set of equations, it is possible to estimate gross energy intake (geI/ IeB) in each animal category: to calculate Ch4 emissions from enteric fermentation, the factor (ym) is applied as recommended by the IPCC (2006) for animals that consume diets that are low in ed (6.5%) or high in grain and ed (3.0%): Besides enteric fermentation, there are Ch4 and N2o emissions from animal manure. For Ch4, the factor of 1 kg Ch4•animal-1•year-1 remains the same. For n2o emission, n excretion (en/ne) is calculated based on IMs, gross protein content of the diet and its digestibility, as well as endogenous nitrogen excretion: N2o production is calculated in the same way as in Level 1, using the rate of conver- sion of 0.2% (Loyon et al., 2008): Level 3 calculations are similar to those of Level 2, except that the equations and coefficients are more specific to Brazilian conditions. For example, the value of Cfi is reduced 10 percent, given the lower maintenance requirements of zebu cattle. how- ever, the ingestion equation includes maintenance, lactation, gestation and weight gain requirements: Based on the ed, the fiber content in neutral detergent (Fdn) as well as lignin (Lig) 76 in the diet is calculated based on data from the nrC (2000): these amounts are used to estimate Ch4 emissions from enteric fermentation, using equation (27) proposed by ellis et al. (2006): Ch4 and N2o emissions from manure are equal at Level 2. once the amount of each productive system in the national herd was determined over time, Ch4 and N2o emissions were calculated as the sum of emissions generated by Technical Synthesis Report | Land Use, Land-Use Change, and Forestry each productive system. Emissions Estimates by Prototypical Systems “Prototypical� farms were defined to study the heterogeneity of productive systems and the possibility of mitigating emissions through changes in the adoption of different types of productive systems, with the following four systems making up a complete system (cow-calf, stocking and finishing): Complete cycle in degraded pastures Complete cycle in extensive pastures extensive cow-calf in pastures + supplemented stocking and finishing in crop-live- stock integration extensive cow-calf in pastures +supplemented stocking and finishing in feedlots. typical zootechnical indices for the simulated productive systems were attributed for each prototypical farm (table 15). the productivity of the prototypical farms (rep- resenting each productive system) was thus calculated based on these indices using Equations 1-11. Table 15: Zootechnical coefficients considered for each productive system extensive cow-calf Complete Complete + supplemented extensive cow- cycle in cycle in stocking and fin- calf + supple- degraded extensive ishing, integrating mented finishing pastures pastures farming and live- in feedlots 77 stock Digestibility of diet during breeding 56.0 62.0 62.0 62.0 Digestibility of diet during stocking 58.0 60.0 60.0 60.0 Digestibility of diet during finishing 58.0 60.0 62.0 72.0 Milk production 1100 1400 1400 1400 Lactation period 7 7 7 7 Birth rate 55% 60% 75% 75% Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Mortality until 1 year old 7% 5% 5% 5% Mortality between the ages of 1 and 2 2% 2% 2% 2% Mortality between the ages of 2 and 3 2% 1% 1% 1% Mortality over 3 years old 1% 1% 1% 1% Rate of cull cows 15% 15% 15% 15% Weight of adult cows 420 420 420 420 Relationship bull/female 30 30 30 30 Male carcass weight 230 250 250 265 Female carcass weight 200 200 200 200 age at first delivery 36 30 30 30 yield male carcass 52% 52% 52% 52% yield female carcass 50% 50% 50% 50% Weight at birth 30 32 32 32 Weight at weaning (males) 160 185 185 185 Weight at beginning of finishing 379 379 379 379 Weight gain during stocking 0.25 0.30 0.40 0.40 Weight gain during finishing 0.40 0.60 0.60 1.20 since Ch4 and N2o emissions in this study are calculated based on the productive system, and it is assumed that the quantity of Ch4 and N2o emissions is dependent on the system and not on the region where the system is located, the proportions of the productive systems were attributed at the national level. as shown in table 16, independent of the level of calculation, the quantity of Co2-e emitted per kilo of carcass equivalent decreases as the system intensifies, being higher in a degraded pasture system, and lower in recuperated pasture systems with supple- 78 mentation, but with finishing in feedlots. In view of this, the accelerated intensification of beef cattle-raising is implicit in the Low-carbon Scenario. Table 16: Greenhouse gas emissions per animal and per carcass equivalent in kg in differ- ent production systems emissions per animal in the herd (kg/year) emissions/product Productive system (kg CO2-e/ kg carcass) 56.38 0.20 1.25 29.65 Ch4 n2o CO2-e Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 51.71 0.22 1.15 21.89 degraded pasture 51.73 0.21 1.15 18.76 extensive pasture 51.53 0.21 1.15 17.64 ILP1 Feedlot 2 Extensive cow-calf and finishing, in crop-livestock integration. For the construction of the reference and Low-carbon scenarios, the size of the na- Extensive cow-calf and finishing in feedlots. tional herd was estimated using an approach similar to that described by Barioni et al. (2007). In this approach, a projection of the amount is adjusted numerically to meet the projected demand for meat (in kg carcass-equivalent). data adopted for the size of the national herd, meat production, and pasture area were the result of simulations from a land-use competition model developed by the ICone consulting firm. Due to the lack of published national statistics on the proportion and geographic distribution of productive systems in the construction of low carbon and Reference scenarios, the size of the national herd was adjusted numerically to meet the projected need for meat (in kg carcass-equivalent). the area, amount and production of each type of system considered were generated based on the following: (a) the sum of meat pro- duction from productive systems equals the production projected by ICone in 2008 and 2030; (b) the sum of the pasture area occupied by productive systems equals that projected by ICone in 2008 and 2030, and; (c) the sum of the amount in each produc- tive system equals that projected by ICone in 2008 and 2030. In 2008 and 2030, the proportion of the systems was interpolated in a linear fashion. Where e represents the national herd (head of cattle), a is pasture area in the coun- try (ha) and Pd is national meat production (t e-carcass/year). the subscript k defines the productive system. PPk is the proportion, in number of animals, of the k-nth produc- tive system (a-dimensional) and tLk is the carrying capacity of the k-nth productive system (head/ha). A substantial linear gain in productivity due to genetic improvement, estimated at 0.3 percent per year (Lobo et al., 2009), was considered. 79 With the aim of providing a basis for calculating the carbon balance in the pasture area for the group responsible for agricultural emissions, carrying capacities provided by ICone for each municipality were used as a proxy for the level of intensification of livestock in the municipality. For mapping low productivity areas, the municipalities were arranged according to carrying capacity and pasture area, adding those with less and more carrying capacity until the pasture area equals 30 percent of the pasture area of the region. All of the municipalities with carrying capacity lower than that which sur- rounds 30 percent of the area were considered municipalities with low productivity. a portion of the areas with low productivity was calculated based on the area of the mu- nicipalities for each micro-region. Estimates were provided as inputs for the analysis of other topics related to LULUCF. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry the quantification of degraded areas also enabled an estimate of emissions over time in spatial terms for other groups. Like production systems in extensive pastures, farm-livestock integration and feedlots have very similar emissions rates per head, enabling emissions estimates for degraded pasture systems for that and other areas for the remaining amount using the following model: the amount of degraded and non-degraded areas (Bd and Bnd) may be calculated as: Where Ld is the rate of carrying capacity for degraded pastures. emissions in systems with degraded (ed) and non-degraded (end) pastures may be estimated as: Where ed is the emissions coefficient per animal (Mg Co2-e head-1 year-1) in a beef cattle production system with degraded pastures, and end is the emissions coefficient per animal (Mg Co2-e head-1 year-1) in a beef cattle production system with non-de- graded pastures. total emissions for the micro-region (et) are thus et = ed + end. 2.1.4.1.2 Reference Scenario Results the expected evolution of livestock production and productivity in the reference scenario is considered, so that the projection of the reference scenario until 2030 includes changes in the composition of productive systems due to variations in the de- mand for beef, area, and herd. the projection of meat production in Brazil for both reference and Low-carbon scenarios anticipates a 35.6 percent increase by 2030, going from 9.7 million tons of carcass equivalent in 2008 to 13.15 tons in 2030. For the reference scenario, a change in pasture area from 205.38 million hectares to 207.06 is projected and from 201.41 million head to 243.2 million head for the herd (table 17). the development of the area occupied by productive systems in the reference scenario is presented in Figure 11, which suggests an increase in the number of cattle, increases in crop-livestock and 80 feedlot integration, and a reduction of the herd in extensive pasture systems and in de- graded pastures. Table 17: Estimates of area, herd, proportion of the herd in productive systems, and emissions for the Reference Scenario area herd Proportion emissions 2008 2030 2008 2030 2008 2030 2008 2030 Productive (million ha) (million head) (% of the herd) Mg CO2-e system 59,524 22,610 22,379 8,500 11,04 3,49 27.974 10.625 degraded Technical Synthesis Report | Land Use, Land-Use Change, and Forestry pastures 132,173 156,858 155,510 184,539 76,69 75,88 178.837 212.219 extensive 5,500 7,141 10,000 12,985 4,93 5,34 11.500 14.932 pastures 8,182 20,500 14,879 37,182 7,34 15,29 17.111 42.759 ILP1 205,380 207,060 202,768 243,205 100,00 100,00 235.421 280.536 Feedlot 2 Total Figure 11: Variation in the pasture area occupied by type of productive system in the Reference Scenario (million ha) 2.1.4.2 Agricultural Emissions greenhouse gas (ghg) emissions in agricultural areas were evaluated, taking into account emissions from soils and fossil energy used in agricultural operations. The fol- lowing were considered ghg emissions from land use: loss of C from the soil; methane emissions from wetland rice fields and biomass burning; and nitrous oxide emissions from burned biomass, fertilizer use, plant residue decomposition, and n mineraliza- tion due to the reduction of carbon stock in the soil. In the case of fossil energy used in 81 agricultural areas, ghg production resulting from the energy consumed in iron and steel production for agricultural machines and tools, and from burning diesel oil for agricultural operations such as plowing, harrowing, pulverization, manure applica- tion, etc., were considered. The energy used to manufacture and transport such inputs as herbicides, pesticides, and fertilizers, is dealt with in the energy chapter of the Low Carbon study, and published in another summary report. It is also considered in the calculations and will be reflected in the twofold accountability of ghg. the quantification of ghg emissions from the soil is an inventory exercise and therefore the IPCC methodology (1966; 2006) was used as a basis. In the case of ghg emissions from the use of fossil energy, technical coefficients were used for crops, for Technical Synthesis Report | Land Use, Land-Use Change, and Forestry which the hours of labor for each agricultural operation were converted into energy units (Megajoules – MJ) and the energy was converted into ghg using as a reference the quantity of Co2 equivalents (Co2, N2o and Ch4) released from burning diesel oil for gen- erating the respective quantity of energy for the agricultural operation (Boddey et al., 2008). ghg were estimated for the more common production systems for cotton, rice, bean, corn, soybean, and sugar cane crops. 2.1.4.2.1 Evaluation of CO2 Emissions from Changes in Soil C Stocks Co2 emissions or uptake resulting from changes in the soil C stocks were estimated using the IPCC methodology (2006), according to which Co2 flows are estimated in- directly through the balance of the net variation in soil carbon stocks due to land-use changes. to estimate net changes in the carbon stocks, the IPCC methodology calls for an estimate of the stocks up to 30 cm deep, distributed by type of use and soil category for a 20-year period. twenty years is assumed to be enough time for the soil C stock to arrive at a level of equilibrium for a specific land use. to calculate C stocks from 2010 to 2030, land-use data since 1990 had to be used. For the methodology to function correctly, the total area of each region to be con- sidered must be the same throughout the entire period. this study estimates ghg emissions in areas used for agriculture as well as for pasture and planted forests, called occupied areas, which are all part of the Reference Scenario. Thus, differences between the years of contraction or expansion of the occupied areas were considered complementary areas, which in reality correspond to the area “under other uses� (e.g. permanent agriculture) and under “native vegetation�. the largest area occupied by agriculture and pastures, between 1990 and 2030 for any given region, corresponded to a zero complementary area. For 1990, it is estimated that, since the complementary area always exceeded the area estimated for “other uses� for that year (data from IBge on temporary summer agricultural areas that were not considered, and permanent ag- riculture areas), the difference corresponded to the area with native vegetation (table 18). starting in 1991, the area’s expansion, making it larger than the sum of the area occupied with “other uses� in 1990, coincided with the reduction of native vegetation. the reduction of the area signified the increase of “other uses�. Table 18: Areas under different uses and total area in 1990 by state 82 Land use Total area state Planted native Farm Pasture Other uses by state 336.89 3,907.44 41.04 4,411.01 1,946.61 10,642.99 forests vegetation 75.12 660.90 11.30 760.20 1,054.33 2,561.86 ro 9.41 436.23 1.11 505.21 2,753.65 3,705.60 aC 10.29 2,579.81 1.41 118.26 0.00 2,709.77 aM 362.01 7,307.78 114.37 8,161.45 12,441.67 28,387.26 rr 1.10 510.92 84.94 167.08 0.00 764.04 Pa 335.14 19,260.40 0.08 0.00 0.00 19,595.62 aP Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 1,372.60 7,530.28 27.84 447.59 0.00 9,378.31 to 863.83 3,404.29 3.01 412.49 0.00 4,683.61 Ma 1,582.17 2,993.38 24.63 832.69 0.00 5,432.87 PI 227.85 1,585.01 5.32 597.03 0.00 2,415.22 Ce 776.01 2,729.40 15.11 99.44 0.00 3,619.95 rn 991.66 2,312.39 13.54 1,060.51 0.00 4,378.10 PB 782.98 848.15 2.24 161.96 0.00 1,795.34 Pe 174.76 1,193.56 2.91 403.53 0.00 1,774.76 aL 1,025.70 16,118.15 297.43 5,072.59 0.00 22,513.87 se 3,110.03 30,585.35 1,707.78 6,932.01 0.00 42,335.18 Ba 231.41 1,389.58 172.74 91.23 0.00 1,884.95 Mg 269.08 1,243.02 25.88 963.03 0.00 2,501.01 es 4,234.49 7,466.69 597.00 4,668.65 0.00 16,966.83 rJ 5,558.22 4,909.67 713.13 3,225.61 0.00 14,406.63 sP 1,776.44 2,586.13 561.55 752.11 0.00 5,676.23 Pr 6,073.59 10,246.60 630.14 0.00 0.00 16,950.32 sC 1,821.76 23,477.38 181.08 3,467.00 0.00 28,947.21 rs 2,377.35 33,185.03 67.83 426.91 0.00 36,057.12 Ms 2,437.54 21,287.09 72.65 0.00 0.00 23,797.28 Mt 76.30 89.47 19.98 52.41 0.00 238.16 gO thus, Figure 12 shows the total area occupied (provided by ICone), as well as the dF complementary area calculated for Brazil through the years, which is also used to cal- culate Co2 emissions/uptake from land-use change. Figure 12: Area of the country occupied by agriculture, pasture and planted forests, and complementary area in the form of native vegetation and other uses, from 1990-2030 83 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry For equation 35, the soil C stock is calculated for the existing land use for a spe- cific year and 20 years earlier (for example: if the area under a specific land use did not change for the current year compared to 20 earlier, the difference between the C stocks is zero, so there are no Co2 emissions or uptake). With ∆C being the variation of C stocks in the soil (soC) for a specific land use, cal- culated for the desired time (soC0) and 20 years earlier (soC0-20); d is the time during which the balance of C in the soil is affected, in other words, 20 years. the soil C stock from a specific land use is calculated in relation to the carbon stock in a reference condition (in other words, native vegetation that was not degraded or im- proved), as shown in equation 36: Where soCref is the soil C stock under native vegetation for the area of interest; F is the change factor of soil C; a is the area under a specific use. the change factor for soil C stock is a product of three factors: (1) land use; (2) land preparation; and (3) residue input. the change factor for soil C stock indicates the amount of C from the native vegetation that remains in the soil after 20 years of land use. the lower the value of F, the greater is the loss of soil C due to the respective land use. according to the IPCC (2006), there is no value for F estimated for different annual crops. however, for the present study, estimates for F are made for each of the six crops analyzed empirically, based on consultations with specialists and published work on C uptake per production system. 84 By analyzing the variations of soil C stock for the different land uses in a given region, one arrives at an estimate of C emissions or uptake. For example, if there is an increase in land use for a specific area with crops with a low F value compared to what was ob- served 20 years before, one arrives at a Co2 emissions estimate through the reduction of the soil C stock. Using this method, the following hypotheses can be made: (i) over time, the C in the soil reaches a level of equilibrium that is specific to soil, climate, land-use, and manage- ment practices; and (ii) linear changes in soil C stocks occur during the transition to- wards a new equilibrium. Changes in soil C stock by crop Technical Synthesis Report | Land Use, Land-Use Change, and Forestry the methodology proposed by the IPCC (2006) enables Co2 emissions to be esti- mated based on the different land uses and recognized areas (country, state, region, micro-region, etc.). C emissions or uptake are calculated based on the balance of C stocks for every land use depending on the size of the area for each one. thus, the final result is expressed in Mg Co2 for the total area. however, the model used to calculate total emissions from agriculture in this study requires an emissions factor per crop for each year of study, expressed as Mg Co2eq ha-1. A new methodology was developed to achieve this. Figure 13 serves as an example: during year n, three crops (a, B and C) occupied a specific part of a known area. during year n+20, crop a lost land to crops B+C. Produc- tion systems for these crops may also have changed over time (ex.: conventional plant- ing vs. zero tillage), in such a way that the respective change factors related to soil C stock, Fa, FB and FC, have also changed from year n to year n+20. Carbon emissions or uptake from Co2 through the change of soil C stocks by crop, based on land-use change, are thus calculated as shown in equation 37: es (emission or uptake of C) = CF + ∆CV (37) Where, CF is the C stock in the area that has not changed over the years and ∆CV is the expansion of the crop area. It is assumed that the crop that expands its area is the one that causes the changes in the soil C stock. thus, for a crop that only loses area, the ∆CV is zero. Figure 13: Fictitious land-use change scheme for three crops (A, B, and C) 85 to calculate CF, the carbon stock calculation methodology was used, like in the IPCC (2006), with an alteration in the use of the change factor for soil C stock (F) (equation 38). Technical Synthesis Report | Land Use, Land-Use Change, and Forestry CF = (Cref x aN+20x ∆F)/(Dt) (38) Where, Cref is the C stock under native vegetation (C in the reference area); aN+20 is the area occupied by crops in the year n+20; and ∆F is the difference in F between years n and n+20. thus, the management of the area does not change over time and the Fs remain the same, with ∆F being zero, hence CF is also zero; Dt is time (according to IPCC, 20 years). To calculate ∆Cv, one must first calculate the sum of the total area that underwent a land-use change for crops in the area or region being studied, which could be the sum of the areas that were reduced, or those that increased, as they are equal in the module. afterwards, for each crop whose area increased, the fraction of this increase (fr) should be calculated in relation to the total area reduced, or increased (always in a module). For crops whose area increased, the calculation of the emissions or uptake of carbon for this crop from the change in soil use (∆Cv) should consider its integrated effect on each crop whose area was reduced, using Equation 39: ∆Cv = {∑ [fra x (-ari) x (Fn+20(g)-Fn(p))) x Cref}/∆t (39) Where fra is the fraction of the increased area; ari is the reduced area for crop i; Fn+20(g) is the change factor of soil C stock in time n+20 for the crop that gained area; and Fn(p) is the change factor in soil C stock in time n for the crop that lost area. estimate of soil C stock under native vegetation for each region Carbon stocks under native vegetation were estimated for regions limited by the crossing of soil classes and vegetation. as the soil map for Brazil (MCt – www.mct.gov. br) shows many different categories, a simplification was done based on IPCC criteria (2006) including texture, clay activity and drainage. For this study, the soil map was simplified in the following categories: Latossols (all latossols); other soils with low activity clay (argissols, Cambissols, Planossols, Plintossols, etc.); soils with high activ- ity clay (Chernossols, Vertissols, alissols); arenous soils (entissols, espodossols, etc.), and hydromorphic soils (Map 6a). the organossol category was initially considered separately, but since it occupies a very small area in the country, it was included in the hydromorphic soils category. For the map of vegetation, the Pampas, Cerrados, shru- 86 bland, amazon rainforest, deciduous forest and atlantic Forest (Map 6B) were consid- ered. Map 6: Simplification of the soils map for Brazil with six soil categories (A); map of veg- etation, with the six categories considered; (B) visualization of soil C stocks under native vegetation in Brazil (C) Technical Synthesis Report | Land Use, Land-Use Change, and Forestry thirty soil x vegetation combinations were created using simplified soil and veg- etation classifications; each was attributed a value of soil C stock based on available published data and soil data bases in the eMBraPa agrogas network, whose map appears in Map 6C. the stocks calculated for the southern and southeastern regions, which were 51.12 Mg C ha-1 and 47.12 Mg C ha-1, are different from those reported for the same regions in the country’s first greenhouse gas inventory (www.mct.gov.br), which were 60.5 and 40.3 Mg ha-1, respectively. Data estimated in this study for other regions seem reasonable. Since the set of data available in this study was inferior to that used in the country inventory, stocks from the inventory from the southern and south- eastern regions were considered. thus, soil C stocks under native vegetation in the country were established for each region of the BLUM Model, as per table 19. Table 19: Soil C stock under native vegetation for each region of the BLUM model BLUM region original C stock (Mg/ha) South 60.5 Southeast 40.3 Central-West 40.1 northern amazon 46.5 northeast Coast 26.2 MaPIto-Bahia 35.0 the stock change factor depends on the type of land use implemented (continuous agriculture, pasture, irrigated rice, etc.), soil disturbance (conventional preparation, zero tillage, etc.) and quantity of residue deposited (basically the quantity of straw and roots returned to the soil after each cycle). It is not easy to attribute values to the change factors for soil C stocks (F) per crop. For example, it is impossible to separate soybean and corn production systems. soy- bean production areas are generally rotated with corn in the summer. In the winter, 87 wheat and oats are the most prevalent crops in southern Brazil, while second-crop winter corn predominates in other regions. Thus, soybean and corn make up one pro- duction system, with few options used effectively for winter crops. For this study, the effect of soybean was considered the same as corn and included the use of winter crops. In the case of cotton, monoculture production systems are still used, and production areas are concentrated in the Cerrado, specifically in Mato grosso, Bahia and goias (Lamas, 2008). It is a crop that produces a little less residue than soybean and corn, which would justify a smaller F. however, they are also rotated with soybean. similarly, the first harvest bean crop produced little residue and was then followed by the second harvest bean crop, principally in Minas gerais, the largest producer in Brazil (eMBra- Pa arroz e Feijão [rice and Beans]) – www.cnpaf.embrapa.br). It was found to have a Technical Synthesis Report | Land Use, Land-Use Change, and Forestry similar effect as the cotton plant with regard to the soil C stock. Wetland rice is also a monoculture, and has great potential for soil C accumulation (IPCC, 2006) when irri- gated or flooded. Under dry conditions, rice is used in areas that have been recently de- forested, or that have been under pasture for long periods of time. For production from residues with a high C/n ratio, and depending upon the conditions of use, a slightly higher value of F is attributed to the rice crop than that of soybean-corn and cotton. the sugar cane crop is continuous and has great potential for maintaining soil C reserves (Cerri et al., 2007; soares et al., 2009). estimate of change in soil C stock for each region soil C stocks for a specific land use require the change factor for soil C stock, in ad- dition to the size of the area under use and the C stock under native vegetation. these factors were not defined for the crops used in this study, so factors for each type of use were estimated based on criteria suggested by the IPCC (2006) and the knowledge of researchers who are part of the study team. statistics on the use of zero tillage in Brazil are not official and the area has not been monitored over the past five years. In the current scenario, about 77 percent of the area under zero tillage is associated with corn and soybean crops, and does not exceed 25 million hectares (dr. Maury sodda, Federação Brasileira de Plantio direto na Palha [Brazilian Federation for Zero tillage into Crop residues) - FeBraPdP, Ponta grossa, Pr, personal communication). Considering the crops used in this study, with the ex- ception of sugar cane and pastures, this area is equal to 66 percent of the area planted in 2008. according to FeBraPdP, this proportion should be maintained if there is no need to change, i.e. if zero tillage is done in a way that is not recommended (saturnino & Landers, 1997), and problems of soil compacting, pests and diseases prompt some farmers to revert to the conventional planting system. Thus, in the Reference Scenario, or baseline, it is considered that 66 percent of the area planted with agricultural crops, except for sugar cane, will be kept under zero tillage until 2030. Zero tillage will be used for 77 percent of the area planted with corn and soybean and 8 percent of the area planted with other crops used in this study. the area under zero tillage considered for the period from 1989 to 2005, was the one available in FeBraPdP (www.febrapdp. org.br). Conventional soil preparation systems used for planting beans generally re- duce the soil C stock (Zinn et al., 2005; Fernside & Barbosa, 1998) more than in the ab- sence of disturbance in the zero tillage system (Zinn et al., 2005; Cerri et al., 2007). In the present study, the change factor for soil C stock was between 10 and 20 per- cent, depending on the region, with greater variations for zero tillage (table 20). In this 88 case, it should be considered that winter crops are diversified and the premises of the system are complied with, thus allowing soil C accumulation (Boddey et al., 2006). Fac- tors considered for conventional planting mean that land use reduces the C stock to lev- els that are 40-70 percent of those under native vegetation, as observed in samplings done in the country (Zinn et al., 2005). Table 20: Change factors for soil C stock for each type of land use Central- Northern Northeast MaPIto- Land use South Southeast West amazon Coast Bahia Cotton 0.50-0.60 0.40-0.50 0.40-0.50 0.40-0.50 0.45-0.55 0.40-0.50 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Rice 0.70-0.90 0.45-0.55 0.45-0.55 0.45-0.55 0.50-0.55 0.45-0.55 Beans 0.50-0.60 0.40-0.50 0.40-0.50 0.40-0.50 0.45-0.55 0.40-0.50 Corn 0.50-0.60 0.40-0.50 0.40-0.50 0.40-0.50 0.45-0.55 0.40-0.50 Soybean 0.50-0.60 0.40-0.50 0.40-0.50 0.40-0.50 0.45-0.55 0.40-0.50 sugar Cane 0.85-0.90 0.85-0.90 0.85-0.90 0.85-0.90 0.85-0.90 0.85-0.90 Degraded 0.80 0.80 0.80 0.80 0.80 0.80 Pasture Productive 1.00 1.00 1.00 0.90 0.90 1.00 Pasture Planted For- 0.80 0.80 0.80 0.80 0.80 0.80 est other uses 0.85 0.85 0.85 0.85 0.85 0.85 Native veg- 1.00 1.00 1.00 1.00 1.00 1.00 etation Note: The impact factor for crops varied depending on the proportion of adoption of zero tillage. The impact Although there are some statistics on pasture area in the country, it is not clear what factor for the sugar cane crop varied depending on the proportion harvested without burning. proportion is considered low productivity or in a process of degradation. In the early 90s, some studies suggested that about 50 million hectares of pasture in the Cerrado areas were showing some degree of degradation (Macedo & Zimmer, 1993; Macedo, 1995). there are no recent estimates available, but it is believed that over 50 percent of the cultivated pasture area is degraded. Under these conditions, the pasture produces less organic matter and the soil tends to present lower C stocks than soil under native vegetation or productive pasture (Fernside et al., 1998; Fisher et al., 2007; Zinn et al., 2005), although reports of soils from pasture areas with more C than those under na- tive vegetation are not uncommon (Fisher et al., 2007; Zinn et al., 2005). In this study, pasture area is considered underutilized based on data on the herd and the pasture area, which would correspond to a carrying capacity of less than 0.55 head ha-1. Based on the work of Fearnside & Barbosa (1998), the change factor of the soil C stock under productive pasture for amazônia is 0.9. For the south, southeast, Central-West and MaPIto, the factor is equal to 1 (Fisher et al., 2007). For the northeast it is 0.9 due to the more limiting climatic conditions. according to Fisher et al. (2007), degraded pasture 89 area would have about 20 percent less C stocked in the soil than productive pastures. thus, the change factor for soil C stock considered for underutilized pasture was 0.8. Lima et al. (2008) concluded that 34 years after removing native vegetation for establishing eucalyptus plantations, the soil C stocks were reduced to 77 percent of the original. no other related studies have been conducted on Brazilian soils. For the representativeness of the study of Lima et al. (2008), the amount of 0.8 was considered for the change factor for soil C stock for planted forests. the area maintained under “na- tive vegetation� has a stock change factor equal to 1. areas under “other uses�, meaning abandoned areas, areas with other crops, or even those not used for agriculture, had change factors for soil C stocks equal to 0.85. Estimate of N2o emissions from the change in soil C stocks Technical Synthesis Report | Land Use, Land-Use Change, and Forestry When the soil loses C due to land use, a certain amount of mineral n is released into the soil. It is considered that 1 percent of mineralized n is released into the atmosphere as N2O (IPCC, 2006). According to the IPCC method (2006), N2o emissions are computed only when the balance of soil C stock for a given region is negative, or when there are Co2 emissions due to a particular land use. To calculate N2O emissions this way, changes C/N ratio of about 12 (Sisti et al., 2004), every 12 units of C that mineralize until Co2 release in soil C stocks are computed as described above. Since the organic material in the soil has a a unit of N in mineral form. Soon thereafter, 1/12 of the quantity of C in the soil lost from a given region represents the quantity of n mineralized. N2O emissions correspond to 1 Estimate of N2o emissions from the soil from the increase in available n with the ap- percent of mineralized N. plication of nitrogenated fertilizers and decomposition of residues N2o emissions increase with the rise of the quantity of mineral n in the soil (Jantalia et al., 2006). according to the IPCC (2006), 1 percent of the n applied as fertilizer or mineralized from residues left after each crop cycle is emitted as n2o. Indirect N-N2o losses are estimated based on the amount of n that enters the soil through fertilizer and the mineralization of residues. since there are no data that en- able the processes of loss per region to be differentiated, a study of the indirect losses would imply the use of a constant loss factor of n (nitrate lixiviation and ammonia vola- tilization) for all areas, and an indirect emissions factor would be applied to this lost quantity, the same for all regions. In the case of fertilizers, it was considered that 1 percent of the quantity of n applied is emitted as N-N2o. If the losses are considered, the factors result in a 10 percent loss of n by ammonia volatilization, with 1 percent lost as n2o, 30 percent of n lost through nitrate lixiviation, and 0.75 percent lost as n2o. this means that, assuming that these processes occur in equal intensity for dryland crops fertilized with n in the country, the emissions factor considered (direct and indirect losses) is 1.325 percent. In the case of residues and mineralized n in the soil, the 1 percent factor was used for direct emis- sions. as these sources do not involve emissions from ammonia volatilization, indirect emissions are attributed to nitrate lixiviation. thus, the emissions factor considered is 1.225 percent. In the case of wetland rice cultivation, 0.3 percent was the emissions fac- tor used for direct emissions (IPCC, 2006). the factor considered for fertilizers for this condition is therefore 0.625, and 5.25 for other sources. 90 Estimate of Ch4 emissions in wetland rice production areas In this study, the wetland rice production area corresponds to the entire area plant- ed with rice in southern Brazil. other regions were considered highland rice producers. In this case, methane production was only calculated for the rice-growing areas of the southern region. results of research conducted in that region by dr. Magda aparecida de Lima, coordinator of one of the sections of this study, suggest that the methane emis- sions factor of 210 kg ha-1 would be reasonable for use in this study (Lima, 2009). the use of a single factor follows the tier 1 of the IPCC methodology of 1996, which is an- nual and is not disaggregated according to the irrigation regime. In fact, ICone’s land- use survey does not define how the land will be used in rice cultivation, so the use of the more complex methodology would not reduce the uncertainty of the study. In the Technical Synthesis Report | Land Use, Land-Use Change, and Forestry wetland rice system in southern Brazil, it was shown that the use of zero tillage reduced methane emissions by about 15 percent (Lima, 2009), an amount considered to be the effect of zero tillage on this crop in the present study. Estimate of N2o and Ch4 emissions from burning sugar cane straw for the harvest the IPCC methodology (2006) was adopted for this estimate, which calculates n2o and Ch4 production as a function of the quantity of dry biomass produced and burned. an 80 percent efficiency rate for burning was taken into account (amount available for sugar cane in IPCC, 2006). With this method, each ton of cane straw burned produces 2.7 kg of methane and 0.07 kg of n2o. the total amount of dry straw (Ms) produced per hectare of cane burned was 13 tons, based on a study done with different plant varieties (Xavier, 2006). these calculations are represented in equations 40 and 41: 2.1.4.2.2 Greenhouse Gas Production from the Use of Fossil Energy this type of ghg production occurs in agriculture from the burning of fossil fuels to generate energy for the synthesis, processing and transport of inputs, as well as from the execution of agricultural operations (soil preparation, planting, crop treatments, and harvesting). nitrogenated fertilizer is especially “costly� in terms of fossil energy consumption, as it is produced using the häber-Bosch process, at high temperatures and pressure, and fed by natural gas. Fossil energy used in urea production was calculated at 5.9 MJ kg-1 n, compared to 3.2 and 5.9 MJ kg-1 of P and K, respectively (Laegreid et al., 1999). other fossil energy inputs come from herbicides and insecticides. Crops such as cot- ton, soybean and sugar cane consume a large quantity of pesticides and herbicides and the use of zero tillage substantially increases herbicide consumption even more. due to the complicated synthesis of herbicides, manufacturing these products industrially requires large quantities of fossil energy, estimated at 452 MJ per kg of active ingredi- 91 ents. however, since the manufacturing and distribution of these products is discussed in other sections of this study, and published in another Summary Report, this section only covers emissions from agricultural operations. Agricultural operations include diesel oil for operating machinery and equipment, as well as fossil energy used in the production, maintenance and eventual disassembly and preparation of equipment (according to international standard Iso 14040), for which the Pimentel (1980) methodology was used as a basis. agricultural operations for each crop were studied for the year 2008 through contacts with specialists and co- operatives, such as CoCaMar (Pr), Fundação Mt, CoMIgo, etc. the total amount of energy expended for each agricultural operation for each crop was converted into Co2 equivalent, considering that the total amount of ghg emitted Technical Synthesis Report | Land Use, Land-Use Change, and Forestry by burning diesel fuel, a standard fuel for generating all the energy for these operations. according to IPCC (2006), each gJ of energy generated by burning diesel oil releases 73.5 kg of Co2eq. 2.1.4.2.3 Synthesis of Emissions from Agricultural Activities total Co2eq emissions from agriculture amount to 2,047.9 million tons for the 2010 to 2030 period (table 21) in the reference scenario. Co2 emissions occur from the emission of fossil energy during crop management, approximately 343.5 Mt Co2, and from the reduction in soil C stocks (585.2 Mt Co2), due to different types of land use, which vary over time. Table 21: CO2, CH4 and N2O accumulated emissions from agriculture from 2010-2030, expressed in CO2eq for the Reference Scenario Co2 N2o Ch4 Total Source Mt Co2eq Variation in soil C stocks 585.2 94.8 - 680.0 Fossil energy 343.5 - - 343.5 Fertilizers - 175.9 - 175.9 harvest residues - 402.9 - 402.9 Burned cane - 12.0 86.5 98.5 Irrigated rice - - 347.1 347.1 Total 928.7 685.6 433.6 2,047.9 N2o emissions amount to 685.6 Mt Co2eq, of which 175.9 Mt Co2eq correspond to n added to the soil through fertilizers; 94.8 Mt Co2eq from the n mineralization in the soil; 12.0 Mt Co2eq from sugar cane burning; and 402.9 Mt Co2eq from N from harvest residues, with soybean residues contributing 51 percent of this total amount. Methane emissions were estimated to be 433.6 Mt Co2eq, only 86.5 Mt Co2eq of which are from sugar cane burning. Eighty percent of methane emissions are from rice grown under flooded conditions. Figure 14 shows how Co2, N2o, and Ch4 emissions occur over the years, expressed in Co2 equivalents. the greatest variations are found with land use-related Co2 emissions. 92 These also include fossil fuel emissions over time, which tend to be relatively constant, compared to the variations of land use-related Co2 emissions. Emissions reductions from burning during the cane harvest had little impact on total Co2eq. It has been noted that, although there are variations, emissions actually increase over time due to the expansion of the area occupied by agriculture, especially soybean, whose increase in the planted area, mainly pastures, has a strong impact on the reduc- tion of soil C stocks, emitting Co2 and N2o. Moreover, as mentioned earlier, residues from this crop contribute a great deal of N2o emissions. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Figure 14: CO2, N2O, and CH4 emissions from agriculture during the 2010-2030 period, expressed in CO2 equivalents in the Reference Scenario the states of Mato grosso and rio grande do sul had the greatest accumulations of ghg emitted from agriculture from 2010-2030 (Map 7). In rio grande do sul, methane emissions from wetland rice represent about 50 percent of total emissions. In Mato grosso, emissions are the result of different types of agricultural land use. the states of Paraná, são Paulo, goiás and Minas gerais, where agricultural activity is intense, also show total emissions of about 100-300 Mt Co2eq. In conclusion, it is estimated that to- tal emissions from land use under agriculture will increase about 42 percent, from 74 Mt Co2eq to 111 Mt Co2eq from 2010 to 2030. Map 7: Total GHG emissions in CO2 equivalent (millions of tons) by state, resulting from agricultural land use 93 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2.1.5 Emissions from Deforestation Brazil has the largest area of tropical forests in the world (representing 56 percent of the national territory – table 22). there is tremendous diversity in the forest for- mations throughout the country, which are distributed in its six different biomes, the main ones being tropical forests (dense and open), which occur mainly in the north; araucaria forests in the south; seasonal forests (deciduous and semi-deciduous), found principally in the southeast; tropical atlantic Forest with a coastal distribution; shru- bland in the northeast; campinaramas in the northwestern part of the state of amazo- nas and in roraima; a variety of savannas and forest formations in the Cerrado in the country’s central region and production forests that represent 1 percent of the forest cover of Brazil, principally in the atlantic Forest biome. With the greatest diversity on the planet, Brazilian forests are essential for maintaining ecological balance on both re- gional and global scales, including the rainfall regime, freshwater supply, biodiversity conservation, preservation of traditional crops, and mitigation of climate change. 94 Table 22: Land use in Brazil between 1990 and 2005 1990 2000 2005 Type (thous. km²) (thous. km²) (thous. km²) Forests 5,200.27 4,932.13 4,776.98 other uses (agriculture, livestock, urban, infra- 3,155.29 3,423.43 3,578.58 structure, etc) Water depth 195.32 159.32 159.32 Total 8,514.88 8,514.88 8,514.88 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry nevertheless, Brazilian forests are being converted for a multitude of purposes at a Source: IBGE (2006) rapid pace (about 420 thousand km over the past 20 years21). the amazon rainforest lost about 18 percent, or a total of 720 thousand km of its original forest area between 1970 and 2007 (InPe, 2009); the Cerrado lost about 20 percent of its original area be- tween 1990 and 2005, and the atlantic Forest about 8 percent during the same period (sosMa, 2005). given the magnitude and complexity of the Brazilian territory, there is a variety of causes and processes involved in the conversion of the native plant cover. the deforestation of the Cerrado biome during the abovementioned period is gener- ally attributed to the expansion of grain and livestock cultivation (Machado et al., 2004; eva et al., 2004; Ferreira et al., 2007), while deforestation in the atlantic Forest biome during the same period occurred due to real estate speculation and the uncontrollable growth of large urban centres (texeira et al., 2009). the causes of the deforestation of the amazon rainforest are complex (soares-Filho et al., 200822) and involve inter-relat- ed regional, national and global socioeconomic and political factors (soares-Filho et al., 2005; nesptad et al., 2006). among the principal causes are the region’s original colo- nization policies and fiscal incentives for the development of activities that triggered an intense migration to the area. Later on, other processes also played a role, such as the expansion of the international market for agricultural and livestock products, sup- ported by the strengthening of the value of the Brazilian real in relation to the dollar; the expansion of wood and livestock exploitation; the increase of agrobusiness; infra- structure development with the opening up and paving of roads; and the absence and inefficiency of state policies that are unable to halt illegal deforestation and regularize land tenure in the region (soares-Filho et al., 2008). 21 about 28.4 thousand km²/year. 22 soares-Filho, B.s. et al. nexos entre as dimensões socioeconômicas e o desmatamento: a caminho de um modelo integrado.In: Mateus Batistella; diogenes alves; emilio Moran, (org.). amazônia. natureza e sociedade em transformação. 1 ed. são Paulo: edusp, 2008, v. 1. As a result of these enormous changes, and despite the recent drop in deforestation rates in the amazon since 2005, as confirmed by InPe (2009), the country’s great heri- tage of forest resources is threatened (table 23). Table 23: Risk of extinction of arboreal forest species in Brazil in 2000 Types Quantity % 95 not threatened with extinction 7,559 95.9% Critically threatened 34 0.4% Threatened 100 1.3% Vulnerable 187 2.4% Total 7,880 100.0% the role of forests in Co2 emissions, the main gas that contributes to the greenhouse Source: FAO (2005) effect, is pivotal, as forests harbor substantial carbon reserves. It is estimated that Bra- zilian forests store a total of approximately 54 billion tons of carbon. In this context, de- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry forestation in Brazil is the process that contributes most to Co2 emissions, with a total of 70 percent of national emissions. the accountability of emissions from deforestation brings Brazil up to fifth place in the list of emitting countries, representing 5 percent of the global total (CaIt_WrI, 2007). estimated deforestation in the reference scenario (Figure 15 and Map 8) is higher than estimates from the BLUM model projections. as explained in the methodology, sIMBrasIL studies amazon deforestation rates that are calculated based on the spatial lag regression model, as well as projections of the demand for land for cultivation. The objective is to incorporate the indirect effect of agricultural expansion into the land-use dynamic in the amazon. Figure 15: Deforestation dynamic in the three main biomes in Brazil in the Reference Scenario (km2/year) 96 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry It is important to note that the modeling for the reference scenario does not incorpo- rate the possible effects of the new objectives of reduced deforestation as announced in the national Plan on Climate Change, such as the strict observance of the Forest Code. These objectives, as well as the observance of new laws on areas of permanent preser- vation and legal reserves, will be considered within the framework of a legal scenario that constitutes one of the elements of the low-carbon scenario proposed by this study, whose results will be presented later on. Map 8: Deforestation in the Reference Scenario (2010-2030) In addition to the spatially explicit land-use change model, there is an online emis- sions accountability model for land use and land-use change. This model incorpo- rates a table of carbon emissions/uptake for each use, and accordingly, the land-use transition. The data was compiled from various sources. A biomass map was used in amazônia (saatchi et al., 2007), while for the rest of the country, data was compiled associating an average biomass amount to each plant physiognomy from the ProBIo map (MMa, 2007), based on recommendations from the Brazilian emissions Inven- 97 tory (MCt, 2004). the uptake of Co2 via secondary regenerating forests and production forests was estimated using maps showing biomass removal potential through the use of natural vegetation and silviculture, respectively, provided by the consulting firm Ini- tiativa Verde. Deforestation-related emissions from the conversion of forests into pastures were calculated using the IPCC methodology (2003). In this case, the difference between car- bon stock from from plant cover from period 1 (t1) and period 2 (t2), in the case of pas- ture, was used. Since biomass from the different plant physiognomies varies spatially, a carbon stock mosaic was assembled (Map 9) and adopted as a basis for calculations. the amounts for this mosaic are between 0 and 276.5 tons of C/hectare. the amount of 4 tons of C/hectare was attributed for pastures. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Map 9: Carbon Stock Mosaic This data was compiled from various sources. A biomass map was used for the Ama- zon (saatchi et al., 2007). For the rest of the country, data was compiled associating an average biomass value according to the recommendations of the Brazilian emissions Inventory (MCt, 2004) for each plant physiognomy from the ProBIo map (MMa, 2007). Co2 uptake from regenerating secondary forests and production forests was es- timated using maps showing the biomass sequestration potential of natural vegetation and forestry, respectively, provided by the Initiativa Verde. Emissions from land-use change are responsible for the positive balance obtained (Figure 16). these emissions are principally the result of the deforestation of forest remnants. about 66.4 to 81.3 percent of total emissions are due to land-use change from the deforestation of the amazon, and 59 to 67 percent of total projected emis- sions. deforestation in the rest of the country is responsible for 30.8 to 13.4 percent of total emissions from land-use change. Emissions from changes in other uses are at- tributed to other types of land conversion, whose participation is between 2.75 and 5 percent of the total amount between the initial and final years of the study, with a maxi- mum participation of 6.38 percent in 2010. 98 Figure 16: Emissions from land use in the Reference Scenario Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2.2 Carbon Uptake Through Reforestation this section presents the databases and the model used to analyze carbon uptake through reforestation, the purpose of which was to evaluate carbon uptake potential estimates in the Cerrado and atlantic Forest biomes. For the systematization of the basic information, the databases were organized in the shP format, which were later exported for the spreadsheet in XLs format. a potential plant biomass model was developed for the Cerrado and atlantic For- est biomes, estimating the atmospheric carbon removal potential through forest res- toration in permanent preservation and legal reserve areas, and establishing carbon uptake potential through the abovementioned reforestation for each micro-region located in these biomes. Parallel to the application of this model to the restoration of native vegetation, the carbon uptake potential through the planting of energy forests, managing the average annual increment also by micro-region (tCo2e/ha/year), was estimated as well. 99 2.2.1 Methodology 2.2.1.1 Details of the Potential Biomass Model the modeling procedures follow the pattern described by Iverson et al. (1994), who developed a spatial model for potential carbon uptake in forests. This means es- timating the potential quantity of plant biomass above and below the soil, excluding anthropic interventions and natural disturbances such as fire, storms, and excessively long dry periods. The model developed for carbon uptake through reforestation as- sumes that the density of plant biomass that a specific region can support is dependent on climatic, topographic and edaphic conditions, but without taking into consideration the cumulative impact of anthropic activities such as pollution, wood extraction, land- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry use change, etc. the application of this biomass potential model is summed up in equation 42, fea- turing the use of four main layers: IBP=I(ICMW)+I(rainfall)+I(topography)+I(soils) (42) Where ICMW = �ndice Climático Modificado de Weck (1970)(Weck’s Modified Cli- matic Index), which includes figures such as temperature and length of the growing season; Precipitation = annual rainfall averages for each locality; topography = altitude and gradient of land; soils = classified according to texture and fertility. an index was attributed to each layer (I) with the maximum value of 25 points, so that the maximum value possible for the model would be 100 points. the climatic index and average annual precipitation represent half of the IBP. The altitude and gradient variables form the topography layer, given that altitude received a maximum number of 13 points and the gradient a maximum of 12. soil type (texture and fertility) represents the remaining 25 percent of the model (Figure 17). Figure 17: Diagram of potential carbon removals by reforestation for the Cerrado and Atlantic Forest biomes 100 Levels Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Weck’s Modified Climatic Index Weck (1961, apud Weck, 1970) developed an empirical model based on climatic data to determine the potential productivity of forests in germany. Later on, the re- searcher expanded his work to include the tropics and developed the following empiri- cal formula related to his index (Weck, 1970) (equation 43), known as Weck’s Climatic Index (ICW): (43) Considering that dt(Celsius) is the diurnal difference between the maximum and minimum average temperatures of the hottest month of the growing season; s (hours) is the average day length during the growing season; P1(dm) is the number of months during which the average annual precipitation is less than 200 mm; P 2(dm) is the number of months during which the average annual precipitation exceeds 200 mm; g(months) is the length of the growing season, which corresponds to the number of months without hydrous deficit; h(%) is average humidity relative to the air; and tm(Celsius) is the average temperature of the hottest month of the growing season. this index is based on the following premises: In the tropics, there is less respiration if the nocturnal temperature is low (dt).the net productivity of biomass is directly proportionate to day length. The relationship between net productivity and amounts of precipitation is not lin- ear. A continuous increase in precipitation above 2000 mm/year will correspond to a successive reduction in the increase of net productivity. Net productivity is directly proportionate to the duration of the growing season. net productivity is directly proportionate to the relative humidity of the air (H) which, in turn, is highly dependent on amounts of precipitation and existing plant cover. Pluviometric precipitation has less of an effect on net productivity if the tempera- ture during the growing season rises. Applying it to the estimate of current biomass in the tropical forests of Asia, Iverson et al. (1994) modified the Weck Climatic Index as follows in equation 44: 101 (44) these authors used the modified index as a basis for evidence that the proportion of biomass production per unit of area for biomass density is constant in all biomes or cli- matic regions (Brown & Lugo, 1982) in mature tropical forests. Moreover, total biomass is the result of the integration of net production as a function of the time it takes to reach maturity. In this study, since the objective is to determine biomass potential, relative air humidity was excluded from the index (h), as it is a variable that is highly correlated with existing vegetation. thus, the WMCI was used in the following simplified formula Technical Synthesis Report | Land Use, Land-Use Change, and Forestry (equation 45): (45) growing season (g) In the tropics, the growing season corresponds to periods during which there is no hydrous deficit. this variable is strongly associated with the dry season, but presents variations depending on plant cover, type of land, and hydrographic basin. As the pur- pose of this study is to provide an estimate for plant biomass production in the Cerrado and atlantic Forest biomes, it was considered that the periods of hydrous deficit would be the months without rain, as this would be an easy-to-obtain and highly reliable vari- able. thus, the growing season (g) is defined as (equation 46): g = 12 - s (46) where s = dry months; Without dry phase – considered an absence of hydrous defi- cit; sub-dry – period of deficit equivalent to one month; 1-2 dry months – deficit equiva- lent to two months; 3 dry months = 3 months of deficit; 4 to 5 dry months = considered 5 months of hydrous deficit. Solarimetry data on daily sunshine (hours) and average day length (hours) during the growing season were used at this level. This section, together with rainfall and temperature data, was used for Weck’s Modified Climatic Index (1970). WMCI values were divided into 25 classes in a non-linear fashion. More values were grouped together in lower classes because the vegetation is more sensitive to WMCI in the dry extreme; a pattern similar to the one used by holdridge (1967). the first 16 classes showed an increment of 25 units; the next 6 classes an increment of 50 units; and the last 3, an increment of 100 units (Figure 18). 102 Figure 18: Points attributed to the WCMI values in the model. Modified by Iverson et al. (1994) Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Rainfall The correlation between precipitation classes and biomass density was assumed to be positive up to the amount of 3,200 mm/year, at which point this correlation begins to have a negative effect on biomass (Brown & Lugo, 1982). according to observations made by Brown et al. (1993), 400 mm annually is the minimum amount needed to sup- port arboreal formations (Figure 19). Figure 19: Points attributed to the amounts of rainfall in the model, according to Iverson et al. (1994) Altitude different authors report that altitudinal zoning changes the vegetation pat- terns, mainly through the climatic variations associated with each altitude class. There- fore, an altitude layer was included in our model. The altitude classes were based on the 103 suggestions of Iverson et al. (1994) and were divided into five classes according to the general variations in vegetation depending on the altitude of the Cerrado and atlantic Forest biomes: 0–15 m – Coastal forest – mangrove 16–50 m – Lowland formation 51–500 m – Sub-mountainous formation 501–1.500 m – Mountain formation + 1.501 – high mountain formation Technical Synthesis Report | Land Use, Land-Use Change, and Forestry the points attributed to each altitude class are shown in Figure 20. the 0-15 meter altitude class received fewer points due to the fact that these plant formations occur throughout the coast and normally have lower biomass values than lowland forests. Figure 20: Points attributed to altitude classes in the model, modified by Iverson et al. (1994) Slope according to Iverson et al. (1994), slope is one of the variables whose correla- tion with forest biomass is extremely variable. Considerable amounts of biomass were already found on relatively sloped lands compared to the amounts found in adjacent flat areas. thus, the slope of the land gets relatively low points in this model, going from 12 points (up to a 10 percent gradient) to 8 points (gradients greater than 20 percent) (Figure 21). Figure 21: Points attributed to the degree of incline of the land, modified by Iverson et al. (1994) 104 Soils Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Different edaphic factors affect the biomass distribution patterns in tropical for- ests (Whitmore, 1984). Forest productivity is generally related to soil fertility, but this potential is much more affected by climatic factors and texture, as large amounts of biomass have been reported in the amazon region in forests growing on nutrient-poor soils, but with suitable texture (Laurance et al., 1999; saatchi et al., 2007). however, to standardize the study, the model adopted the IBge soil fertility map in Brazil, where different degrees of fertility assume the amounts presented in table 24. Table 24: Points of the different IBGE fertility classes Points Fertility 25 Average to high 22 Low to medium 18 Low to very low 15 Low 12 Low to very low 7 Very low Thus, for the development and application of the model, the following data need to be examined: rainfall; solarimetry; soils; temperature; growing season; Weck’s Modified Climatic Index; altitude; Declivity. 105 Data bases used in the model the environmental data bases consist of a set of grids (geo-referenced matrixes) standardized by the overlapping of layers, with auxiliary data in vector format, such as hydrography, official boundaries of the Cerrado and atlantic Forest biomes, political division of the territory and official mapping of the Brazilian ecosystems. Sources used in the data bases vary. The environmental data base stores informa- tion extracted or processed from official cartographic data (IBge), thematic maps (eM- BraPa and JrC) and spatial models (srtM, WorldClim). Environmental databases Technical Synthesis Report | Land Use, Land-Use Change, and Forestry data availability for the different environmental layers in Brazil in 2008 is much greater than that found by Brown & Iverson (1994) for southwest asia, or by Brown & gaston for tropical africa; for the Cerrado and atlantic Forest, data is currently avail- able in greater spatial resolution and thematic diversity. Two important sources, which were not available in the 90s and are available today for almost the entire planet, are based on srtM and WorldClim topography. the srtM Base (shuttle radar topography Mission) consists of a topographic in- vestigation done through orbital radars. The international topographic mission was lead by nasa, and its measures were realized in the year 2000. the resulting digital land model has approximate spatial resolution of 60 m and a margin of error of ± 6.2 in South America. the WorldClim database was developed by the University of California and pub- lished in 2005. data from tens of thousands of meteorological observation posts around the planet were geo-referenced and interpolated, resulting in climatological maps derived from temperature and precipitation data. the data base has 55 grids that describe each month: minimum temperature, maximum temperature and precipita- tion, and bioclimatic variables that are relevant for ecological modeling. The grids cov- er almost the entire globe, except for the polar regions, with a spatial resolution of 30’. although this is a consistent data base established for the international scientific community, we feel that an analysis of the specific area of study might be more accurate. It was thus decided to review the WorldClim data in order to evaluate its accuracy for the study area. In parallel, the re-sampling of the digital model of the srtM land was audited for a 30’ slope. the schedule for the WorldClim audit included the following steps: tabulation of available climatic data from the eMPraPa database “Brazilian cli- matic database� for 35 small cities that are distributed regularly along WorldClim Zone 34, which includes the study area. seventeen meteorological posts were selected from this initial group, and these were described with regard to coordinates, altitude, aver- age temperature, annual precipitation and number of months with precipitation lower than 50 mm. Consultation with WorldClim for the cell amounts (pixels) which include coordi- nates from the 17 meteorological posts selected. the amounts consulted were: alti- tude (re-sampled from srtM), average temperature, annual precipitation, number of 106 months with less than 50 mm of precipitation (this last layer was produced by the re- classification of precipitation for each month [amount<50mm=1, value>50mm=0] and for the summary of all the months of the year). the amounts were collated and tests comparing the results were conducted (ano- Va). all the consultations at WorldClim, analyses of reclassification and overlap were conducted using the software arcView 9.3. For all the comparisons realized, the WorldClim database is not statistically dif- ferent from the eMBraPa environmental database (Levene test, p>0.92). thus, the WorldClim database can be used for modeling, as can be observed in Figure 22. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Figure 22: Graph of the box-plot where the distribution of the amounts for altitude, rainfall, and months of hydrous deficit may be observed for the EMBRAPA and WorldClim databases The climatic variations considered as described in the potential biomass variation were organized in grids measuring 30° latitude by 30° longitude. the spatial resolution is 30 seconds (0.93 km at the equator), resulting in a grid with 3,600 X 3,600 cells. Wgs 84 datum was adopted once Brazil officially abandoned the use of sad 69 and adopted sIrgas. Wgs 84 datum is standard for WorldClim and srtM, and presents only about a 1cm difference for sIrgas. 107 Table 25: Entries in the environmental database ID Variables Unit Base Digital land model 1 Altimetry (m) srtM-nasa 2 Slope (%) srtM-nasa II Precipitation 3 Annual precipitation (mm) WorldClim 4 Duration of dry weather (month) WorldClim 5 Monthly precipitation > 2m – 2m (month) WorldClim Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 6 Monthly precipitation < 2m, >2m=2 (month) WorldClim III Temperature Average temperature of the hottest 7 (°C) WorldClim month IV solarimetry 8 Daily insolation (hs) Solarimetric atlas V soil 9 Fertility Classes IBge/eMBraPa VI soil cover 10 Map of plant cover in Brazil (Fao) JrC / eMBraPa VII Complementary data 11 Limits of biomes of interest IBge Biomes Databases - Maps the study is limited to the Cerrado and atlantic Forest biomes. according to the IBge, the Cerrado biome has an area of 2,063,001 km2 and the atlantic Forest biome 1,112,170 km2. the two have very different floristic compositions: arboreal formations predominate in the atlantic Forest, and savannas and cerrado fields are more common in the Cerrado, presenting, nevertheless, very distinct potential for carbon uptake. Map 10 shows the limits of Cerrado and atlantic Forest biomes. Map 10: Boundaries of the Cerrado and the Atlantic Forest, extracted from the Map of Bra- zilian Biomes, produced in 2004 by IBGE in cooperation with the Environment Ministry. This map indicates the boundaries adopted for the area of the study. 108 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Altitude the relationship between altitude and biomass density is not linear: the extreme altitudes have low amounts of biomass density potential, such as in coastal formations and mountain tops. The high altitudes tend to have little impact on the study area, as Brazil’s altitudes are low on average and the highest are well below the limit for forest formations (3,750m). the altitude with the greatest biomass productivity would be between 16 and 750 meters, which covers almost the entire country. only areas in bright blue or brown have restrictions for biomass development (Map 11). Map 11: Altimetry based on the digital SRTM land model. Original data offer an aver- age altitude with 3’’ resolution. The model presented was re-modeled for 30’’. Declivity according to Iverson et al. (1994) large amounts of biomass density can also be found in areas of accentuated declivity, but there is a tendency for flatter areas to pres- ent higher average densities than sloping areas. the flatter areas are shown in yellow 109 in Map 12. the low declivity levels found are justified by the smoothing effect of the re- modeling for 30’’; however this is not an average declivity model. Map 12: Declivity in percentages based on the digital land model re-modeled for 30’’ Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Rainfall This is one of the main variables for modeling biomass potential. The methodology proposed by Iverson et al. (1994) anticipates two entries for this variable: average an- nual precipitation participates in Weck’s Modified Climatic Index and is alone in the final overlap (Map 13). Map 13: Average annual precipitation in millimetres, from the WorldClim climatic da- tabase. Annual precipitation is directly related to potential productivity; biomass density tends to be greater in blue areas than in red. according to Iverson et al. (1994), an increment in precipitation above 3,000 mm would have a negative effect on biomass productivity. however, average precipitation in the study area does not exceed that amount. growing season The growing season corresponds to the number of months where monthly rainfall is over 50 mm. during periods of low hydrous availability, plants close their stomata to avoid loss of humidity, and often lose their leaves. Co2 absorption from the atmosphere 110 is either non-existent or low at this time, when hardly any plant growth occurs. In the Cerrado biome, the growing season lasts a maximum of nine months, such as in Campo grande and Cuiabá. the most common duration is seven months, as is the case in Brasília, goiânia and Belo horizonte. the growing season in the Caatinga lasts less than five months, but it is outside of the study area (Map 14). Map 14: Length of the growing season indicated by the sum of months with higher precipi- tation than 50 mm, obtained from the climatic WorldClim model Technical Synthesis Report | Land Use, Land-Use Change, and Forestry average values obtained for the growing season in the atlantic Forest are much higher than those of the Cerrado, although they vary greatly. some parts of the atlantic Forest present low values, like the area between the cities of rio de Janeiro and Belo horizonte, and between aracaju and Maceió. This grid was calculated based on the previous map so that the variation in the dis- tribution of rainfall in the WMCI may be incorporated. Weck’s Modified Climatic Index states that the duration of the growth period is directly proportionate to net productiv- ity. In equation 47 (WMCI), the letter “g� represents the number of months of the grow- ing season. Average temperature of the hottest month of the year This item takes into consideration the average temperature of the hottest month of the growing season. It negatively influences carbon uptake potential because the higher the temperature, the more respiration and less net absorption of Co2 (Map 15). Map 15: Average temperature of the hottest month of the year (Celsius) 111 Database - Soils Technical Synthesis Report | Land Use, Land-Use Change, and Forestry the soil map used was the IBge fertility map. a hierarchical fertility organization may be applied for most classes, which would be directly proportionate to the contribu- tion of the edaphic component to biomass potential. the predominance of “very low fertility� soils in the Cerrado is very high, above 70 percent. this class is very rare in the atlantic Forest, as it is associated more with coast- al regions, suggesting a high fertility x potential biomass correlation. on the other hand, more fertile soils from the rio Paraná basin are equally divided between the Cerrado and atlantic Forest (Map 16). Map 16: Soil fertility map for Brazil database – plant cover this map shows the soil cover classification for Brazil. Both anthropic and natural areas adhere to the Fao classification system. the mapping was done by eMBraPa as an integral part of the gLs2000 (global Landcover 2000), a Joint research Center project, responsible for managing an accurate soil cover data base for International Conventions (Climate Change, Combatting desertification, ramsar, and Kyoto Proto- col), and serving as an initial register (Map 17). there is a predominance of agricultural areas in the focus biomes. according to the Fao legend, the Cerrado’s features fall under the savanna classification, and are not characterized as forest formations in most of the remnant areas. For the atlantic Forest biome, natural areas are classified principally as humid and swamp forests, but also with savannas. This map does not participate 112 directly in the IBP calculation, but serves as a reference for calibration as it indicates potential biomass distribution, which is still greatly obscured by anthropic pressure. Map 17: Map of plant cover in Brazil for 2000 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Map 18: Map of Brazilian ecosystems, IBGE, representing an estimate of the distribution of “original� plant formations with a simplified legend indicating areas of transition 113 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2.2.1.2 Carbon Uptake Potential through the Restoration of the Le- gal Reserves the potential for carbon uptake from the atlantic Forest biome was between 183 and 661 tCo2e/ha. The micro-regions that show the greatest potential are generally lo- cated in the southern region and in the state of são Paulo, but specifically in the serra do Mar. as expected, there was less potential in micro-regions with greater hydrous deficit and consequently a shorter growing season, such as in sergipe. For the Cerrado biome, the carbon uptake potential was between 195 and 467 tCo2e/ha in Minas gerais and Mato grosso do sul, respectively (Map 19). Map 19: Map of carbon uptake potential through the forest restoration of the Legal Re- serve in the Cerrado and Atlantic Forest in tCO2/ha 114 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2.2.1.3 Carbon Uptake Potential through the Restoration of River- side Forests With regard to carbon uptake potential in riverside forests in the atlantic For- est biome, the lowest amount observed was in the state of sergipe (231 tCo2/ha) and the highest amount was in Paraná (720 tCo2/ha). For the Cerrado biome, carbon uptake potential was between 220 and 552 tCo2/ha in the micro-regions of Barra (Ba) and Itu- verava (sP), respectively (Map 20). Map 20: Map of carbon uptake potential through the restoration of riverside forests in the Cerrado and Atlantic Forest biomes in tCO2/ha Uptake through forest vegetation was only studied outside of the amazon rainfor- est, due to the uncertainty about its natural carbon balance (nobre et al., 2001). howev- er, secondary forests play a considerable role in the Cerrado and atlantic Forest biomes, as a good part of these remnants constitute secondary forests that are about 50 to 60 years old. A logistical function for calculating potential biomass accumulation was as- sumed that takes into consideration plot location, and current and peak age. The latter 115 variable is assumed to be 200 years (Figure 23) for natural forests, while for reforested areas it is 20 years. Annual uptake, however, takes into account the regeneration of these forests at an average initial age of 60 years. Figure 23: Logistical function of biomass uptake using local biomass potential and age of vegetation as parameters Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Note: The local potential in this example is equal to 100 t/ha. 2.2.1.4 Carbon Uptake Potential through Energy Forest Plantations in the Cerrado and Atlantic Forest Biomes Co2 absorption values by energy forests were defined in the productivity data obtained in the CCaP (2006), which presents minimum and maximum productiv- ity (tCo2/ha/year) for this type of reforestation in Brazil. although data from aBraF (2007) shows significantly higher average productivity rates, they were initially not considered for the modeling, as they incorporate land management and correction practices that were not integrated into this model. Lower amounts were generally observed in the northeast of Brazil, with a minimum of 28.85 tCo2/ha/year in the micro-region of Brumado in Bahia. the northeast average was about 39.83 tCo2/ha/year. For the other extreme, the greatest potential for annual carbon renewal was observed in the state of Paraná, in the micro-region of Capanema. the average for the southern part of the country was 48.45 tCo2/ha/year, while the av- erage for the southeast was 41.28 tCo2/ha/year (Map 21). Map 21: Forest productivity (tCO2/ha/year) for the Cerrado and Atlantic Forest biomes 116 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2.2.2 Reference Scenario for Forest Restoration Even with credit lines at reduced interest rates for native forest restoration, this activity has not been adopted voluntarily. For most rural property owners, reforesta- tion or even simply fencing off areas of riparian forest or legal reserves implies a loss of productive areas. even with non-recoverable financing, farmers are reluctant to give up PPAs for forest restoration and the subsequent elimination of this environmental liability. This is particularly relevant for small rural properties, of which permanent preservation areas may occupy a large part. Forest restoration generally occurs due to legal obligations in most cases, with terms of behavioral adjustments and forest com- pensations. Thus, on the scale of this study, the evolution of native forest restoration in the Refer- ence Scenario may be considered negligible. The largest private initiative of this type was in the state of são Paulo, with the restoration of 12.7 thousand hectares of riparian forest, implemented by the aes tietê along the boundaries of reserves where it is the electricity generation concessionaire. restoration occurs at a rate of 250 hectares/ year23. at the public level, the government of são Paulo has made major efforts to re- store native vegetation through the Projeto de recuperação de Matas Ciliares (riparian Forest restoration Project - PrMC), whose resources come from the global environ- ment Fund (geF) and is expected to restore 1,500 hectares of riparian forest. how- ever, the two initiatives combined do not cover 1.5 percent of the state’s current deficit, 23 http://www.aestiete.com.br/content/doc/aneXo_III_reflorestamento.pdf which is over 1 million hectares. Based on these observations, it was determined that although the contribution of forest restoration in the reference scenario is quantitatively rather limited, the expe- riences themselves are essential from the qualitative point of view in order to under- stand how to overcome the obstacles identified. 117 2.2.3 Non-renewable Charcoal and Planted Forests for Renewable Charcoal The Reference Scenario for the additional use of renewable charcoal was created based on the analysis of two groups of causal factors and their respective impacts on the participation of the three thermo-reduction agents in the Brazilian iron and steel industry. the first group, defined as the main group, constitutes the fundamental axis of the analysis, which is the generalized maintenance of the different obstacles identi- fied, including investments; management; institutional and technological obstacles; and market gaps. the second group, the auxiliary group, is subordinate to the first and involves the degree of legislative deterrence or control over the use of non-renewable charcoal. In this group, two types of sub-scenarios were analyzed: (i) the enforcement Technical Synthesis Report | Land Use, Land-Use Change, and Forestry of a legislative structure that tolerates the large-scale use of non-renewable charcoal and (ii) the enforcement of a legislative structure with a low tolerance level for the large-scale use of non-renewable charcoal. the first group deals with the degree of scarcity of renewable charcoal and exposes the insufficient supply of inputs and absence of additional policies and incentives. the objective of the second group is to determine the thermo-reduction agent to be used in a scenario with scarce renewable charcoal, in other words, either coal or non-renew- able charcoal. the second group thus assumes a purely auxiliary role, as the lack of additional planted forest, as shown in the first group is a strong indicator of the occur- rence of net ghg emissions and the non-occurrence of net uptake. these two legal sub- scenarios essentially serve as key references regarding specific sources of emissions to be avoided in a low-carbon scenario. two possible reference scenarios were identified based on these two groups and are presented below. The resulting emissions appear in the summary report on energy, which only deals with the impacts of land-use change. Reference Scenario with a low level of legal restrictions: In this Reference Scenario, the principal group of premises was adopted in combination with the sub-scenario (i) of the auxiliary group, which was the scarcity of planted forests combined with a low level of legal restrictions. as a result, it was assumed that a 3.7 percent increase per year for iron and steel production would be in keeping with the actual participa- tion of thermo-reduction agents on the market24 in which 66 percent of the process of thermo-reduction necessary for producing iron and steel would continue to be based on the use of coke, 24 percent on the use of non-renewable charcoal, and 10 percent on the use of renewable charcoal, as illustrated in Figure 24 below: 24 According to estimates presented in the report on other topics of this study. The assumption of 3.7% growth per year was adopted based on the national energy Plan. Figure 24: Reference Scenario for charcoal with a low level of legal restrictions; participa- tion of thermo-reduction agents in the Brazilian iron and steel-producing market 118 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Source: Research by AMS, IBGE, ABRAF, SINDIFER therefore, the renewable charcoal deficit will increase in absolute terms, despite a modest growth in the planted area due to the proportional distribution over time. Reference Scenario with a high level of legal restrictions: In this Reference Sce- nario, the principal group of premises was adopted in combination with the sub- scenario (ii) of the auxiliary group of premises: scarcity of planted forests for the production of renewable charcoal combined with a high level of legal restrictions. As a result of the increased legal restrictions regarding the use of non-renewable charcoal, the gradual decrease in the use of this thermo-reduction agent was as- sumed until its complete elimination starting in 2017. this scenario is based on the growing tendency to enforce legal deterrence against the use of non-renewable charcoal observed throughout the country, especially in the state of Minas gerais, which is responsible for over 60 percent (sIndIFer) of iron and steel production us- ing charcoal in Brazil25. Although this scenario takes into consideration the increased rigor in the applicable legislation regarding the use of charcoal, it is important to note that the prevalence of the main group of barriers means that planted forests will continue to be scarce. Thus, the market segment that is currently based on non-renewable charcoal would then be based on coal. This scenario is based on the economic premise that the simple deter- rence of the use of non-renewable charcoal does not automatically generate a relative increase in the supply of renewable charcoal. Figure 25 below illustrates the reference Scenario with a high level of legal restrictions. 25 there are already stringent regulations and legal restrictions in Minas gerais regarding the use of non-renewable charcoal, considering a period of transition that could last 7 to 10 years. these restrictions are the result of a sustainability pact signed between the productive sector, the state government and different non-governmental organizations. Figure 25: Reference Scenario for charcoal with a high level of legal restrictions: participa- tion of thermo-reduction agents in the Brazilian iron and steel-producing market 119 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry given the growing tendency to impose legal restrictions on the use of non-renew- Source: Research in AMS, IBGE, ABRAF, SINDIFER able charcoal, this scenario is more likely to occur than the previous one. There is a stronger argument that one cannot assume that iron and steel producers, regardless of scale, would make the necessary investments in the expansion of the sector based on a non-sustainable option from the environmental and legal point of view (non-renew- able charcoal). It is therefore possible and more conservative to assume that the oppor- tunity cost for new investments in the iron and steel sector must be based on the use of mineral coke or renewable charcoal from new investments in planted forests. Estimates of net and adjusted emissions generated in the aforementioned Refer- ence Scenarios were calculated on this basis. The conclusion was reached that uptake in the scenarios with a high or low level of legal restrictions was the same, as the volume of planted forests is the same in both (table 26). Table 26: Projection of CO2 Emissions and Uptake in the Reference Scenario (use of coal and/or non-renewable/renewable charcoal) - 2010 to 2030 (in thousand tCO2) 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 57,917 62,283 66,977 72,025 77,453 83,291 89,568 96,319 103,579 111,385 119,780 emissions 120 (net) 64,096 68,927 74,122 79,708 85,716 92,176 99,124 106,595 114,629 123,268 132,559 emissions 6.2 6.6 7.1 7.7 8.3 8.9 9.6 10.3 11.1 11.9 12.8 (gross) Uptake* Source: Adapted from data presented in “topic O� report (industry-related emissions) *Note: Uptake by planted forests for the production of renewable charcoal in the Reference Scenarios Figure 26: CO2 emissions projection for the Reference Scenario (charcoal) Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Source: Adaptation of data presented in the “topic O� report (industry-related emissions) 2.3 Reference-Scenario Emissions Results the study team generated an integrated reference scenario for LULUCF based on subsectoral analyses, using the emissions calculation methods indicated above, which were then integrated into the sIM Brazil model. Using these models made it possible to generate maps and tables that registered annual emissions and carbon uptake over the study period, calculated for each 1-km2 plot and integrated by micro-region, state, and country (Figure 27). Figure 27: Reference Scenario results, emissions from land use and land-use change, 2009-30 121 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Emissions from land-use change via deforestation account for the largest single share of total emissions from LULUCF—up to 533 Mt Co2e per year by 2030. direct annual emissions from land use (agricultural production and livestock activities) in- crease over the period up to an annual rate of 383 Mt Co2e. The model shows a decrease in the annual carbon uptake rate, from 28 Mt Co2e in 2010 to 20 Mt Co2e in 2030. For the entire period considered, the net balance between land use, land-use change, and carbon uptake results in increased emissions, reaching about 895 Mt Co2e annually by 203026. 26 When calculating national carbon inventories, some countries consider the contribution of natural regrowth to carbon uptake; therefore, although this study does not compute the contribution in the carbon balance of LULUCF activities, it would be fair to add that information for purposes of comparison. If the carbon uptake from the natural regrowth of degraded forests were to be included, then the potential uptake would increase by 109MtCo2 per year, thus reducing net emissions. 3 Mitigation and Carbon Uptake options Based on the projected evolution of LULUCF-sector emissions in the reference sce- nario (Chapter 2), the study explored opportunities for reducing emissions and scaling up carbon uptake. The study proposes a Low-carbon Scenario for land use and land-use change in 122 Brazil focused mainly on (i) containing national demand for land for crop and pasture expansion in order to reduce emissions from deforestation, (ii) scaling up the identified mitigation options for agriculture and livestock, and (iii) maximizing carbon uptake potential associated with legal forest reserves and production forests. sections 3.1 to 3.6 identify mitigation options through agricultural production (zero tillage), charcoal (forestry-based carbon uptake), carbon uptake potential through re- forestation (Cerrado and atlantic Forest) and livestock and deforestation, respectively. each of these five sections examines obstacles to the adoption of the respective mitiga- tion measures, exploring ways to overcome them and accompanying measures. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 3.1 Mitigation Options in Agriculture: Zero tillage Zero tillage was found to be the most promising strategy in the agricultural sec- tor for reducing agriculture-related ghg emissions. the adoption of the zero tillage system, which entails the elimination of soil disturbance, crop rotation and soil cover maintenance (saturnino & Landers 1998), could increase soil C stocks to a higher level than the conventional land preparation system (sisti et al., 2004; diekow et al., 2005), close to native vegetation levels (Jantalia et al., 2007). Monitoring of C in the soil has been done using the IPCC methodology (1996, 2006), which establishes a depth of 0-30 cm as a reference. Variations of C stocks depend on the area’s history. Published reports resulting from studies that evaluated the effect of zero tillage for cereals and soybean on the accumulation of C in the soil found that rates were 0.5 MgC ha-1 year-1 for the most superficial soil layer. however, this figure is dif- ficult to extrapolate for the entire country, as areas under the same production system for over 20 years (IPCC, 2006) are not experiencing any more changes in soil C stock. Although other studies have shown greater variations in the length of time needed for establishing stocks (Coleman et al., 1997), 20 years may be enough for the tropical re- gion, where the C cycle is faster. another important point has to do with the way zero tillage is conducted. In southern Brazil, soybean and wheat are the main crops used in summer and winter, respectively, while in the Cerrados the system is based on the soy-corn mini-harvest. Problems related to pests and diseases in zero tillage may appear in monocultures in summer as well as winter (derpsch, 1997), which is why crop rotation is essential for ensuring its success. Crop rotations under this system are also essential for reducing problems related to erosion, pests, and diseases, and to make the best use of accumu- lated organic matter in the soil. While zero tillage may increase soil C stocks, it can also provide conditions for great- er denitrification activity and increase n2o emissions (smith & Conen, 2004), although this does not seem to occur in all soils. Jantalia et al. (2008) did not find any differences in the N2o in soils under zero tillage compared to those under conventional planting. With well-draining soils, such as latossols, which are common in most agricultural ar- eas in the country, the use of zero tillage does not favor n2o emissions (rochette, 2008). In addition to favoring soil C accumulation, the use of zero tillage can reduce meth- ane emissions from wetland rice, a recommended mitigation strategy for the produc- tion system (Wassmann et al., 2000). studies done in wetland rice areas show different 123 levels of reduction in Ch4, amounting to an average of 48 percent (table 27). the effect of zero tillage may be explained by the increase in electron receptors (hanaki et al., 2002) and phototrophics (harada et al., 2005), which reduce methanogenous activity. research on wetland rice systems in Brazil shows that the absence of soil rotation in the zero tillage system can reduce methane emissions by about 15 percent (Lima, 2009). Table 27: Methane emissions from conventional planting and zero tillage in irrigated rice areas in different locations and a comparison of emissions reductions between the two uses Technical Synthesis Report | Land Use, Land-Use Change, and Forestry % Reduction in Local Conventional Zero methane References (time measured) planting tillage g Ch4 m-2 emissions Philippines (cycle) 27.2 19.6 28 Wassmann et al., 2000 Japan (2 years) 48.4 15.6 68 hanaki et al., 2002 Japan (2 years) 45.8 18.8 59 hanaki et al., 2002 China (year) 117.9 19.7 83 shao et al., 2005 China (year) 117.9 68.4 42 shao et al., 2005 Japan (year) 17.9 10.2 43 harada et al., 2005 China (cycle) 17.9 15.7 13 Xiang et al., 2006 Average 48 -- the zero tillage system is characterized by the elimination of soil disturbance, Benefits of zero tillage maintenance of soil continuously covered with crop residues, and the use of crop rota- tions. one of the great benefits of zero tillage is the decrease in soil erosion. In southern Brazil, which is more mountainous, loss of soil from erosion can be substantial. studies show that soil loss is reduced an average of 70 percent, and water loss an average of 20 percent (Figure 28) with zero tillage. Maintaining residue on the soil helps reduce the effects of wind erosion, although the extent of this type of erosion in the country is not clear. one of the changes caused by zero tillage is related to soil structure. the use of ma- chinery on the soil can cause compaction, principally in the more superficial layers, which can be serious depending on the crops used in the rotation. Species such as for- age gramineae from the Brachiaria and Panicum genus can remedy this situation due to their deep and abundant root system. Well-conducted zero tillage has positive effects on capping the soil temperature, im- proving the structure and capacity of water storage, and increasing nutrient retention sites in the layer occupied by plant roots (gassen & gassen, 1996). The reduction of agricultural operations for soil preparation, which took a month or more before one could plant, was another benefit of zero tillage. It made it possible to have two or three harvests per year, thus economizing on fuel and labor for operations and maintenance (Figure 28). 124 Figure 28: Percentage of reduction of soil and water losses from zero tillage (ZT) com- pared to conventional planting (CP) Technical Synthesis Report | Land Use, Land-Use Change, and Forestry table 28 shows costs for investments and o&M for agriculture in the reference (Adapted from De Maria 1999) scenario, as well as the accumulated income earned between 2010 and 2030. approxi- mately 68 percent of the revenue includes total costs for the reference scenario. Values for the Low-carbon Scenario appear in the same table, with a 100 percent adoption rate for zero tillage. total costs represent 44 percent of the income for the Low-carbon sce- nario. Investment costs are reduced 29 percent and o&M costs, 8 percent. Cost reduc- tions stem from the elimination of tools and materials, such as fences and plows, and the decrease in the use of fuel (about 40 l/ha for each harvest). Table 28: Cumulative costs and revenue in the reference and Low-carbon Scenarios with the adoption of zero tillage from 2010 to 2030 Proposal Income (without Cumulative Reference scenario Options Mitigation or Carbon Uptake Options Investment Cost o&M Income Investment Cost o&M considered carbon) Zero tillage 473,851,746.00 2,324,026,541.00 4,114,575,626.00 335,574,435.00 2,129,349,512.00 5,618,152,902.00 125 thus, the adoption of zero tillage for mitigating greenhouse gas emissions will not generate additional costs, as it will automatically lead to an increase in income. 3.1.1 Emissions Reduction Potential Associated with Zero Tillage In the Low-carbon Scenario, 100 percent of the cotton, rice, beans, corn and soybean production area will be converted into zero tillage starting in 2015. the principal result of this process is the reduction of Co2 emissions from land use, which is 237 Mt Co2eq (table 29). the reduction of n2o in the soil through the mineralization of organic n is 55 Mt Co2eq. Zero tillage in areas with irrigated rice contributes to reducing emissions by 10 percent compared to the reference scenario, or 43 Mt Co2eq. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry the amount of fossil energy saved is 21 Mt Co2eq, mostly from economizing on the use of diesel oil in agricultural operations. Total avoided emissions amount to about 356 Mt Co2eq, the equivalent of a 17 percent reduction compared to the reference sce- nario. Table 29: Greenhouse gases produced in the Low-carbon Scenario: adoption of zero till- age in 100 percent of the agricultural area from 2015 to 2030 difference compared to the low- ghg emissions (Mt carbon scenario emission source CO2e) % reduction Co2 produced with a reduc- Mt CO2e 348.4 236.8 40.5 tion in soil C stock N2o from fertilizers, residue (including burning of sugar cane) and mineralization of nitrogen in the soil with a 631.0 54.6 8.0 reduction in C stocks Ch4 produced from wetland irrigated rice and burning 390.8 42.8 9.9 of sugar cane Use of fossil energy to pow- er agricultural operations 322.4 21.1 6.3 Total 1,692.5 355.5 17.0 Results per unit of the federation are presented in Maps 22 and 23 below. Map 22: Mitigation by crop, 2010 to 2030 126 Cotton (Mt CO2e) Rice (Mt CO2e) Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Bean (Mt CO2e) Corn (Mt CO2e) Soy (Mt CO2e) Sugar cane (Mt CO2e) Positive Effect Negative Effect Source: EMBRAPA, Word Bank Brazil Low Carbon Case Study Map 23: Total emissions from agriculture, 2010 to 2030 127 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 3.1.2 Obstacles Limiting the Expansion of Zero Tillage the use of zero tillage requires three basic actions to ensure the system’s sustain- ability: a) continuous planting without traditional soil rotation; b) the use of crops that leave a sufficient amount of litter to cover the soil the entire year; and c) crop rotation in summer and winter to break cycles of pests and diseases and improve nutrient recy- cling in the soil. surveys on the use of zero tillage in Brazilian agriculture, with the support of rural extension agencies and farmers, show that the system was widely adopted up until the beginning of this decade. the following years, adoption estimates for zero tillage in production areas were limited to the analysis of trends, resulting in an estimate of 25 million hectares in 2005. Consultations with specialists from the Brazilian Federation of Zero tillage into Crop residues and the Zero-tillage Farmers association of the Cer- rado (aPdC) provided information on the stagnation of zero tillage adoption. Farmers seemed to be increasingly using conventional practices, such as light terracing, subso- lators, or even completely returning to conventional planting (dr. John Landers, aPdC, personal communication). however, there is some agreement among specialists that the area under a “well-executed� zero tillage system, based on the premises outlined, is actually well below the 25 million hectares that appear in the statistics. Changing from conventional planting to zero tillage has never been easy. Myths about the use of zero tillage, such as the risk of soil compaction and low efficiency lim- ing, and the occurrence of pests and diseases as a result of poorly planned systems or not following recommendations, discourage farmers from even trying. There are also still a number of technical obstacles that have yet to be surmounted. access to technology. Farmers do not generally adopt technologies that they are not familiar with. They also resist acquiring the necessary knowledge, which is a major obstacle for small-scale farmers, who are responsible for much of grain pro- duction such as beans and corn, and who, for economic and cultural reasons, have little to no access to professional support to help adapt their production systems. This is less of an obstacle for large-scale farmers. 128 Costs of conversion/economic advantage. depending on each situation, the be- ginning of the zero tillage system can be more onerous due to the need for machines and more inputs, and there is no consensus that the use of zero tillage is always eco- nomically advantageous for all parts of the country. Available knowledge. The degree of knowledge and technologies available for areas with more amenable climates in southern Brazil means that the use of zero tillage is much more widely disseminated there. Nevertheless, for regions including the north/northeast of the state of Paraná and other parts of the country, especially the Cerrados, more research has been requested on cover crops for the period fol- lowing the summer harvest, which ensures that enough residues are produced to Technical Synthesis Report | Land Use, Land-Use Change, and Forestry cover the soil during the year. Logistics and infrastructure. the Brazilian farmer generally suffers greatly from the handling and storage of farm products. The higher value of soybean means that there is no way to store crops such as corn, one of the most important options for crop rotation in the summer. There is also no guarantee that these alternative cere- als used in summer rotations will be purchased. These factors only serve to stimu- late the soybean monoculture and weaken one of the links for the success of zero tillage: diversification. 3.1.3 Proposals for Overcoming Obstacles the obstacles that hinder the expansion of zero tillage in the country need to be sur- mounted. The following measures are recommended to achieve this: Encourage basic research and technology in order to generate a continuous infor- mation flow to ensure the sustainability of zero tillage in different parts of the country. restructure the rural extension system by training technicians who act as a link be- tween research institutions, universities and the different actors in the productive sec- tor. It is essential that technical universities and schools include the zero tillage system in the minimum professional training curriculum. Facilitate differentiated priority credit for farmers who adopt the zero tillage sys- tem; e.g. increase agricultural credit with lower interest rates for farmers who practice zero tillage; offer rural insurance with the possibility of reduced premiums depending how long it takes to adopt the system, etc. Increase storage area and guarantee the purchase of suitable products for zero till- age, such as corn and rice. develop financial “hedge� instruments for prices for essential inputs for the zero tillage system (e.g. herbicides). 3.2 Carbon Uptake through the Increase of Planted Forests for as discussed in Chapter 2, Brazil’s main available options for carbon uptake are Renewable Charcoal planted forests and native forest recovery—particularly reforestation of riparian for- ests and legal reserves. the next two sections identify the carbon removal potential 129 of these options, first for production forests and second for native forest recovery, and analyze and explore ways to overcome the obstacles to their implementation. 3.2.1 Carbon Uptake Potential Associated with the Increase in Re- newable Charcoal Production one mitigation option considered here was the additional use of renewable charcoal in the Brazilian iron and steel production sector. this section deals with the potential of net greenhouse gas uptake or “sequestration potential� as a function of a possible increase in renewable charcoal production. the national Plan for Climate Change studies this type of mitigation option and ref- erence is made to the need to double the actual area of planted forests in Brazil (MMa, Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2008). however, since the Plan is still at an early stage, no exact amounts have been mentioned regarding the additional use of renewable charcoal in the iron and steel pro- duction sector. This topic has also been studied by the Productive Development Policy, which is still being developed and coordinated by the Ministry of development, Indus- try and Commerce (MdIC). As mentioned earlier, the additional use of renewable charcoal as a thermo-re- duction agent in the iron and steel production process may result in two types of cli- matic benefits: (i) emissions reductions in the industrial process and (ii) an increase of carbon stocks generated by additional stocks of sustainable forest plantations. a low-carbon scenario was developed within the framework of this study (LCCCs ). this report compares the reference scenario with the uptake potential of the new forest plantations in two situations. In the first (scenario 1), the hypothesis would be that the participation of charcoal in iron and steel production would be maintained at current levels, approximately 33 percent until the year 2030, assuming that all of the charcoal used by the sector would be from plantation forests. At the present time, less than half of the charcoal used in the sector comes from planted forests (see aMs, 2009, aBraF). In the second situation (scenario 2), participation would increase from the current 33 percent to approximately 46 percent by 2030. In both cases, charcoal would continue to be the main thermo-reduction agent used in iron and steel production, but in the second case, the relative participation of charcoal would increase about 13 per- centage points. however, since both scenarios are very ambitious and are dependent on different structural changes in current production conditions (to be discussed later on), it was decided to adopt them as hypotheses for projections to keep the demand for land in Brazil at a more conservative level, while incorporating a considerable expan- sion in areas of planted forest. But these are not the most likely scenarios in the absence of new measures. For the two situations, the following premises were used: A substantial decrease in investment, management, institutional and technological barriers, generating a significant increase in the supply of planted forests for renew- able charcoal through the different measures presented earlier in this report, including the use of the Clean development Mechanism of the Kyoto Protocol or similar instru- ments. With the decrease in the aforementioned barriers, the stock of planted forests would represent an area of 2,352 to 2,663 million hectares in scenario 1 and 3,276 to 3,663 million hectares in scenario 2. The main obstacles to using renewable charcoal for iron and steel production are: (i) scarcity of planted forests, (ii) higher transaction costs for renewable charcoal com- 130 pared to mineral coke and non-renewable charcoal and (iii) technical and logistical lim- itations of using renewable charcoal in large blast furnaces. given the aforementioned technical and operational limitations, it will be necessary to promote the injection of charcoal powder in large blast furnaces that run on coke (especially in the integrated sector) to stimulate new productive arrangements based on the use of smaller blast furnaces (especially in the independent sector). In conclusion, maintaining the current arrangements would make potential changes impracticable for different companies in the sector, especially in the integrated sector, which makes up the better part of Brazil- ian iron and steel production. As a result of the implementation of the new measures, such as those suggested here, and the gradual increase in the availability of planted forests starting in 2010, Technical Synthesis Report | Land Use, Land-Use Change, and Forestry there may be a measured change in the participation of each thermo-reduction agent in the growth projected for the sector starting in 2018 (3.7 percent per year ). In scenario 1, the annual growth of iron and steel production using coal will be about 2.8 percent, and the annual growth of production using renewable charcoal will be approximately 0.9 percent. In scenario 2, the reverse will occur, in which case the an- nual growth of iron and steel production using coal will be about 1 percent, and the an- nual growth of production using renewable charcoal will be about 2.7 percent. In other words, both Low-carbon scenarios were completely based on the sector’s expansion, including the consolidation of the different investment decisions. Possible co-benefits and negative effects: In addition to the climatic benefits, a variety of co-benefits may be attributed to the expansion of renewable charcoal for iron and steel production. one indirect but ex- tremely important co-benefit is its contribution to reducing pressure on native forests in Brazil. historically, native forests have met most of the demand for wood in the coun- try, which has contributed to the deforestation of native forests in different biomes. another co-benefit is the significant reduction in the country’s dependence on out- side energy sources due to its external dependence on coal. approximately 80 percent of the material is imported due to the scarcity as well as the properties of coal produced in the country (Brito, 1990). If the measures taken help avoid the use of coal in the fu- ture to some extent, there may also be a significant reduction in emissions from other polluting gases, such as so2/sox, as well as in the net consumption of atmospheric o2. In the process of manufacturing iron and steel using coal, “1,376 kg of o2 are consumed for each ton of pig iron produced. on the other hand, when charcoal is integrated (…) there is practically no o2 removal from the atmophere� (Bonezzi, Cadeira-Pirez and Brasil Junior, 2004; apud sampaio, 1999). according to Castro (2000), for iron and steel pro- duction using charcoal from planted forests “the productive cycle shows a negligible consumption of 8 kg o2/ ton of pig iron, with the permanent restoration of 14,120 kg o2/ ton of pig iron in the atmosphere�. on the other hand, to produce a ton of pig iron, it was found that “10 kg of sox pro- duced by the trajectory of coal� is also emitted (Castro, 2000) and about 9.5 so2 from the same process (Bonezzi, Cadeira-Pirez and Brasil Junior (2004), apud sampaio (1999)). these emissions occur basically due to the chemical composition of coal which contains sulfur and “other undesirable substances such as heavy metals, which are only partially removed from combustion emissions. The combination of these sub- 131 stances with water vapor in the atmosphere can form sulfuric acid precipitation� (Cas- tro, 2000). The effects of activities related to the production of planted forests can also be pre- sented as co-benefits if they occur on degraded or less productive lands, and if they comply with Brazilian environmental legislation. these possible gains include great potential to contribute to sustainable development in rural areas by generating new jobs, or the establishment, monitoring and preservation of areas of native biomes ad- jacent to the plantations, contributing to biodiversity conservation, compared to what might occur in degraded areas. Potential negative effects of the hypotheses of Low-carbon Scenarios are directly Technical Synthesis Report | Land Use, Land-Use Change, and Forestry linked to some of the basic premises adopted in this report, such as respect for environ- mental legislation, especially with regard to deforestation-related provisions. Negative environmental impacts may occur in cases where the law is not complied with, either in the socio-environmental management of forest plantations and carbonization practic- es on different scales, or in the unsustainable conversion of native forests in production areas. Potential negative effects would need to be studied within the framework of a more thorough analysis, including the implementation of the Brazilian environmental legislation as a whole, which is beyond the scope of this report. there is a definite need to examine these risks, although there are fewer now than in the past. It is also important to emphasize that a conclusive analysis should compare the substantive consequences of not complying with the law with the potential nega- tive effects of using alternative products, in other words, the possible negative effects of using coal and charcoal from unsustainable deforestation practices. If not, the analysis could either not reflect, or reflect in an unbalanced way, possible trade-offs regarding the use of the three thermo-reduction agents in question. Quantification of the potential net uptake of Co2 in the Low-carbon Scenarios: The hypothesis of the implementation of the Low-carbon Scenarios would result in the generation of net Co 2 uptake from the atmosphere commensu- rate with additional stocks of planted forests. Estimates of the uptake poten- tial were based on the average pluriannual stocking capacity of 190 tCo 2 e per hectare, as presented above and in Chapter 2. this is a conservative ap- proach in that average Co 2e stock per hectare only includes the average quan- tity of live biomass during the seven years of the wood planting and harvesting cycle . expected gains in productivity are also considered with possi- b l e o p e ra t i o n a l o r te c h n o l o g i c a l i m p rove m e n t s i n fo re s t p l a n t a t i o n s , and in the processes of carbonization and thermo-reduction over time, in a way that is coherent with the gains in productivity obtained in the past . Considering that the total area required may vary between 2,352 and 2,663 million hectares in scenario 1, the stocking potential in 2030 would be between 446.8 MtCo2e and 499.6 MtCo2e, as shown in Figure 29 and table 30. Figure 29: CO2e stock from forest plantations for renewable charcoal in Scenario 1 132 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Table 30: CO2e uptake from forest plantations for renewable charcoal in Scenario 1 Low-carbon scenario 01 2010 2015 2020 2025 2030 Uptake of tCO2e Reference Scenario 144,364 146,258 175,394 179,407 178,503 Low-carbon Low 159,973 170,484 264,639 376,007 446,875 Scenarios Medium 170,506 178,120 276,493 396,691 473,202 (by productiv- ity) high 181,642 189,335 293,902 419,567 499,681 Net uptake potential in the highest productivity scenario 37,278 43,077 118,508 240,160 321,178 (highest) in tCo2e total area required in scenario 2 would be between 3,276 and 3,663 million hect- ares and the stocking capacity would vary between 622.4 MtCo2e and 695.9 MtCo2e, in 2030, according to Figure 30 and table 31 below: Figure 30: CO2e stock in forest plantations for renewable charcoal in Scenario 2 133 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Table 31: CO2e uptake in forest plantations for renewable charcoal in Scenario 2 Low-carbon scenario 02 2010 2015 2020 2025 2030 Uptake of tCO2e Reference Scenario 144,364 6,258 175,394 179,407 178,503 Low-carbon Low 159,973 179,677 343,677 531,261 622,440 Scenarios Medium 170,506 187,726 359,071 560,484 59,110 (by productiv- ity) higher 181,642 199,545 381,680 592,806 695,992 Net uptake potential of the higher productivity scenar- 37,278 53,287 206,286 413,399 517,489 io (tCo2e) Net uptakes that were hypothetically generated by the two Low-carbon Scenarios in 2030 were estimated based on the difference between the stocks of planted forests in the respective scenarios and stocks in the Reference Scenario that are the equivalent of approximately 1 million hectares in 2030. thus, the maximum net uptake potential in 2030 would be approximately 321.1 M tCo2e in scenario 1 (see table 30) and 517.4 MtCo2e in scenario 2 (see table 31), as illustrated in Figure 31 below. Figure 31: Comparison of CO2e stock in Scenarios 1 and 2 and the Reference Scenario. 134 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 3.2.2 Obstacles to the Expansion of Production Forests for Renew- able Charcoal Brazil is one of the few countries capable of producing iron and steel using charcoal on a large scale27. however, the increase in the use of renewable charcoal from planted forests in an effort to avoid using mineral coke or non-renewable charcoal in the future is now faced with a number of obstacles and market flaws, which are listed below: Lack of adequate funding: although forest plantations in Brazil are highly produc- tive compared to other countries, they require substantial long-term investments (ex: land, labor, etc.). the first profits are usually seen only after the seventh year, meaning that the necessary loans for this activity require a grace period of at least seven years, and a minimum duration of 10 years for fast-growing species such as eucalyptus. This credit structure is non-existent in commercial Brazilian banks and rather rare in pub- lic ones. Most federal funding programs (ProPFLora, PronaF) are geared towards small-scale production, which, although important, is not enough to reverse the deficit of planted forests in Brazil. difficulty in obtaining credit: In addition to the scarcity of financing, access to credit is also a major obstacle. Many banks still have difficulty accepting planted forests as a loan guarantee, although it is allowed for other agricultural crops. often the land alone 27 Brazil, 2007 can be considered a guarantee. on the other hand, the irregular situation of some firms in terms of environmental licensing often makes it more difficult to grant funding, fur- ther proof of the need for coordination between public funding policies and economic agents within the framework of environmental licensing processes. according to aBraF (2009, p. 93), “the forest sector currently has lines of credit at its disposal for small-scale forestry projects that are implemented by public federal banks.� these funds are concentrated basically in two programs, ProPFLora and For- 135 est PronaF, developed by the national Bank for economic and social development (Bndes) and by the partnership between the environment Ministry (MMa) and the agricultural development Ministry (Mda), respectively. although substantial, such resources are not enough to meet the financial needs necessary for reversing the deficit of planted forests in the country. In 2007, ProPFLora spent approximately r$52 mil- lion, while PronaF spent about r$12 million the same year (aBraF, 2009: 94-95). the amount needed for financing within the framework of the Low-carbon scenario is ap- proximately Us$ 6 billion. another important program is the BB Florestal which, besides investing in forests, also covers cost defrayal and marketing. Despite its rather considerable resources, with nominally about r$76 million in operation in 2008, contracts are concentrated Technical Synthesis Report | Land Use, Land-Use Change, and Forestry in the state of são Paulo, with over 73 percent of the total amount of applied resources (aBraF, 2009) and are being used for other industrial sectors. some state-level experiences have met with relative success, such as the Proflores- tas Program of the Minas gerais state development Bank (BdMg), but suffer from the same scarcity of resources. There are still some resources available in the so-called Constitutional Funds (Fno-Basa; Fne-BnB; FCo-BB), but there is no program that specifically targets the use of charcoal from planted forests for iron and steel produc- tion. higher transaction costs than alternative products, and the capital market’s greater aversion to charcoal from planted forests: Transaction costs from planting and manag- ing production forests are significantly higher for renewable charcoal than for the main alternative products. Compared to the global substitute (mineral coke), renewable charcoal has going against it: a long maturation period, with a 14-21 year production cycle; greater need for labor for planting and carbonization processes; high costs for land and difficulty obtaining environmental licencing and financing. on the other hand, coal is a global commodity with an established international market, whose produc- tion costs and logistics structures are widely known. the product also benefits from increasing returns to scale and network externalities (see Krugman & obstfeld, 2000). Compared to non-renewable charcoal from deforestation, renewable charcoal may face dishonest and often illegal competition, giving rise to negative externalities. non- renewable charcoal does not require substantial investments for land and plantations, thereby drastically reducing its production costs. In this context, with its market shortfalls, international investors tend to have great- er risk aversion to renewable charcoal, a long-term investment with higher transaction costs than coal. Lack of security in the sustainable supply of renewable charcoal: Brazil has expe- rienced a historic deficit of wood from forests that have been planted for different purposes, especially for the supply of renewable charcoal28. From 1967 to 1988, the federal government had a fiscal incentive program (FIset), which, despite a number of problems, contributed to the significant increase of the area with planted forests (Bra- zil, 2007). With incentives coming to an abrupt end, the prevalence of the abovemen- tioned obstacles became even more pronounced, followed by an increase in the deficit of renewable charcoal on the market. on the other hand, the opening up of the Brazilian market in the early 90s facilitated access to coal (Brazil, 2007). 136 the country’s chronic renewable charcoal deficit, commonly known as “forest blackout�, makes companies in the sector highly vulnerable. this problem is to a great extent the result of the different obstacles listed in this section, as well as market flaws due to negative information, and the respective challenges in evaluating the risks of al- ternative products (coal, and non-renewable and renewable charcoal), as mentioned in the last section, exposing companies to periods of shortage in renewable charcoal. historic variations in the price of pig iron, which – strictly in the financial sense – could make the establishment of plantations more attractive, have not resulted in a proportionate increase in plantation establishment. on the contrary, the current gap in plantation establishment has actually grown. Throughout history, positive changes in the price of pig iron had no impact, doing nothing to reverse the deficit situation be- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry tween the annual establishment of plantations and the effective consumption of reduc- tion agents. rather, the deficit increased when the price of iron increased significantly (research by sIndIFer/aLICeWeB/aMs 2007, 2008). this rather inflexible relation- ship between the establishment of plantations and their final use corroborates the sizeable risks perceived in investing in plantations, which helps explain the predomi- nance of coal, an easily accessible and abundant global commodity. Prevalence of inefficient carbonization technologies: Most of the conversion of wood into charcoal in Brazil is still done using inefficient technologies that emit large amounts of ghgs, including Ch4. These are still common processes even if the wood comes from renewable sources. In cases of illegal wood carbonization practices using wood from deforestation, the environmental damage can be even greater, with clandes- tine situations making working conditions even worse. Social communication: There is a considerable lack of information and communica- tion between the economic players involved in the productive chain and civil society with regard to positive impacts and ways to mitigate possible negative ones associated with large-scale wood cultivation and charcoal production. This could result in prob- lems for public policy formulation and in the definition of regulations. Risks related to regulations: Despite the fact that the production logic behind planted forests (silviculture) is the same as for other crops, the regulations for the sector are 28 different governmental and non-governmental organizations have published reports on the status of plantations and the sources of wood supply, as well as on specific deficits of plantations for the purpose of producing charcoal in Brazil, including the Instituto Brasileiro de geografia e estatística (Brazilian Institute for geography and statistics - IBge), Banco de desenvolvimento social e econômico (national Bank for economic and social development), the environment Ministry (MMa), the Brazilian silvicultural society (sBs), the Instituto Brasileiro de Pesquisa Florestal (the Brazilian Institute for Forestry research - IPeF), the associação de silvicultura de Minas gerais (silvicultural association of Minas gerais - aMs, antiga aBraCaVe), the Universidade de Viçosa (UFV), Universidade de são Paulo (esaLQ/UsP), stCP engenharia (engineering), the associação de defesa do Meio ambiente de Minas gerais (association for environmental defense of Minas gerais -aMda), among others. different. one example is the need for a licence to harvest and transport wood from planted forests, which is not the case for the harvesting of other crops, thereby gener- ating additional limitations. overcoming this type of obstacle is not only linked to the simple need for less bureaucracy, but also to measures that ensure the wood’s origin control. environmental regulations and tree-planting laws in Brazil are generally complex and the environmental licencing process takes at least six months, despite recent at- 137 tempts to simplify it. Some characteristics of the regulations are inherent to the nature of the subject: land use in Brazil and the need to combine economic and socio-environ- mental development. Nevertheless, although these characteristics are seen as natural and unavoidable obstacles, they seem to carry more weight than the regulatory charac- teristics used for alternative inputs. Investors who opt to import coal, for example, are not subjected to this type of environment-related procedure, which impacts the analy- sis of opportunity costs. another example is the necessity of acquiring a large quantity of land to preserve legal reserve and permanent preservation areas, as explained ear- lier. Measures such as those mentioned above are necessary for guaranteeing the sus- tainability of the process, but they influence decisions on the use of the different ther- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry mo-reduction agents. Internalization of environmental costs for the use of charcoal is not necessarily accompanied by an equivalent marginal income to complement the environmental benefit generated. therefore, the evaluation of the opportunity costs for the use of renewable charcoal often suffers, characterizing the trade-offs between economic and environmental aspects. 3.2.3 Measures for Overcoming Obstacles Table 32 shows some suggestions for overcoming the aforementioned obstacles. Table 32: Measures proposed to surmount obstacles Identification of proposed Category Description revise current national financing instruments, includ- 138 measures ing the Bndes and the BB Florestal and Constitutional Funds, with the objective of facilitating and expanding (i) the availability of credit for the productive chain in the use of renewable charcoal in iron and steel produc- Rectifying tion, including forest plantations, carbonization tech- nologies, use of by-products and thermo-reduction. encourage public and private banks to recognize for- (ii) est assets as guarantees during the risk evaluation process for operations in the sector. Rectifying Technical Synthesis Report | Land Use, Land-Use Change, and Forestry encourage and support the use of the Clean develop- ment Mechanism of the Kyoto Protocol as an addition- (iii) al source of resources and financing, using methodolo- gies that cover most of the productive chain. Incremental revise the sector’s regulations in light of the nature of wood cultivation, aiming at synergy between federal and state entities, with the objective of improving the environmental licencing process, especially: simplifi- (iv) Rectifying cation of the licencing process for wood cultivation in areas that are degraded, underutilized or previously used for other crops; and simplification of harvesting and transport licencing, without harming environ- mental integrity. Strengthen the monitoring structure for the illegal use of non-renewable charcoal from illegal deforestation practices, based on (i) command and control mecha- nisms within the governmental framework and (ii) (v) measures that stimulate the use of products based on sustainable forest cultivation, and the depreciation Incremental of the value of products from deforestation practices, from buyers in the productive chain all the way to the final consumer. Include renewable charcoal (solid biofuel) and its de- riviatives (tar and biogas from carbonization) in the (vi) Brazilian biofuel policy and in the agendas of the re- spective agencies for its regulation, stimulation and Incremental promotion in Brazil and abroad. Develop an environmental communication and educa- tion program in a partnership between government and civil society, including the private sector, with the (vii) objective of informing the Brazilian population about alternatives such as the sustainable use of planted for- Incremental ests for iron and steel production. stimulate applied research on more efficient process- 139 es for converting wood into charcoal and for making (viii) the best use of by-products from the process (e.g. use of tar and combustion gases). Incremental Implications: “Stakeholders� - winners and losers In addition to climate-related environmental co-benefits, and the possible negative effects analyzed above, an analysis of the gains and losses for the different stakeholders involved in the Low-carbon Scenario is presented as follows: Companies from the iron and steel production chain: Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Charcoal producers: With the increase in demand, companies that produce re- newable charcoal from planted forests would benefit from the possible creation of a structured input market. this market is not very structured at present, and most firms that use non-renewable charcoal (from native forests) internalize production. on the other hand, non-renewable charcoal producers would incur considerable losses, as incentives for forest plantations combined with the strict legislation would practically eliminate activities based on deforestation and the use of non-renewable charcoal. In an efficient market economy, this flow of gains and losses would not result in net losses of taxes and jobs. on the contrary, those involved in unsustainable activities could be absorbed by new sustainable ones. however, the implementation of public policies that support this transition would be necessary to ensure the logistics of the process and reduce resistance from the political side (see energy report). Coal producers: given that the Low-carbon scenario contemplates a significant absolute growth in iron and steel production using coal, and that most of the coal used in iron and steel production in Brazil is imported, national producers would not suffer any significant negative impacts, as there is already room for a considerable increase in national production, subject to technical feasibility. Iron and steel industry: Iron and steel producers who use or can use charcoal would naturally also benefit from the new policies, as the implementation of the Low-carbon Scenario would reduce uncertainties and obstacles associated with the use of renew- able charcoal in iron and steel production, increasing supply security and decreasing investment risks. Those who would lose out in the long run are producers who deliber- ately use non-renewable charcoal on a large scale in order to avoid considerable invest- ments, generating unfair competition. As in the case of coal producers, coal-based iron and steel production would not be negatively impacted, as the fundamental premises of the Low-carbon Scenario do not involve any change in investments that are already consolidated or being implemented. Even in Low-carbon Scenarios, coal will continue to be the predominant thermo-reduc- tion agent due to an increase in production and the lack of techno-economic feasibility of its use in most cases. however, the main target, as well as the principal challenge of any hypothesis on the Low-carbon Scenario would be the elimination of the differ- ent obstacles while generating additional incentives so that future expansions, which have yet to be planned, may be based on renewable charcoal. Should this not occur, any hypotheses for Low-carbon Scenarios would be impracticable and the Reference Sce- 140 nario that appears in item 2.2.3 of this study would prevail. If current economic, tech- nological, logistical and institutional conditions are maintained, the potential increase in the relative participation of renewable sources of charcoal would not be viable and any restrictions on the use of coal could negatively affect the competitiveness of the na- tional iron and steel industry. Climatic benefits associated with renewable sources of charcoal production and the potential increase in the participation of this thermo-reduction agent in the market by 2030 could help bring about a balance in emissions resulting mostly from national production, which would continue to be based on coal. If emissions from the Brazilian iron and steel industry were analyzed as a whole, national iron and steel production using coal could actually benefit from the increase in the use of renewable Technical Synthesis Report | Land Use, Land-Use Change, and Forestry charcoal, as the sector’s carbon intensity could do an “about-face� and establish itself as one of the lowest, if not the lowest, in the world. If the net uptake potential were to be considered and attributed to the sector, the positive impact on the balance of emissions would be even more significant. therefore, charcoal from renewable sources could be a complementary and non-exclusive alternative to the use of coal, within the context of a partnership for the sustainable development of the national iron and steel sector. Nevertheless, despite the positive marginal costs, it is essential that the posi- tive climatic externality generated by the Low-carbon scenario be priced and convert- ed into marginal revenue so that the scenario does not generate economic losses for the different companies in the sector, further proof of the importance of the CdM, the carbon market, and other similar mechanisms in this context. other members of civil society: other likely winners in a new scenario with the policies proposed would be the rural populations in regions located in the immediate vicinity of iron and steel production hubs (within a 300-500 km radius). these populations would benefit from the increase in rural job opportunities related to the productive/ forest chain of renewable charcoal, such as solid biofuel. It would thus be of the utmost importance to ensure the imple- mentation of best working practices in regions where silviculture and charcoal produc- tion activities are expanding. In addition, municipalities where forest plantations and charcoal production are established could benefit from the increase in tax revenue due to the expansion of economic activities in rural areas. government: a major benefit for public management, which can be attributed to the implementa- tion of the Low-carbon Scenario, would be the increase in the rasterability, transpar- ency and monitoring capacity in relation to the fiscal and socio-environmental aspects of the productive chain in iron and steel production, especially on a small scale. This could result in the significant reduction of government inspection costs, especially with regard to measures that seek to control biomass origin. 141 3.3 Carbon Uptake through Native Forest Recovery as illustrated in Chapter 2, there is some potential for Co2 uptake through the natu- ral regrowth of degraded forests, which has already been mentioned in the Reference Scenario. But because of the botanical obstacles mentioned earlier, the carbon-capture potential associated with natural regrowth remains limited. Despite these challenges, various studies and projects have demonstrated that forest plantings can promote the accelerated reestablishment of native plant cover. Such plantings induce microclimatic changes favorable to germination, the establishment of plantlets and the generation of a layer of litter and humus, which increases soil fertility. In addition, shade from young trees helps to suppress invasive grasses. Because of the large areas of degraded ecosys- tems, such as abandoned pasture and croplands, where native forest recovery activities Technical Synthesis Report | Land Use, Land-Use Change, and Forestry could be implemented, such activities can represent significant carbon uptake poten- tial in Brazil. to assess Co2 uptake potential through native forest restoration, the study devel- oped a biomass potential model in the most promising biomes, the Cerrado and atlan- tic Forest. these biomes, which had large forested areas in former times, have suffered greatly from deforestation over the past two centuries. 3.3.1 Carbon Uptake Potential Resulting from a“Legal Scenario� for Forest Restoration To estimate the carbon uptake potential through forest restoration, the establish- ment of goals for these activities is required. As a result of consultations with govern- ment representatives, this study adopted compliance with forest law as a target with regard to forest preservation areas and reserves. The costs of implementation are ana- lyzed in Chapter 7. the greatest reforestation potential for carbon uptake in Brazil considered in this study revolves around a “Legal scenario� involving compliance with and enforcement of laws governing the management and use of riparian forests and legal reserves (Box 2). a two-step calculation is required to estimate that potential: (i) determining the area needed to comply with the legislation, and (ii) estimating the Co2 uptake potential result- ing from native forest restoration in this area. To estimate the amount of land needed for reforestation to comply with the Legal Reserve Law, the study used the area of the mu- nicipality as the basis for calculating the percentage of legal reserve. the study excluded conservation units, (CUs), indigenous lands, PPas of major watercourses, areas with declivity above 15 percent, unfit soils and urban areas. Legal reserve percentages defined by the Forest Code were used. also excluded were areas with native plant cover, including secondary vegetation, savannas, and forests. The area left equaled the intended area for forest recovery in compliance with the Legal Reserve Law. Box 2: Moving towards a “Legal scenario�: Main areas for Protection Permanent Preservation areas (aPP) are forested areas found on the banks of riv- Permanent Preservation areas ers, lakes and other water bodies, which preserve the water resources, prevent soil erosion, maintain the landscape and geological stability, ensuring the welfare of hu- man beings. In the case of riparian forests in Brazil, the width of aPP depends on the 142 width of the river (table a). Table a: Comparison of the width of the river and the aPP: Up to 10 30 Width of the river (m) Width of aPP (m) 10-50 50 50-200 100 200-600 200 Larger than 600 500 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Legal Reserves Legal reserves are areas in Brazilian rural properties (except aPP) that are vital to the sustainable use of natural resources, conservation and rehabilitation of ecologi- cal processes and biodiversity conservation. The percentage of land set aside as legal reserve varies by biome: • 80% in rural property located in the Legal amazon; • 35% in rural property located at the Legal amazon and Cerrado areas; • 20% in rural property located in forest areas or other forms of native vegeta- tion in other regions of the country, especially the atlantic Forest. To estimate the uptake potential, the study team assumed that the legal reserve ar- eas to be restored would be reforested gradually until 2030, when full legality would be achieved. Starting in 2010, 1/21 of the total area for reforestation would be deducted every year from the area available for agricultural production. The environmental li- ability for the country was estimated at about 44 million ha, about one-third of which would be located in the amazon region (table 33). Table 33: Area needed for reforestation under Brazil’s Legal Reserve Law, by state area for re- area for refor- state forestation state estation (ha) Mato grosso do sul 3,398,792 Acre 721,161 (ha) Mato grosso 9,465,888 amazon 34,848 143 goiás 2,611,730 Roraima 46,757 distrito Federal 0 Pará 11,369,199 Maranhão 40,959 amapá 0 Piauí 0 Tocantins 1,644,537 rio grande do norte 3,062 Paraná 1,711,257 Paraíba 27,167 santa Catarina 398,679 Pernambuco 58,239 rio grande do sul 1,184,241 Alagoas 91,861 Minas gerais 2,682,095 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Sergipe 118,800 Espirito Santo 205,436 Bahia 242,079 rio de Janeiro 178,087 Rondônia 4,794,589 são Paulo 3,314,927 total for Brazil: 44,344,390 ha Sources: ICONE, UFMG. the study estimated the carbon uptake potential for the Legal scenario at about 2.9 gt Co2 over the study period; that is, about 140 Mt Co2e per year (Figure 32).29 Figure 32: Carbon uptake potential of forest recovery activities and production forests 29 If the carbon uptake from the natural regrowth of degraded forests were to be included, then the potential uptake would increase by an average of 112Mt Co2 per year. In the meantime, since biomass accumulates over an extended period of time in recently planted native forests, often over 100 years before peaking, the forest restora- tion of the legal reserve is a measure whose uptake potential is beyond the scope of this study. Map 24 shows the uptake potential obtained for 2030 and total potential, per state. 144 Map 24: CO2 uptake potential through forest restoration by 2030 and total CO2 uptake potential Technical Synthesis Report | Land Use, Land-Use Change, and Forestry It is important to note that enforcing forest legal reserves implies releasing the cor- responding land currently occupied by other activities (i.e., crops or pastures). this means that the land use and land-use change projected in the Reference Scenario (Chapter 2) would need to be revised. such a revision would be significant since the area released for legal enforcement of the forestry law would equal more than twice the estimated deforested area under the Reference Scenario. This runs the risk that the benefits gained from carbon uptake resulting from forestry activities could be partially lost via increased conversion of native vegetation to accommodate crops and pastures displaced by restored legal reserves. 3.3.2 Obstacles to Forest Restoration and Ways to Overcome Them obstacles to forest restoration can be divided into two main categories, ecological and economic, and are described as follows. ecological obstacle: the main ecological factor of natural recomposition is that the predominant species should regenerate naturally, without the need for long-term planting. however, depending on the degree of degradation of the ecosystem, and if re- generation does not occur, the colonization of the area by arboreal species and the sec- ondary succession would be negatively impacted. The following obstacles are consid- ered limiting factors for natural regeneration in areas such as pastures and abandoned agricultural fields: • Absence of propagulum: lack or little availability of an adequate seed bank on the ground, absence of dispersers and the seeds’ difficulty in reaching the soil due to the quantity of biomass from gramineae. • Lack of plant establishment: in this scenario, even if there is an adequate seed bank, seed predation and the herbivorous consumption of young plants, in ad- dition to competition with gramineae, makes the natural reestablishment of the plant cover difficult. • other factors: in addition to the aforementioned causes, burning, over-exploi- 145 tation of the areas and the absence of symbiotes and pollinators are also consid- ered major obstacles to natural regeneration. Although these factors impede the natural regeneration of native forests, a number of scientific studies have shown that forest plantations can eliminate these barriers, facilitating and accelerating the reestablishment of native plant cover. The positive effect of plantations is in the micro-climatic changes they generate, favoring plant ger- mination and establishment, and creating a layer of litter and humus that increase soil fertility. In addition, the shade of the young trees helps suppress the growth of invasive gramineae. economic obstacle: Currently observed in the state of são Paulo, which has a deficit Technical Synthesis Report | Land Use, Land-Use Change, and Forestry of over a million hectares of riparian forests. Despite the fact that the state government created reforestation programs, for example the secondary Forest restoration Project, with funding from geF, and that the Federal government has made lines of credit avail- able for the native forest restoration on rural properties, less than a thousand hectares in total have been registered for the native vegetation restoration by rural property owners. As mentioned earlier, the high costs of restoration, combined with the loss of productive area, are the main economic obstacles to the implementation of large-scale forest restoration activities. The voluntary participation of property owners in reforestation programs that fo- cus on the restoration of these areas is rare. Forest restoration generally occurs in most cases due to legal obligations that entail changes in behavior and practices, and forest compensation. however, the following measures can provide major resources to help develop forest restoration projects: Establishment of volunteer market for emissions compensation: The volunteer market for corporative greenhouse gas emissions compensation has great potential to contribute to the restoration of forest areas that are part of the Low-carbon Scenario (legality). this market is currently not regulated, and is without clear emissions fac- tors in many inventories. Moreover, there is no standard for carbon uptake potential through planted trees for compensation, as this type of compensation is customar- ily awarded through actual tree planting. also, the organizations and companies that award the compensation have no idea where the trees are planted, so the creation of legal mechanisms that restrict tree planting for ghg-related compensation to perma- nent preservation areas would be essential for implementing the long-term legal sce- nario. simplification of the environmental licencing of forest restoration activities: the requirements of forest activities that aim at restoring degraded areas through the use of native species must be simplified with regard to the environmental licencing of this activity, eliminating bureaucracy and facilitating the implementation of this type of project. derogation of the Forest servitude regime: the Forest servitude regime estab- lished by the Forest Code permits the compensation of the environmental liability of a specific property with another property located in the same micro-basin if it has a forested area that exceeds the limits established by the Forest Code. however, with the atlantic Forest Law, which prohibits the deforestation of any area that is in an interme- 146 diate or advanced stage of regeneration, this regime takes on a different character, as the owner registers forest excess on his property to comply with the obligation of the other property owner, who in turn does not reforest within his property. stimulate the CdM forest program modality: the CdM forest program modality is considered major progress within the CdM in the implementation of large-scale forest restoration programs through different activities. however, due to the perceived risks and long validation and approval process, it has not developed to the extent it should have, by combining new restoration areas. These areas should be added to the Program of Activities during project implementation, without going through the validation and approval process to which a traditional CdM project is subjected. thus, government support to the CdM forest program would be a public policy that could generate ad- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry ditional resources for native forest restoration programs, and is also essential for over- coming resistance to this modality. Foster the market for non-wood forest products: the harvesting and marketing of native fruits, plant resins, honey, and other non-wood forest products considerably increases the value of native forests and should stimulate the restoration of legally pro- tected forest areas, generating an income for local communities. These efforts should be encouraged. Fiscal incentives: rural property owners whose properties do not present deforestation in the Permanent Preservation area (PPa) and the Legal reserve (Lr) receive fiscal incentives that are as close as possible to the opportunity costs of the land. these incentives may be in the form of a tax exempt status or even the availability of lines of credit at reduced interest rates and with longer grace periods. This measure would motivate rural property owners with vegetation coverage liabilities to eliminate them by reestablishing natural plant cover on their properties. Environmental education: Environmental education and awareness-raising pro- grams for rural populations and land owners, teaching them about the importance of forests for the environment and for the entire productive chain in the agriculture and livestock sectors. In addition, awareness-raising campaigns should be conducted with appropriate forest restoration models through educational programs, while building capacity among the rural population and land owners regarding the importance of for- ests in our society. 3.3.3 Reforestation Support Policies Within the framework of reducing and eliminating carbon emissions in the Cer- rado and atlantic Forest biomes, the principal legal instruments that provide any kind of guarantee for the maintenance of these biomes and the preservation of the carbon stock are the Forest Code and the atlantic Forest Law. the Forest Code the Forest Code establishes two distinct types of areas within the property that can- not be deforested: the Permanent Preservation area (PPa) and the Legal reserve (Lr). the PPas are defined based on geographic aspects, such as riparian forests, which are 147 30-3,000 meter wide areas of forested land that are adjacent to a body of water; and forests located on top of hills and slopes with declivity above 45 degrees. the Legal re- serve (Lr) is a part of the property, excluding the PPas, that cannot be deforested. the limits of the Legal reserve are defined depending on the biome such as: • 80 percent of rural property located in the Legal amazon; • 35 percent of rural property located in the Cerrado area in the Legal amazon; • 20 percent of rural property in forested areas, or other forms of native vegeta- tion in different parts of the country, principally the atlantic Forest. Forest servitude regime Technical Synthesis Report | Land Use, Land-Use Change, and Forestry the Forest Code has a flexibility mechanism that is frequently used in the regular- ization of the Legal reserve: the Forest servitude regime. through this mechanism, the rural land owner who does not have the minimum property required by the Legal Re- serve legislation can compensate for this deficit through the Forest servitude regime. this means that the property owner can transfer a forest area that is equal to his deficit from another property that has an excess of vegetation, as long as it is located in the same micro-basin. the atlantic Forest Law the atlantic Forest Law goes beyond the Forest Code, establishing limits for the de- forestation of vegetation that exceeds the Legal reserve. In practice, it imposes limits on deforestation for areas of vegetation at different stages of regeneration including: • advanced regeneration: may be deforested only in exceptional cases, when nec- essary for the implementation of projects that are of public interest, scientific research and preservation practices; • Mid-level regeneration: in addition to what is stated above, when necessary for the implementation of activities and crop-livestock-related uses for the small- holder farmer and traditional communities, that are essential for their subsis- tence; • Initial regeneration: in states where the total remainder of atlantic Forest is less than 5 percent, this stage of regeneration will comply with the same legal regime for vegetation at mid-level regeneration. It should be emphasized here that, at any stage of regeneration, the cutting and re- moval of vegetation should be authorized beforehand by the relevant state agency. national Forest Program - nFP Launched by the Federal government in 2000, the general objective of the national Forest Program – nFP is “the promotion of sustainable development, reconciling utili- zation with the protection of ecosystems and making the forest policy compatible with other sectors in such a way as to promote the increase of internal and external markets and the institutional development of the sector�. thus, the nFP combines the environmental, social, and economic aspects of the Bra- zilian forest sector, including, among its specific objectives: 148 • to stimulate the sustainable use of native and planted forests; • to encourage reforestation activities, namely on small rural properties; • to restore permanent preservation forests on legal reserves and in modified areas; • To support economic and social initiatives of populations that live in forest ar- eas; • to penalize illegal deforestation and the predatory extraction of forest products and sub-products, containment of accidental burning and forest fire prevention; Technical Synthesis Report | Land Use, Land-Use Change, and Forestry • To promote the sustainable use of national, state, district or municipal produc- tion forests; • to support the development of forest-based industries; • to expand internal and external markets for forest products and sub-products; • To promote social, environmental and economic aspects of the services and benefits provided by public and private forests; • To promote the protection of forest biodiversity and ecosystems. the nFP makes lines of credit available through the funds listed below. Principal lines of credit PronaF – national Program for the strengthening of Family agriculture the purpose of the national Program for the strengthening of Family agriculture (PronaF) is to financially support agriculture and livestock and non-livestock-related activities through the direct employment of the labor force that includes the rural farm- er and his family. It has a specific modality for the forest sector, the PronaF-FLoresta, which deals with the funding of projects involving silviculture, agricultural systems and sustainable extractivist exploitation for rural farmers. • the following are eligible for credit in PronaF-FLoresta: • Family farmers and rural laborers who: • Cultivate a plot of land as owner, homesteader, lessor, partner or concessionaire of the national agrarian reform Program; • reside on or near the property; • do not have title to an area greater than four fiscal modules, quantified accord- ing to current legislation; • earn at least 80 percent of the family income from crop-livestock and non-crop- livestock operations; • Perform and run overall operations, outsourcing only when necessary based on seasonal demands for agricultural and livestock operations; • earn gross family income between r$1,500.00 and r$10,000.00, excluding re- 149 muneration from social welfare benefits as a result of rural activities. Family farmers and rural laborers who: • Cultivate a plot of land as owner, homesteader, lessor, partner or concessionaire of the national Land reform Program; • reside on or near the property; • do not have title to an area larger than four fiscal modules, quantified according to the current legislation; • earn a minimum of 80 percent of the family income from the crop-livestock and Technical Synthesis Report | Land Use, Land-Use Change, and Forestry non-crop-livestock operations of the establishment; • Perform and run overall operations; maintaining 2 permanent employees, out- sourcing whenever the seasonal nature of the activity requires; • earn gross family income over r$10,000.00 annually and as much as r$30,000.00 excluding remuneration linked to social welfare benefits stem- ming from rural activities. PronaF-FLoresta has an interest rate of 4 percent per year and a 12-year time frame for repayment, with an 8-year maximum grace period. Funding limits are r$6,000 for the beneficiaries of group (a) and r$4,000 for those of group (B). • ProPFLora – Commercial Planting and Forest restoration Program • the credit line made available by Bndes, ProPFLora has the following general objectives: • Planting and maintenance of forests for industrial use; • Recomposition and maintenance of areas of permanent preservation and legal forest reserves; • Implementation and maintenance of forest species for the production of wood to be burned for use in the process of drying agricultural products; • Implementation of silvopastoral (combination of livestock with forests) and agroforestry (combination of agriculture with forests) projects; and • Planting and maintenance of palm forests for biofuel production. Beneficiaries include rural farmers (physical or legal entities), associations and cooperatives. the ceiling for beneficiaries is r$200,000 per year, with an interest rate of 6.75 percent per year. the time frame for repayment and grace periods adhere to the following regulations: • Up to 144 months, from the grace period until the date of the first cut, can be in- creased from 6 months to 96 months, for forest planting and maintenance proj- ects to be used for industrial purposes and in the production of wood to be used for burning for the process of drying agricultural products; • Up to 144 months, including a 12-month grace period for projects for the resto- ration and maintenance of permanent preservation areas and legal reserves; 150 • Up to 48 months, including an 18-month grace period for other projects for es- tablishing forest seedling nurseries. Constitutional Finance Funds the 1988 Federal Constitution earmarked 3 percent from income taxes and any type of profit, as well as industrialized products, to be applied to financing programs for the productive sectors of the northern, northeastern and central-west regions. By us- ing part of the income tax for the more needy regions, the Union prompted the creation of the Constitutional Fund for Financing the north (Fundos Constitucionais de Finan- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry ciamento do norte - Fno), northeast (nordeste - Fne) and Central-West (Centro-oeste - FCo), with the objective of promoting the economic and social development of those regions through programs that fund productive sectors. rural farmers, individual firms, legal entities and associations and production coop- eratives that develop activities in the crop-livestock, mineral, industry, agro-industry, tourism, infrastructure, commerce and service sectors may solicit financing from the Fno through the Banco da amazônia s.a. for the northern region; the Fne through the Banco do nordeste do Brasil in the northeast; and from the FCo through the Banco do Brasil s.a, in the central-west. Credit lines made available through each of these forest- sector-related funds are summarized as follows: FCo – Constitutional Fund for Financing the Central-West FCo PronatUreZa: this credit line made available by the Bank of Brazil is part of the Constitutional Fund of the Central West, where a good part of this region is within the limits of the Cerrado biome. Its potential beneficiaries are rural farmers (physical or legal entities), associations and cooperatives. this line of credit has the following general objectives: • sustainable forest management; • reforestation for energy and wood; • agroforestry systems; • restoration of degraded areas; • acquisition of machines and equipment; • Integrated rural and industrial projects; • Promotion of the market. Fne – Constitutional Fund for Financing the northeast Made available by the Bank of the northeast, this credit line has a line item, the Fne Verde, to be used to finance productive activities, with an emphasis on environmental conservation and protection in productive activities in general including organic ag- 151 riculture and livestock operations, such as the conversion of traditional systems into organic ones, forest management, reforestation, agro-silvicultural and agroforestry systems, alternative energy generation, collection and recycling systems for solid resi- dues, environmental studies, implementation of environmental management systems and certification, clean technologies and restoration of degraded areas. Fno – Constitutional Fund for Financing the north the actions of the Fno cover the states of acre, amapá, amazonas, Pará, rondônia, Roraima and Tocantins. This fund offers credit at interest rates that vary depending on the borrower, from 8.75 to 14 percent per year. operations related to the industrial, agro-industrial, tourism, infrastructure, commercial and service sectors include sus- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry tainable forest management activities, which correspond to line item Fno Floresta. Demonstrative Project Sub-Program - PDA Implemented by the environment Ministry starting in 1996 within the framework of the Pilot Program to Conserve the Brazilian rainforest (PPg7), the Pda supports innovative initiatives of civil society organizations in the sustainable use and preserva- tion of natural resources in the amazon and atlantic Forest biomes, aiming at improv- ing the quality of life of the populations involved. the program has supported initiatives in amazonia and the atlantic Forest, and in their associated ecosystems. Between 1996 and 2003, the Pda supported 194 projects, with 147 in the amazon and 47 in the atlantic Forest. the projects developed actions in areas including agroforestry systems and environmental restoration (including nurs- ery construction), forest and water resource management and environmental preser- vation. The PDA currently supports projects that are divided into three components: 1. alternatives to deforestation and Burning Project (PadeQ): with 49 projects contracted in the states of Pará, Mato grosso, rondônia, roraima and tocantins; 2. Consolidation: aims at strengthening experiences previously supported by the PDA through the more integrated consolidation of environmental, economic, social and institutional sustainability, and currently supports 31 large projects, with 12 in the atlantic Forest and 19 in the amazon; 3. actions to Conserve the atlantic Forest: 99 approved large and small-scale proj- ects throughout almost all the states where this biome is present. 3.4 Mitigation Options for Livestock Activities over the past few years, research on mitigating greenhouse gas emissions caused by ruminants has been stepped up and different technological options have been ex- plored, particularly with regard to manipulating the animals’ diets. however, since Brazilian livestock is raised predominantly on pastures, many of the mitigation options tested in developed countries are not applicable on a large scale. 152 due to the magnitude of degraded pasture areas in Brazil, their restoration has taken on a key role. The reduction in the productivity and quality of fodder and carbon stocks, combined with the low level of animal productivity, result in a higher level of emissions per product unit in this system. In addition, this type of system is extremely demanding with regard to the need for land, and its maintenance or expansion is asso- ciated with the need to open up new areas with native vegetation. More intensive systems, such as integrated crop-livestock and feedlots also have great potential to mitígate emissions and increase animal productivity, while reducing emissions per product unit as well as the demand for land. this alternative will be prioritized due to the large amount of emissions caused by Technical Synthesis Report | Land Use, Land-Use Change, and Forestry deforestation in Brazil. the potential to reduce emissions if these more technological systems are associated with research programs and incentives for the genetic improve- ment of both fodder species and animals will also be taken into consideration. A techni- cal summary on the main alternatives that are applicable to Brazilian production sys- tems is presented in the following section. 3.4.1 Main Options Considered for Mitigating Emissions from Live- stock Pasture restoration It is estimated that about 60 percent of pastures in the Cerrado region are currently at some stage of degradation. Pasture degradation results in a decrease in the soil car- bon stock, a reduction in carrying capacity, an increase in soil loss due to erosion, and a significant increase in Co2-e emissions per kg of meat. however, the restoration of these areas may reverse these characteristics, resulting in an increase in carbon sequestra- tion from the soil as well as carrying capacity, a decrease in soil loss from erosion, and a reduction in Co2-e emissions per kg of meat. Adoption of integrated crop-livestock systems Crop-livestock integration consists of the implementation of different productive systems with grains, fibers, meat, dairy and fuel in the same area, in association, either sequentially or in rotation. on-farm land use is alternated in time and space between agriculture and livestock (Vilela, 2008). The integration of meat and grain production systems is one of the viable options for the development of alternatives for re-establishing the productive capacity of culti- vated pastures. according to the author, out of all the benefits of the potential synergy between pastures and annual crops, the following stand out: a) improvement of the physical, chemical and biological properties of the soil; b) a break in the cycle of disease, pests and damaged plants; c) a reduction of economic risks through the diversification of activities, and d) a reduction in the cost of restoration/renovation of degraded pas- tures. The increase in animal productivity by maintaining pasture areas in better condi- tion makes it possible to improve zootechnical indices, while reducing greenhouse gas emissions per product unit. according to Kichel et al. (2003), integrated crop-livestock systems can attain car- 153 rying capacity rates of 3 animals/ha when combined with supplementing and feedlots, whereas the national average is approximately 1 head/ha. In a recent report, Martha Junior et al. (2006) concluded that the amount gained in terms of live weight on the pasture the first year in integrated farming-livestock systems was between 9 and 40 arrobas of carcass/ha/year, depending on the edaphoclimatic conditions and local management, while extensive systems typically produce between 3 and 5 arrobas of carcass/ha/year. In systems of stocking and finishing males only on pasture, Magnabosco et al. (2003) obtained carrying capacity rates of 2.68 animal units/ha and 1.48 units during the dry season, with an average weight gain of approximately 6@/year, enough to make the Technical Synthesis Report | Land Use, Land-Use Change, and Forestry slaughter of pasture animals viable in less than 30 months. expanding the adoption of finishing in feedlots the use of this technology for terminating animals in a short period of time (60 to 120 days) before slaughter has a number of advantages, among which are their lower age at time of slaughter, an increase in carcass weight and meat quality, improvement in the herd’s rate of reproduction, and greater land-use efficiency. typically, the animals are fed a moderate diet or canned concentrated grains, together with more voluminous feed (ex. silage corn or sorghum, chopped cane). thus, an animal that gains 1.6 kg/day in a feedlot produces in only 90 days what an animal that gains 0.4 kg/day in a pasture would produce in a year. however, in most regions of Brazil, extensive livestock is sub- ject to seasonal variations due to differences in temperature and precipitation. During periods of fodder scarcity, the animals’ performance is low or even negative, which needs to be recuperated during the next rainy/hot season. thus, using feedlots in the final phase of termination can reduce the age at time of slaughter by another year, with a significant reduction in Ch4 production from enteric fermentation. Improvement of fodder quality according to the Fao (2007), the potential to mitigate greenhouse gas emissions through the genetic improvement of fodder species has been relatively unexplored. however, according to the report, there are strong indications that this approach could, a priori, meet with a relatively high degree of success. Two aspects are under discussion with regard to the genetic improvement of tropical fodder species for the reduction of methane emissions. the first has to do with nutritional quality. An increase in digestibility and in the amounts of soluble carbohy- drates, as well as its voluntary consumption by the animal, can generate a significant reduction in the amount of methane produced per product unit. For example, Lovett et al. (2004) demonstrate that increasing the soluble carbohydrate content in forage grass by 33 g/kg generated a 9 percent reduction in methane production in vitro. It is estimated that there may be a reduction of about 15-28 percent in emissions from the use of improved fodder species (Fao, 2007). The second aspect has to do with the occurence of antimethanogenic compounds in fodder plants. Johnson & Johnson (2002) cite different compounds that appear to have such an effect. Woodward et al. (2001) determined a reduction in methane emissions for sheep and dairy cows consuming fodder that was rich in condensed tannins. Ulyatt et al. (2002) found that methane emission was substantially reduced for sheep and 154 dairy cows that consumed kikuyu grass (Pennisetum clandestinum), under specific conditions, suggesting the presence of compounds that have not been identified thus far. genetic improvement of the beef cattle herd Zootechnical indexes (e.g. birth and weaning rates, weight gain, age at first pregnan- cy and at time of slaughter, etc.) obtained in Brazil are still lower than those of devel- oped countries for different reasons, such as the quality of tropical fodder, the climate, the presence of diseases and parasites, and the herd’s genetic potential. the latter as- pect is due to the fact that the taurine cattle used in countries with temperate climates have undergone genetic selection over centuries. Their productive potential is there- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry fore high. In comparison, zebu cattle used in Brazil and in other tropical countries have only been utilized for 130 years in the americas. genetic selection did not take place in India, their country of origin, until 20 years ago, nor in Brazil. today there are different genetic evaluation programs in Brazil, including the Zebu Cattle Improvement Program (aBCZ/eMBraPa), the nelore Program of Brazil (national association of Breeders and researchers) and the PaInt (Lagoa da serra), among others. although these programs are all different, each one studies aspects such as weight, sexual precociousness, and maternal ability. none of these programs includes the nutritional efficiency aspect, although it is of significant economic interest. Moreover, more efficient animals consume less feed and produce less methane with the same performance (nkrumah et al., 2006). It is estimated that for most characteristics of economic relevance, genetic progress could reach 1 percent per year, although in Brazil progress is generally about 0.3 percent per year (Lobo et al., 2009). genetic improvement has not achieved its potential for several reasons: low rate of technology adoption, lack of availability of improved animals, non- utilization of the most advanced selection techniques, prioritization of other character- istics (eg. fur, breed, etc). this means that the Brazilian herd is much larger than neces- sary for satisfying the demand for meat, which implies that the production of Co2-e/kg of meat is way above that of competing countries. For example, Brazil and the United states are the biggest meat producers in the world, with 9.47 and 11.98 million tons in 2007, respectively, although the beef cattle herd in the Usa is half the size of the Brazil- ian herd (nass, 2010). Part of this difference is related to production systems, which are extensive in Brazil and extensive/intensive in the Usa, with most of the animals being kept in feedlots. another reason for this difference is the Brazilian herd’s lower genetic potential. an alternative for reducing Co2-e emissions while increasing meat produc- tion would be to improve the genetic quality of the national herd. 3.4.2 Obstacles and Proposals for Overcoming Them Adoption of more productive systems the first obstacle for pasture restoration and the adoption of more intensive sys- tems is the need for initial investment capital for the transition to the productive sys- 155 tem. since the activity is not very economically attractive (table 34), the availability of credit is essential, particularly for financing the purchase of animals to increase car- rying capacity. In case financing does not include the purchase of animals, the breeder will most likely underutilize available fodder resources due to the lack of capital for purchasing animals. there is also the activity’s low rate of return, which would require low interest rates to make it more attractive. The rather favorable economic performance of crop- livestock integration justifies the actions the Brazilian government has taken over the last five years (ProLaPeC, ProdUsa) to promote the adoption of these systems, to re- duce business risks and increase rural incomes, and to restore degraded pasture areas, thereby enabling the expansion of crop-livestock systems in already anthropized areas. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Incentive policies for the early slaughter of animals can also generate productivity gains and emissions reductions. a good example of this is the early Bullock Program in Mato grosso do sul. In this program, animals slaughtered in meat processing plants with carcass and teeth typification (as a way of evaluating the animal’s age), and the minimum desired carcass weight, provide the registered cattle breeder with a finan- cial incentive through the reduction of the ICMs payment from 16.67 percent to 66.67 percent. The adoption of a similar policy at the national level could be an incentive for the adoption of more intensive systems, not only in the stocking and finishing phases, but also in the cow-calf phase, as the better remuneration for animals for slaughter will be reflected in improved remuneration for the calves, enabling better intensification of the cow-calf phase. In addition, positive externalities such as a reduction in clandestine slaughter, better carcass conformity and more tender meat are expected for this type of policy. Another aspect to be considered in the adoption of more intensive systems is a greater need for effective management. Public policies that promote rural extension and training for cattle breeders are important for surmounting this obstacle. genetically improved fodder species since all the scenarios evaluated in the present report examine the extensive use of pasture, and cattle-farming in Brazil currently occurs predominantly in this system, the use of fodder species with less potential for methane production for ruminants would have a significant impact on methane gas emissions in the atmosphere. however, the priority of fodder species improvement programs in Brazil currently is to develop ma- terial with favorable agronomic characteristics and resistance to pests and diseases; they do not include the evaluation of emissions levels. on the other hand, ongoing research is testing evaluation techniques for methane production in vitro for fodder plants. It is currently estimated that a research program on genetic improvement for the launching of a cultivar would cost approximately r$4 million over a 12-year period. Within the context of genetic improvement with a focus on management and quality, public policies that promote the financing of projects in this area, which has not been a priority in crop-livestock research thus far, would steer efforts towards research uni- versities and institutions in order to select fodder species that are of higher nutritive value, as well as improved management strategies for its use, resulting in the launching of cultivars with better methane emissions potential for ruminants. Use of genetically superior bulls 156 genetic improvement, which has a longer period of return, is often not considered a priority by breeders in extensive systems. thus, programs that provide incentives for evaluating bulls, and subsidies for acquiring tested animals of good lineage, may contribute to greater sector efficiency in the medium term and also help reduce ghg emissions. assuming that 2.3 million bulls are needed to maintain the national herd (a bull-to-cow ratio of 30:1), a 50-percent premium for improved animals above their slaughter value, and four years of useful life for the bull, the total value of subsidies for the national herd would amount to r$350 million per year. Positive externalities for adopting such a measure include increased productivity, better quality carcasses and an increased calving rate (assuming andrological testing of improved bulls). Technical Synthesis Report | Land Use, Land-Use Change, and Forestry The mitigation potential for direct emissions from livestock, combined with the mitigation options proposed depends on the scenario adopted to substitute low-pro- ductivity production systems with higher productivity production systems: the greater this substitution, the greater the reduction of direct emissions from livestock. Since more productive systems enable the same level of production to be achieved on smaller pasture areas, this scenario was constructed based on the need to free up pasture areas as part of the strategy for reducing deforestation, which will be presented in the next section. As a result, calculating the potential for mitigating emissions resulting directly from livestock can only be done once this deforestation reduction strategy has been quantified at the end of the next section, after determining the amount of meat produc- tion allocated for each productive system in the Low-carbon Scenario. 3.5 Reduction of Emissions from Deforestation Deforestation appears to be the main source of emissions in the Reference Scenario. While significant, the mitigation and carbon uptake potential described in the forego- ing section remains limited compared to the large volume of ghg emissions resulting from deforestation. As mentioned above, a main trigger of deforestation is the need to convert native vegetation into land to accommodate crops and pasture expansion. the land-use modeling developed by this study makes it possible to estimate the volume of additional land needed and associated deforestation in the Reference Scenario. To avoid emissions from deforestation, ways would need to be found to reduce global demand for land, while maintaining the same level of products supply as the Reference Scenario. In systemic terms, mitigation of emissions through land-use change could be achieved by absorbing the expansion of these activities via the increased productivity of other ones. Brazil’s major agricultural activities already show high levels of productivity and consequently do not offer opportunities to increase productivity on the scale required to absorb these additional levels of demand for land. For example, the productivity of a soybean plantation in Brazil was 2.86 tons per ha in 2008, compared with 2.81 tons per ha in the United states (table 34). Table 34: Average productivity of selected crops in different countries (tons per ha), 2008 157 Crop (tons per ha) Country soybean Corn Cotton Rice Argentina 2.78 Bangladesh 3.93 China, People’s republic of 1.61 5.17 1.30 6.43 eU-27 5.67 India 1.06 2.3 0.57 3.31 Indonesia 4.66 Mexico 3.22 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Pakistan 0.65 Paraguay 2.62 Thailand 2.76 United States 2.81 9.46 0.99 Uzbekistan, republic of 0.83 Brazil 2.86 3.99 1.49 4.20 Beef-cattle farming shows much greater potential for increasing productivity per hectare, which can be applied to a much larger pasture area, since pastures occupy 207 million ha compared to 70 million ha for agricultural activities in 2030 in the reference scenario. Consequently, increasing the technological level and the intensification of livestock-raising can play an essential role in reducing the need for land for this activity, while releasing the land required for the expansion of other activities. In Chapter 2, in the section related to the calculation of emissions from livestock in the reference scenario (see 2.1.4.1.2 emissions estimates by prototypical systems), we saw that there are low-productivity productive systems, particularly: • Complete cycle in degraded pastures • Complete cycle in extensive pastures • there are also high-productivity production systems in Brazil, but only on a lim- ited scale, particularly: • extensive cow-calf in pastures + supplemented stocking and finishing in crop- livestock integration • extensive cow-calf in pastures + supplemented stocking and finishing in feed- lots. the modernization of the Brazilian livestock sector through the accelerated expan- sion of systems 3 and 4, to substitute for the low productivity systems 1 and 2, would enable the same amount of meat to be produced in a much smaller pasture area. Due to the large area currently occupied by low productivity systems – about 200 million ha – this substitution opens up the possibility of vacating very large volumes of pasture area compared to the expansion of other agricultural activities, which are already highly productive and only occupied about 52 million hectares in 2008. a 5 percent increase 158 in the agricultural area, or 2.5 million hectares, corresponds to only 1.25 percent of pasture area used by low productivity livestock systems. Thus, pasture areas may be vacated to accommodate the expansion of agricultural activities, virtually eliminating the need to clear new areas. the BLUM/sIMBrasIL tool that was developed in this study to model land use and its future changes helps qualify the substitution of low productivity systems with more productive systems year after year, and simulates the location of pastures that could be vacated to accommodate the economic growth of projected crop-livestock systems in the Reference Scenario, as well as new land uses considered in the Low-carbon Sce- nario. details of this quantification appear in the next chapter, which has a land-use scenario that is compatible with ghg emissions mitigation and uptake proposals con- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry sidered in the Low-carbon Scenario of this study. 4 Low-Carbon Land-Use scenario in Brazil a key conclusion from the study’s investigations on emissions mitigation is that in order to reduce deforestation, the main source of emissions, enough land from existing pastures must be vacated to accommodate new activities and thus avoid the conversion 159 of native vegetation. earlier sections presented opportunities for avoiding ghg emissions and carbon uptake associated with land use and land-use change, particularly emissions from ag- ricultural production and livestock activities, and carbon uptake via production forests and native forest recovery. But putting together a low-carbon land-use scenario is not simply a matter of adding (in the case of avoiding emissions) or subtracting (in the case of uptake) the volumes of greenhouse gases associated with these opportunities. For example, while increasing the land area allocated for forest recovery and production forests leads to carbon uptake and a reduction in emissions from iron and steel produc- tion, it also decreases the amount of land that is otherwise available for the expansion of agriculture and livestock activities. The potential conversion of more native vegeta- tion areas for the expansion of these agriculture and livestock activities would generate Technical Synthesis Report | Land Use, Land-Use Change, and Forestry carbon leakage. To avoid this situation, ways must be found not only to reduce the ad- ditional amount of land needed under the Reference Scenario, but also to release land for the mitigation and removal activities envisioned while maintaining the same level of products. 4.1 Additional Needs for Land for Carbon Uptake Activities and Biofuel Export In the Low-carbon scenario, more than 53 million ha is the amount of additional land needed for total emissions reductions and carbon uptake. of that amount, more than 44 million ha—twice the land expansion projected under the reference sce- nario—is for forest recovery under Brazil’s legal reserve law. the total volume of ad- ditional land required is over 70 million ha, more than twice the total amount of land planted with soybean (21.3 million ha) and sugar cane (8.2 million ha) in 2008 or more than twice the area of soybean projected for 2030 in the reference scenario (30.6 mil- lion ha) (table 35). Table 35: Mitigation and carbon uptake options for a Low-carbon Scenario and associ- ated needs for additional land Reference Scenario: additional expansion of agriculture and livestock production to scenario additional land needed (2006–30) volume of land required for the meet the needs anticipated in 2030: expansion of agriculture and >16.8 million ha 160 livestock activities Low-carbon Scenario: additional elimination of non-renewable charcoal in 2017 and the volume of land required for miti- participation of 46 percent of renewable planted char- gation measures coal for iron and steel production in 2030: > 2.7 million ha expansion of sugar cane to increase gasoline substitution with ethanol to 80 percent in the domestic market and supply 10 percent of estimated global demand to achieve an average worldwide gasoline mixture of 20 percent ethanol by 2030: > 6.4 million ha Technical Synthesis Report | Land Use, Land-Use Change, and Forestry restoration of the environmental liability of “legal forest reserves�, calculated at 36.2 million ha in 2030: > 44.3 million ha Total 70.4 million additional hectares one possible consequence is that land-use expansion for activities that promote lower levels of emissions, fossil-fuel substitution (as detailed in Chapter 4), or even car- bon capture may provoke an excess in demand for land use, which could in turn gener- ate deforestation, causing a lower net carbon uptake balance. 4.2 Toward a New Pattern of Productivity for the Livestock In- dustry The study simulated the new distribution of productive systems for livestock that should be promoted in order to liberate enough pasture land to accommodate all the needs for additional land due originating from crop expansion in the reference scenar- io, and for the implementation of new emissions reduction and carbon uptake options proposed under the Low-carbon Scenario. Figure 33: Mitigating measures for the construction of the Low-carbon Scenario 161 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry to increase livestock productivity per hectare—thereby absorbing agricultural ex- Source: ICONE (2009) pansion and other low-carbon activities without causing deforestation, while reducing emissions per unit of meat—five options were considered: (i) promoting the recovery of degraded pasture; (ii) stimulating the adoption of productive systems with feedlots for finishing; (iii) encouraging the adoption of crop-livestock systems; (iv) developing genetic improvement programs for higher quality, lower emissions forage adapted to Brazil, and (v) developing incentive programs for the use of genetically superior bulls. The projected effects of the productive systems considered for the reference and Low-carbon scenarios are compared below (Figure 35). a change in the pasture area is projected for the Low-carbon scenario, from 205.38 million hectares to 137.82 million hectares. a variation in the herd is also estimated to go from 201.41 (IBge, 2009) to 214.27 million head. these variations helped estimate the proportion of productive systems for 2008 and 2030 as presented in Figure 34. Figure 34: Change in pasture area occupied according to type of productive system (mil- lion hectares) 162 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Reference Scenario Low-carbon Scenario The new distribution of livestock production per production system can also be seen in Figure 35 below by the number of head of cattle in the different systems in the reference and Low-carbon Scenarios. Figure 35: Variation in number of head of cattle in productive systems, 2009-30 163 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Reference Scenario Low-carbon Scenario 4.3 Mitigation Potential of Direct Emissions from Livestock in the Low-carbon Scenario the variation in the systems’ composition generated substantial gains in land pro- ductivity. Projections indicate an increase in productivity from 47.22 kg of carcass equivalent /ha in 2008 to 63.51 and 95.42 kg carcass equivalent /ha in 2030, respec- tively, for the reference and Low-carbon scenarios, enabling a reduction of 68,239 mil- 164 lion hectares in the pasture area. Figure 36: Projection of Brazilian herd productivity between 2009 and 2030 for the refer- ence and Low-carbon Scenarios Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Figure 37: Projection of pasture area in Brazil from 2009 to 2030 (Low-carbon Scenario) 165 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry As a result of the substantial rise in meat production projected for the period, there was an increase in direct emissions for the sector. on the other hand, there is a consid- erable reduction in direct emissions in the Low-carbon scenario (34.1 thou Mg of Co2-e per year in 2030) due to the reduced emissions per production unit in the Low-carbon scenario (Figure 36). The transition from a lower to a higher productivity system alone has little effect on ghg emissions per animal (1.25 tCo2e in the degraded-pastures scenario versus 1.15 tCo2e in other scenarios). But higher productivity in more intensive systems generates a significant reduction in the size of the herd projected for 2030 (208 million head in the Low-carbon scenario versus 234.4 million in the reference scenario), which would in turn generate significant emissions reductions per unit of meat (Figure 38) and in the total value (Figure 39). Figure 38: Comparison of methane emissions from beef-cattle raising (Mt CO2e per year), 2008–30 166 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Figure 39: Comparison of methane emissions per unit of meat (kg CO2e per kg), 2008–30 167 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry The combination of improved forage and genetically superior bulls, together with the proposed increase in livestock productivity, would reduce direct livestock emis- sions from 273 to 240Mt Co2 per year by 2030, thereby maintaining emissions at about the 2008 level. 168 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Map 25: Number of heads of cattle Map 26: Total cumulative emissions from livestock, 2010-2030 169 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 4.4 A New Land-use Scenario for the Low-carbon Scenario With the data provided by BLUM on land requirements for the reference scenario, the land-use change simulation model was constructed again. In addition to the entry data, the difference between the reference and Low-carbon Scenarios is that for the lat- ter, when there is environmental liability in the micro-region, the deforestation rate is reduced to zero and environmental restoration is implemented through forest restora- tion. It is important to note that in both scenarios, the projection model for deforesta- tion in the Legal amazon is activated, simulating additional deforestation from indirect causes. With new data provided by the economic modeling team on the need for land in the Low-carbon scenario—the development of which is based on a wide array of im- provements in zootechnical livestock indices and the subsequent decrease in pasture area, more area for sugar-cane production, restoration of environmental liability with regard to legal reserves and PPas, and greater use of charcoal for ironworks—the land- use change simulation model adopted in the Reference Scenario was run again. Increased carrying capacity rates associated with greater herd productivity, as a combined effect of the recovery of degraded areas and the adoption of more intensive livestock stocking and finishing systems (integration of crop-livestock systems and feedlots), are reflected in an accentuated reduction in the demand for land, projected at about 137.82 million hectares in the Low-carbon scenario, compared to 207.06 million ha in the reference scenario for the year 2030 (table 36). the difference would be suf- ficient to absorb the demand for additional land associated with the expansion of agri- 170 culture and livestock in the reference scenario, as well as the expansion of mitigation and uptake activities in the Low-carbon scenario (Figure 40). Table 36: Comparison of land-use results for the reference and Low-carbon Scenarios (millions of ha) Reference Low-carbon difference scenario scenario (2030) be- tween low- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Var. Var. carbon and 2030– 2030– Reference grains (harvest) 38.94 37.79 47.92 8.98 47.86 8.92 (57) Land use 2006 2008 2030 2006 2030 2006 scenarios Sugar cane 6.18 8.24 12.70 6.52 19.19 13.01 6.49 Production forest 5.27 5.87 8.45 3.18 11.17 5.90 2.72 Pasture 208.89 205.38 207.06 (1.83) 137.82 (71.07) (69.24) Total area for agriculture and livestock1 259.27 257.28 276.13 16.85 216.04 (43.23) (60.08) Restoration - - - - 44.34 44.34 44.34 Balance 1.11 2 (15.74) herd (per 1,000 head) 205.890 201.410 234.460 28.570 208.000 2.120 (26.46) 1 Total area allocated to cotton, bean (1st harvest), corn (1st harvest), soybean, sugar cane, production forest, and pasture. 2 Represents expansion of agricultural area between 2006 and 2008 in the northern and northeastern regions. Source: ICONE Figure 40: Evolution of Brazil’s demand for land by crop, 2006-30 (millions of ha) 171 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Reference Scenario Low-carbon Scenario Source: Adapted from ICONE (2009) the first map (Map 27) shows that sugar cane areas undergo major land-use chang- es in relation to the Reference Scenario due to the great increase in demand for land for sugar cane for the expansion of ethanol production. however, geographic distribution patterns are maintained, with some intensification and, in traditional areas, spread more towards the central-west and interior of Bahia. Map 27 also shows the dynamics of the cotton crop in the Low-carbon scenario. Like in the reference scenario, yellow represents areas where cotton is cultivated in 2007 172 and 2030, remaining constant. areas in blue indicate where the cultivation of the prod- uct declined, and areas in red show where it expanded during the period modeled. the map shows that there were no significant changes vis a vis the reference scenario for cotton, as the demand for the product did not change between the two scenarios. Scenario (2010 – 2030). Yellow = crop permanence; blue = crop decrement; red = crop Map 27: Dynamics of sugar cane cultivation (left) and cotton (right) in the Low-carbon increment Technical Synthesis Report | Land Use, Land-Use Change, and Forestry results for rice are shown in Map 28. there have not been any significant changes in relation to the reference scenario. geographic distribution patterns, showing the crop’s expansion and regression, remain practically unchanged. the dynamics of the bean crop in the Low-carbon scenario are also shown in Map 28. there are no noticeable changes of any significance in relation to the reference sce- nario, as there have been practically no variations in the need for land for this product. Map 28: Dynamics of the rice (left) and bean crops (right) in the Low-carbon Scenario (2010-2030). Yellow = crop permanence; blue = decrement; red = increment 173 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Map 29 shows the results for corn. there is a significant change in the occurrence of this product compared to the Reference Scenario. Areas where the crop has decreased are in the western part of the states of rio grande do sul, santa Catarina, são Paulo and northern Paraná. on the other hand, there it has also increased in other parts of the same states, as well as in Minas gerais. With regard to soybean, a few changes can be seen (Map 29). Just like in the refer- ence scenario, there are some areas where the crop area has expanded and is growing near its original areas. Its geographical distribution pattern has not changed either, as it still found in the states of the south, central-west, Minas triangle and western Minas, western Bahia, Piauí, and Maranhão. A decrease in the soybean crop may be observed in são Paulo, which occurs to a lesser degree in the reference scenario, and can be justi- fied by the competition with sugar cane. Map 29: Dynamics of corn crop (left) and soybean (right) in the Low-carbon Scenario (2010-2030). Yellow = crop permanence; blue = decrement; red = expansion Map 30: Dynamics of planted forests (left) and pastures (right) in the Low-carbon Scenario (2010 – 2030). Yellow = remained constant; blue = crop decrement; red = crop increment 174 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Map 30 shows the forestry dynamics. due to the increase in the demand for pro- duction forests to neutralize deforestation for charcoal, there are many differences between the Reference Scenario and the Low-carbon Scenario. While in the Reference scenario there are practically no areas of expansion, in the Low-carbon scenario, these are rather clear, especially near regions with previously established plantations. Map 30 also shows pasture dynamics, again with considerable differences com- pared to the Reference Scenario. Since the low-carbon scenario includes land-use intensification, pasture area becomes the main donor of cropland, especially in the central-south and northeast of the country. With the exception of a few areas of expan- sion scattered throughout the northeast of Minas, rio grande do sul, Paraná and santa Catarina, a decrease in pasture areas predominates in this vast part of the country, a map of forest regrowth (Map 31) was created for the Low-carbon scenario. de- pending on the assumptions made for this scenario regarding the restoration of envi- ronmental liability according to the current Forest Code, restoration of the native vege- tation is being stimulated in the micro-regions where there is environmental liability at the beginning of the simulation. This regrowth occurs up to the limit required by law for the legal reserve (PPas were not considered in the regrowth due to the limits of spatial resolution adopted by the study). the permanence of areas of wild grass between 2010 and 2030 occurred mainly in the state of Maranhão, as well as in Minas gerais and Ba- hia, although to a lesser degree. only Maranhão presented areas where wild grass is de- clining. In the case of this study, the information used for developing the land-use map indicated a strong occurrence of areas with wild grass, and the same rule applies for this type of plant cover: deforestation only ceases when the legal barrier is overcome. Map 31: Forest regrowth in the Low-carbon Scenario 175 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Map 32 below synthesizes the evolution of the area occupied by agricultural and livestock activities in the reference and Low-carbon Scenarios. Map 32: area used for agriculture, pasture, and reforestation by region 4.5 Reduction of Deforestation in the Low-carbon Scenario A decrease in the need for land, which was calculated based on assumptions gener- ated by the Low-carbon Scenario, will lead to a reduction in deforestation rates com- pared to the Reference Scenario. New soil-use and deforestation maps were produced with the same spatial emissions model for land use developed with the ego dynamic platform (Map 33). the model for the Low-carbon scenario works like a legal scenario; 176 that is, when there is environmental liability, deforestation rates are set to zero and a simulation of a regeneration process for the micro-region in question is started. Map 33: Comparison of Cumulative Deforestation, 2007–30 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Reference Scenario Low-carbon Scenario Map 34: Total area deforested, 2010-2030 177 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Model-based projections indicate that, under the new land-use dynamic, deforesta- tion would be reduced by more than two-thirds (68 percent) compared to the refer- ence scenario; in the atlantic Forest, deforestation would be reduced about 90 percent, while the amazon region and Cerrado would see reductions of 70 percent and 65 per- cent, respectively. In the amazon region, the level of deforestation would drop quickly to about 17 percent of the historic annual average of 19,500 km2.30 (Map 34). It was expected that, with demand for pasture land reduced to zero as projected by the ICone module, deforestation rates would also be reduced to zero; however, that was not the case. deforestation still continues in certain parts of the amazon states of acre and Pará, with the model’s incorporation of indirect causes, through spatial lag regression (as in the reference scenario). thus, in micro-regions where the legal de- forestation limit was not reached in 2009—where there is still room for legal deforesta- tion and where the indirect dynamics modeled are the determining factors—defores- tation will continue unabated. Moreover, although residual deforestation is not quite at zero, the remaining amount is compatible with the 70-percent amazon deforestation-reduction target that the PnMC set for 2017, having as its baseline the historic average of 19,500 km2 per year. therefore, average annual amounts of 4,000 km2 produced by the model are be- low the 5,000 km2 per year threshold established as a final target for Brazil (Figure 41). 30 Between 1996 and 2005, the historical rate of deforestation in the amazon region was 1.95 million ha per year, according to the PnMC. Figure 41: Evolution of deforestation in the Low-carbon Scenario (curve) (km2 per year) 178 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry source: UFMg (2009) Figure 42: Evolution of deforestation in the low-carbon (LCS) and Reference Scenarios (RS) (thousands of ha per year) Map 35: Total cumulative emissions from deforestation, 2010-2030 179 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 4.6. Additional Measures for Protecting the Forest from Deforesta- tion Although the low carbon land-use scenario offers solutions for bringing the need for additional land virtually to zero, it is expected that complementary forest protection measures would also be required for two major reasons. First, the legal limit for defor- estation (up to 20 percent of properties located in the amazon region) has not yet been reached. thus, where the complex dynamic of deforestation is powered by the financial value of the wood or cleared land (along with with the need for cropland, pasture and production plantations), deforestation would continue. second, there may be a sig- nificant delay between the time demand for cropland, pasture or production forests is reduced and the time one could effectively observe a behavioral change among defores- tation agents at the frontier (i.e., since they may continue to speculate on demand that has already dried up far upstream in the land market chain). as already mentioned in Chapter 2, this reflects the fact that other indirect factors, in addition to the concrete need for additional land for the crop-livestock sector expan- sion, also play a role in the deforestation process. The model therefore includes indirect causes that are not captured by the land availability variables. These results support the urgent need to adopt additional measures to contain deforestation. The Low-carbon Scenario thus proposes to implement additional forest-protection measures in forested areas where deforestation is illegal. given the many ongoing pro- grams and abundance of literature available on this topic, including the Plan of Action for the Prevention and Control of deforestation in the Legal amazon (PPCdaM), this study was limited to analyzing existing proposals. Below is a list of the principal mea- sures, policies, programs and actions that aim directly or indirectly at reducing defores- 180 tation and its associated emissions. Protected areas expansion and Consolidation arPa Program – Continuation and expansion. In 2003, the Brazilian govern- ment initiated the implementation 31 of the amazon region Protected areas Program (arPa). over 30 million hectares32 of conservation units (CUs) have been created as Integral Protected Areas and Protected Areas with Sustainable Use under this program by means of an initiative supported by national (MMa and ICMBio) and international (World Wildlife Fund, World Bank, and KfW) partners through the Protected areas Fund. the program is being implemented in three stages (2003-2008; 2009-2013; and Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2014-2016) and will create about 50,000 ha of protected areas (table 37). Table 37: Snapshot of protected areas in the Amazon biome and ARPA participation Portion Protected area su- Protected or area of biome pported by aRPa military area no. (km²) (%) (%) 6 0.6 - 26,235 Military area 282 23.4 - 987,219 Indigenous land 44 3.3 22.5 137,385 state 37 5.5 80.6 Total 231,072 protection Federal 72 4.8 13.2 201,918 state 80 5.5 26.2 sustainable 233,523 use Federal 521 43.0 16.8 1,817,355 Total Source: Soares-Filho et al. (2008) 31 dF nº 4.326. 32 a v a i l a b l e a t < h t t p : / / w w w. m m a . g o v. b r / s i t i o / i n d e x . p h p ? i d o = c o n t e u d o . monta&idestrutura=154>. Last access on 10/05/2009. soares-Filho et al. (2008b) confirm the importance of protected areas and of ARPA in particular in helping to avoid deforestation. A decrease in the historic rates of deforestation in the amazon as of 2004-05 can be attributed, in part, to a series of measures that are part of the Plan of action for the Prevention and Control of de- forestation in the amazon, including the creation and consolidation of CUs. accord- ing to these authors, the probability that deforestation will occur around protected 181 areas is 10 times greater than in the interior. Based on an analysis of historic rates of deforestation around protected areas, the study demonstrated that there is no significant redistribution of deforestation in other areas due to the creation of pro- tected areas. Nevertheless, the consolidation of protected areas is a strong mitigating measure against the deforestation process observed in the amazon at a relatively low cost. The creation and consolidation of protected areas thus becomes part of a national deforestation reduction strategy once its maintenance can be ensured at relatively low cost. the same authors (personal data) estimate a cost of 1.3 to 10 billion dollars (nPV) for the consolidation and management of the network of pro- tected areas in the amazon over a thirty-year period. according to estimates made by amend et al. (2008), the maintenance cost for these areas will be Us$3.72 per ha. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry The PROdes program – Monitoring the Brazilian amazon by satellite, imple- mented by the national Institute for space research (InPe) since 1988 is funded by the Ministry of science and technology, with the collaboration of IBaMa and MMa. the analyses, carried out mainly based on the use of images from the tM sensor onboard the North American satellite Landsat and provides annual deforestation rates in the region, increments and decrements of deforested areas and specialized data in vector and raster formats. The results are widely used by the national and international sci- entific community and were important for raising awareness about the deforestation process in the region. deTeR – (detection system for deforestation in Real Time), another program developed by InPe, is based on data from the ModIs sensor from the Land/Water sat- ellite and WFI sensor from the CBers satellite (the data is less refined than Prodes data). the deter system aims at the rapid monitoring of the deforestation dynamic in the Legal amazon, in an effort to provide support to supervisory activities. there are monthly reports during the dry periods and trimestrial reports during rainier periods, due to cloud presence. however, the program only manages to identify areas larger than 25 ha. a third program, degrad – Mapping of Forest degradation in the Brazil- ian amazon maps degraded (i.e. partially deforested) forest areas in the amazon using CBers and Landsat satellite imaging. studies dating back to 2007 enable areas of up to 6.5 hectares to be identified at different stages of degradation (see Figure 43). Figure 43: Identification of forest degradation patterns in the Amazon within the frame- work of the DEGRAD program. Source: INPE, 2009 182 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry According to the National Research Institute management report, the following resources have been available for monitoring projects via satellite for the amazon (in- cluding the three aforementioned programs) for the last three years (table 38): Table 38: Resources of the INPE for monitoring the Amazon by satellite 2006 1,415,506.00 456,708.55 Year Budget (R$) Total spent (R$) 2007 2,750,000.00 2,072.634.00 2008 2,850,000.00 2,077.178.20 Source: INPE, 200933 B. Development of Integrated Projects in the amazon, coordinated by the President’s office, is implemented through the PPCdaM – The Plan of action for the Prevention and Control of deforestation 33 available at coordinated action of 13 ministries. the general aim of PPCdaM is to reduce deforesta- tion rates in the Brazilian amazon through a set of integrated actions including territo- rial and land ordinances, monitoring and evaluation to foster sustainable production activities involving partnerships between federal agencies, state governments, may- oral offices, civil society and the private sector. PPCdaM has three main axes around which activities are conducted: (i) land and territorial ordinances, (ii) environmental 183 monitoring and evaluation, and (iii) productive and sustainable activities. during 2008-11, the government plans to invest approximately Us$500 million in PPCdaM- related initiatives. The sustainable amazon Program (Pas): strives for a new development land- scape by focusing on environmentally sustainable, economic solutions. Its targets and directives are based on a current diagnosis of the amazon. the program is implement- ed according to an agreement between federal and state governments and promotes the integration of promotion and production. one of the driving forces behind the ap- plication of resources from the amazon Fund together with PPCdaM, it is based on the principal that more efforts are needed to ensure the sustainable development of the forest’s socioeconomic potential if impact mitigation, through the creation of CUs, does not prevent the deforestation of the amazon. actions and strategies must be imple- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry mented with greater local government participation. The Program also helps regulate space appropriation dynamics while providing suitable conditions for populations and communities by guaranteeing their social rights. The participation of private capital is essential for providing conditions for the implementation of these projects (PPg7, FaM, etc.) (MMa, 2008). C. Creation of Forest Protection Funds amazon Fund (FaM): established by decree nº 6.52734, the fund aims to secure donations through non-reimbursable investments for actions to prevent, monitor and combat deforestation, and to promote the conservation and sustainable use of forests in the amazon biome. It entails different activities: management of public forests and protected areas; environmental monitoring, evaluation and supervision; management; economic activities based on the sustainable use of the forest; Zee territorial and land regularization system; conservation and sustainable use of biodiversity; and restora- tion of deforested areas. actions should comply with Pas and PPCdaM directives, as the fund was developed within the broader context of Brazilian public preservation policies. the fund’s resources (managed by Bndes) come from donations from certi- fied donors, and equivalent value in tons of avoided carbon, calculated according to a methodology to be established by a technical team. the fund has a technical Commit- tee (responsible for testing MMa emissions calculations) and an orientation Commit- tee (responsible for the implementation and preservation of the fund’s initiatives and goals). norway signed a contract to donate Us$110 million to the fund and the country plans to donate the first installment of Us$1 billion by 2015. D. Sustainable use of forest resources and payment for environmental services: Public Forest Management: To promote forest conservation, the concession for the sustainable use of public forests aims to increase forest appreciation. In support of this goal, Law 11,284 was created in 2006 to regulate forest management in public 34 available at . Last access on 09/05/2009. areas; the law also established the Brazilian Forest service (sFB) and national Fund for Forest development (FndF). this legal document establishes three types of man- agement for sustainable forest production: (i) the creation of conservation units for sustainable forest production, such as FLonas; (ii) the non-onerous use of forests for sustainable and social development; and (iii) paid forest concessions35, based on a public bid, to guarantee access to products traditionally used by local populations. The sFB will be responsible for the public forest management system, for stimulating sus- 184 tainable forest development and for the management of the fund (FndF). In the case of concessions, decisions are made with the aim of ensuring that this process benefits society as much as possible. thus, the choice of concessions (which cannot exceed 20 percent of the area to be conceded the first 10 years) use criteria such as best price, less environmental impact, greater socioeconomic benefits, improved efficiency and aggre- gation of local value, and restriction to national companies. Table 39: Implementation of the public forest management systems: benefits and losses Beneficiaries of the new public forest management Measure Beneficiaries Losses Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Public Forest system will be the local communities who live off of Areas opened up Management: forest products, and who want to participate in the through defores- Implementa- regional economic dynamics, formalizing their en- tation; it is hoped tion of the trance into the market, expanding the multiple uses but not guaran- Brazilian For- and enjoying non-onerous conditions; and private teed that this est System - business people who prefer not to buy land, who want process will be Law to use the forests legally, so that they may have access slowed down. 11.284/06 to credit, export, tourism, reforestation of degraded areas, certification, jobs and income. the Forest grant Plan annually defines the public forests that may be subject to con- Source: (Azevedo, T., Tocantins, M.A., 2006) version, as identified in the national register of Public Forests. It also defines the nec- essary management resources, especially with regard to monitoring. The table below presents estimates of the resources necessary for implementing activities planned for 2009: 35 Do not imply rights of ownership, only use of resources. Table 40: Summary of expenditures anticipated for Public Forest Management services in 2009 national register of Public Forests 8 activities anticipated (summary) Resources (million R$) Support forest management activities 7.8 185 Forest concessions 10 Monitoring of public forests 15 Creation of the national system of Forest Information (sistema 5.4 nacional de Infomações Florestais) national Forest development Fund (Fundo nacional de desen- 2.5 volvimento Florestal) Implementation of the sFB administrative structure 8 Total 56.7 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Source: Plano Anual de Outorga Florestal (Annual Forest Grant Plan), 200936 Forest allowance (Bolsa Floresta): one of the first programs to apply the concept of paying for environmental services in Brazil. Implemented by the state government of amazonas, it plans monthly payments of r$50.00 to families registered by the project, and residents of state CUs. the families’ permanence in the program is linked to the de- velopment of sustainable activities in these areas, which principally revolve around the reduction of deforestation activities. the state target covers about 60,000 families in the program and extends access to indigenous communities. Program resourcescome from the state Fund for Climate Change, environmental Conservation, and sustainable development, which was created by the state Law for Climate Change nº 3,13537. Program for the socio-environmental development of rural Family Production in the amazon (ProaMBIente): initiated by social movements representing small- holder farmers in association with IPaM in 2001, having been adopted earlier by the environment Ministry and included in the Pluriannual Plan. It seeks public policy in- novations for the development of smallholder farmers in the amazon region, but is also compatible with the current environmental paradigms. It thus seeks to overcome the impression of rural credit as the only economic instrument for development, and suggests compensation for environmental services38 performed by farmers linked to the program as a tool, once they have transitioned to sustainable production systems. according to the environment Ministry’s report, the program should already be ben- efitting almost 4 million families from traditional communities in 148 groups, which, in turn form 11 poles that are distributed throughout amazonia. In this program, the Production Unit becomes the basic unit, each one represented by a community group. 36 available at http://www.mma.gov.br/estruturas/sfb/_arquivos/paof_2009_vf_95.pdf Last access on 11/05/2009. 37 available at < http://www.florestavivaamazonas.org.br/download/Lei_est_n_3135_de_050607. pdf > 38 For example, the reduction of deforestation, the recuperation of environmental liabilities, soil, water, and biodiversity conservation, reduction of the use of agrochemicals, reduction of the risk of fire, more sustainable energy matrix, transition to agroecology. Agreements made between groups from each pole establish environmental service tar- gets to be achieved in each unit. The condition for payment to the farmers from the 11 poles is the achievement of the goals established in the agreements. The program also seeks the qualification of participating farmers through ater (Assistência Técnica e Ex- tensão Rural - technical assistance and rural extension), which is composed of agents selected from the community itself, and aims at the formation of a single cooperation network. In addition, the project receives support from the Brazilian national environ- 186 mental Fund – FnMa and the embassy of the netherlands. With regard to the amount of resources involved, in one of the pioneer poles of the program, the transamazon Pole (which includes families from three municipalities in the state of Para), received a com- pensation payment for 6 months for 340 families in the amount of r$100.00 /month per family, an additional compensation of r$126.00 per family for buying material and tools, and salaries of r$380.00 for new community agents for eight months (nepstad et al., 2007). e. environmental certification socio-environmental register: the socio-environmental Commitment register (CCs) is a voluntary register of properties whose owners are committed to improving the “socio-environmental performance� of their properties (http://www.yikatuxingu. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry org.br/projetos/ver/48). the CCs already has over 1.5 million hectares of property, a large part of which is located at the headwaters of the Xingu river. With regard to the CCs, registered properties receive preferential treatment from meat-processing plants in the region (e.g., the Independência and Bertim meat-processing plants already pay a better price for an arroba (15 kg) of beef cattle from properties listed in the CCs. 4.7 Balance of emissions from land use and land-use change in the Low-carbon Scenario Based on mitigation options for direct emissions, and the reduction of emissions linked to deforestation and carbon uptake through plantations, the sIMBrasIL model calculated annual emissions for the 2007-2030 period resulting from land use and land-use change for each micro-region. Compared to projections in the reference scenario (Figure 44), emissions from de- forestation are considerably lower under the new land-use dynamic considered in the Low-carbon scenario (Figure 45), at about 170-190 Mt Co2e per year over much of the period. This decrease is due to less demand for pasture area and the subsequent drop in the need to convert land via deforestation. annual land-use emissions (i.e., agricul- ture and livestock) increase 310 to 340 Mt Co2e over the period, with agricultural emis- sions accounting for most of this increase. still there is a 6 percent overall reduction in emissions compared to the reference scenario. Ch4 emissions from beef cattle remain relatively stable at 236 to 249 Mt Co2e per year, since the gains from reduced Ch4 pro- duction per unit of meat are offset by increased production. of these emissions caused by different types of land use, livestock is the greatest emitter, surpassing emissions caused by the deforestation of the amazon. Finally, carbon uptake shows a growing trajectory, presenting an initial rate of ap- proximately 133 Mt Co2 per year for 2010 and a final rate of 213 Mt Co2 per year for 2030, as a function of the growth in forest plantation cover and recuperation of envi- ronmental liabilities of legal reserves and PPAs. The resulting balance between use, change, and uptake shows a decrease in the amount of net emissions between 2007 and 2030, reaching a rate of approximately 321 Mt Co2e per year in 2030, a reduction of 187 nearly 65 percent compared to the reference scenario39. Figure 44: Reference Scenario results: emissions from land use and land-use change, 2009–30 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 39 If the carbon uptake from the natural regrowth of degraded forests were to be included, then the potential uptake would increase by 112Mt Co2 per year on average, thus reducing net emissions. Figure 45: Emissions from land use and land-use change under the new land-use dy- namic in the Low-carbon Scenario 188 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Map 36 below compares net emissions generated by land use and land-use change in the Reference Scenario and in the Low-carbon Scenario for each unit of the federa- tion. Map 36: total cumulative emissions from land use (agriculture, livestock, deforesta- tion, and reforestation) 2010-30 Emissions reduction from deforestation and carbon uptake through plantations and forest recovery are much greater than the reduction of emissions from all the other sectors considered in the global low-carbon study for Brazil (energy, transport and waste). the decrease in deforestation and increase of forest plantations are two areas where the Low-carbon Scenario proposed had the greatest success in reducing emis- sions. together, these two areas represent 67 percent of the net reductions registered 189 during the 2010-30 period (table 41). It is more difficult to reduce emissions from the transport and energy sectors, as they are already low compared to the international standards, mainly due to the considerable amount of hydroelectricity and bioethanol in the current energy matrix. Table 41: Comparison of cumulative emissions distribution among sectors in the refer- ence and Low-carbon Scenarios, 2010-30 Reference sce- Low-carbon sce- nario (2010–30) nario (2010–30) Reduction % of refer- ence Scenario Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Mt % of Mt % of to- Mt % of (2010–30) Co2e total Co2e tal Co2e total Land use 16,709 55 9,228 48 7,481 67 44 sector Waste 1,633 6 375 3 1,258 12 78 Transport 4,101 14 3,614 19 487 5 13 Energy 7,587 25 5,763 30 1,824 16 24 Total 30,030 100 18,980 100 11,050 100 37 Consequently, the distribution of ghg emissions among the different sectors in the Low-carbon scenario differs significantly from the distribution observed in the refer- ence Scenario, mainly because the amount of emissions from deforestation is reduced to approximately 70 percent compared to the reference scenario (Figure 46). Figure 46: Comparisons of gross emissions distribution among sectors in the reference and Low-carbon Scenarios, 2008–30 190 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 4.8 Key Uncertainties for Emissions Estimates Since the reference and proposed Low-carbon Scenarios are subject to uncertain- ties, the results are indicative and should be used to inform stakeholders of future emis- sions if the study’s assumptions, which were based on a broad and ongoing consulta- 191 tive process, are verified. some of the uncertainties result from calculations related to either the reference or Low-carbon Scenario, while others concern both. This section first outlines overall uncertainties for the four main areas and then addresses more sector-specific ones. Macroeconomic Projections For emissions-generating activities, both the reference and Low-carbon scenarios depend heavily on the macroeconomic projections of the 2030 national energy Plan (Pne 2030) published by the ePe in 2007. the plan’s B1 scenario, adopted as the refer- ence case, estimates that the Brazilian economy’s average growth rate at 4.1 percent annually. as a consequence of the recent financial crisis, the Brazilian government expects lower gdP growth, particularly in the near term. If so, decreased supply and demand for a variety of services and products would slow the pace of deforestation and Technical Synthesis Report | Land Use, Land-Use Change, and Forestry energy consumption, including the demand for transport services. however, given the longer–term timeframe of the study, medium-term projections for emissions growth under the Reference Scenario are less affected by the crisis and would remain about the same. The same short- and medium-term trends would also apply to the Low-carbon Scenario. Land-use Questions With respect to uncertainties for projected land-use emissions, one must distin- guish between the gross volume of ghg emissions and net emissions obtained after taking into account carbon uptake activities involving mainly production forests and native forest recovery. Uncertainties for gross emissions differ between the first and second stages of calculations: (i) projecting land use and land-use changes and (ii) con- verting the results into emissions. the economic modeling developed for the first stage of calculations benefited greatly from the wealth of historical local data, which allowed for robust calibrations of the key parameters and equations (Box 3). Based on the results, it was assumed that the main uncertainties are linked to the abovementioned macroeconomic projections, which directly affect projections for expanded cropland and meat production and thus deforestation. If cropland and meat production expand more than expected under the Reference Scenario, then more effort will be required under the Low-carbon Scenario to release enough pasture; otherwise, the additional deforestation that would result would lead to increased emissions. For the second stage of calculations, the main uncertainties are based on available data for soil carbon content and the vegetation converted, which drive the conversion of deforestation into ghg emissions. estimates of the above- and below-ground carbon content of biomass depend on the accuracy of the data, which can only be improved by intensive field research. the uncertainty of the data used for this national study is esti- mated at about 20 percent, which mainly affects the Reference Scenario, since conver- sion of native vegetation is at very low levels in the Low-carbon Scenario. Under the Low-carbon Scenario, an added uncertainty is the pace of releasing pas- ture for expanding agricultural crops to avoid deforestation and comply with the legal scenario adopted as a target for forest recovery–based carbon uptake. the rapid fall of deforestation-based emissions entails considerable efforts to improve livestock productivity to free up pastures for other activities. to the extent that the release of pasture keeps pace with the annual need for additional land for crop expansion, the conversion of native vegetation would no longer be needed; in theory, deforestation 192 and related emissions would then be brought to zero. Key questions are whether the pace of pasture release and agricultural expansion will match and whether the neces- sary conditions will be created to ensure that the pace of agricultural expansion is not too rapid. Achieving the right pace on the livestock side and providing the right incen- tives—positive or negative—for forest protection are critical. If the required financial disbursements are not made on time, deforestation and its related emissions will con- tinue unabated. another uncertainty involves the expected effect of productivity gains on the growth of livestock. In the study, the Brazilian share of the international market is taken as an exogenous projection from FaPrI (Box 3). Increased productivity could im- prove competition and thus spur increased production. Since productivity gains mean Technical Synthesis Report | Land Use, Land-Use Change, and Forestry less need for pasture area, such a rebound effect should not cause more deforestation, provided such gains are limited to the areas of the former low-productivity systems. Uncertainties inherent to the economic modeling of future land-use scenarios are related Box 3: Uncertainties for economic land-use scenarios to the modeling of (i) domestic demand (a function of income, ultimately linked to mac- roeconomic projections and equilibrium prices determined by the modeling), (ii) exports (a function of macroeconomic parameters and prices), and (iii) production (a function of costs and productivity per hectare). Price elasticities were calibrated from a historical se- ries (1996–2008), while production costs and per-hectare productivity for various crops were based on data from the national supply Company (ConaB); the Brazilian Institute of geography and statistics (IBge); and agroconsulta and scott Consultoria, two private firms that update estimates for the sector on an annual basis. Brazilian export projections are exogeneous and were based on global projections of the Food and agricultural Policy research Institute (FaPrI), the same source used by the U.s. department of agriculture; FaPrI projections were used to calibrate export projections for 2009–18 and 2019–30. It was thus assumed that the key uncertainties are linked to macroeconomic projections. Under the reference scenario, projections for meat exports and pasture are relatively conservative. With the exception of the amazon region, where significant growth in pas- ture is expected, volume nationwide remains fairly stable, which can be attributed to the continued stability in the global meat demand. stabilization—or even a slight decrease in meat exports, observed over the past several years—is difficult for Brazilian industry to reverse, following the impressive development of the previous decade (1997–2006). Source: ICONE Under the Low-carbon Scenario, the main carbon uptake potential resides in the recovery of legal forest reserves. Indeed, the proposed Low-carbon Scenario consid- ered full compliance with the Forest reserve Law—including an enormous effort to re- cover riparian and native forests—as a target for carbon uptake. this “Legal scenario� would imply a break with the past. a fully Legal scenario may be difficult to implement and flexibility mechanisms are already being discussed, especially regarding legal re- 193 serves, which may reduce the net area reforested. For example, in such amazon states as rondônia and Pará, which have already developed economic and ecological zoning, the legal reserve can be reduced from 80 percent to 50 percent, particularly for rural properties located along the main roads. In exchange, landowners would commit to fully restoring the 50 percent legal reserve, with the abated 30 percent converted into “agriculture consolidation areas.� Therefore, the carbon uptake volume indicated in this study may be at the upper limits of the range. Building flexibility into target setting would reduce the volume of carbon sequestered. At the same time, it would facilitate the effort of releasing the cor- responding amount of pasture and thus mitigate the risk of inducing carbon leakage. That is, conversion of native vegetation would occur somewhere else as a result of the domino effect triggered by the induced net reduction of land available at the national Technical Synthesis Report | Land Use, Land-Use Change, and Forestry level for crop and livestock expansion. In terms of carbon balance, avoiding the release of the full carbon stock of one hectare of burned forest in the atmosphere is preferable to the progressive removal of ghgs from the atmosphere through the restoration of one hectare of forest. It is thus essential to ensure consistency between efforts to release pasture and enforce the restoration of legal reserves. 4.9 Benefits Related to Reducing Aerosol Emissions Resulting from Deforestation by Burning A study was conducted with the pupose of generating estimates for aerosol emis- sions from burning in the projected scenarios, the effects of land-use change and soil cover on surface flows, and lastly, how these changes affect the hydrologic cycle of south america, especially amazonia. Burning, which occurs mainly in the tropics, is a major source of atmospheric pollut- ants (artaxo et al., 2002, andreae, 1991). In south america, during the winter months, hundreds of thousands of fires are set principally in the cerrado and forest ecosystems. this burning occurs mainly in the central and amazon regions, although the spatial distribution of the smoke covers an extensive area amounting to about 4-5 million km², far greater than the area where the fires are concentrated (Freitas et al., 2005, 2006, 2007). gases are released into the atmosphere during biomass combustion, includ- ing some greenhouse gases, tropospheric ozone precursors and aerosol particles that interact efficiently with solar radiation, affecting microphysical processes, cloud for- mation dynamics and air quality. The effects of these emissions are widespread, affect- ing the composition and physical and chemical properties of the atmosphere in South America and nearby oceanic areas on a regional scale and potentially on a global scale. Emissions from burning change the atmospheric radiative balance both region- ally and globally through the direct effect of aerosol particles when they reflect and scatter solar radiation back in space, reducing the quantity absorbed by the land area, and when they absorb solar radiation, thereby heating up the atmosphere. Jacobson (2001) suggests that atmospheric warming due to black carbon aerosols could balance the cooling effect associated with other types (sulfates), and that their direct radiative force could exceed that of Ch4. Thus, aerosol particles produced from incomplete com- bustion processes would only come second after Co2 in terms of contributing to atmo- spheric radiative heating. 194 The balance between radiation and the hydrologic cycle can also be affected indi- rectly by emissions from burning through micro-physical alterations and the dynamics of cloud formation (Kaufman, 1995), due to the greater availability of cloud condensa- tion nucleii (CCn) and ice in the atmosphere, which cause changes in cloud drop spec- tra (andreae et al., 2004; Koren et al., 2004; rosenfeld, 1999; Cotton and Pielke, 1996) and in thermo-dynamic stabilization (Longo et al., 2006). the increase in the concen- tration of aerosol particles results in the production of a greater number of smaller cloud drops, with two outcomes: first, the greater quantity of drops reflects more solar radiation back to space (although it cools the atmosphere), and second, the smaller size will be less favorable for rain production, as the tiny droplets tend not to stick together to form large drops that become rain. on the other hand, thermo-dynamic stabilization Technical Synthesis Report | Land Use, Land-Use Change, and Forestry caused by the direct interaction of aerossol particles with solar radiation (reducing the heat in the low atmosphere by reducing solar radiation), restricts the rise of convective cells generated close to the surface, thus inhibiting cloud formation. This set of factors suggests that the effects of burning can have an impact on a local scale, with a major impact on the regional hydrologic cycle, as well as the planetary energy redistribution pattern in the tropics for medium and high latitudes. on the other hand, land-use change can lead to variations in the balance of energy, water and momentum on the surface, due to the corresponding changes in its albedo, the evapotranspiration capacity associated with the plant cover and its spatial struc- ture. In particular, the substitution of forested areas with deep root systems for pas- ture areas results in an increase in the albedo and less accessibility to deep soils with substantial water storage potential. This change generally leads to an inversion in the Bowen ratio-energy balance, producing drier, hotter and deeper planetary layer limits, mainly during the dry season. Thus, land-use changes cause alterations in the pattern of the hydrologic cycle, which can be evaluated using numerical values and land-use scenarios. another relevant aspect that has not been studied extensively is the effect of land- use change on dust aerosols. With more exposed soils and intense winds (which can be expected with the decrease in land rugosity when forests are replaced by pastures) there may be a significant increase in the production and removal of dust from the soil, impacting the radiative balance as well as the cloud microphysics and hydrologic cycle. 4.9.1 Methodology: Numerical Modeling with CCATT-BRAMS The methodology used was based on numerical atmospheric modeling using the emissions model, chemical reactivity, transport and deposit of gases and CCatt aerosols (Coupled Chemistry-aerosol-tracer transport) combined with the atmospheric BraMs model (Brazilian developments on the regional atmospheric Modeling system). the BraMs is a numerical meteorological model that simulates atmospheric circulation from hemispheric scales to scales of major turbulences in the planetary boundary layer. The model has a multi-layer arrangement that enables the different spatial and temporal scales (Walko et al., 2000) to be resolved simultaneously, with state-of-the-art physical parametrizations and a modern parametrization of cumu- lus clouds developed in the formalism of the ensemble (grell and devenyi, 2002). 195 the CCatt is a numerical system that is designed to simulate and study emissions, transport, deposits and physical and chemical processes associated with trace gases and atmospheric aerosols. It is a Euleriano transport model that merges completely with the BraMs, enabling the simultaneous numeric provision of time, air quality and impact of aerosols and land-use changes on atmospheric development (Freitas et al. 2005, 2006, 2007; Longo et al., 2006, 2007). It has a transport model that re- solves phenomena on the grade and sub-grade scale (the main processes are shown in Figure 47), besides a complete chemical mechanism for the prognosis of chemical reagent species. Figure 47: Transport processes simulated by the CCATT-BRAMS, including plume rise, deep Technical Synthesis Report | Land Use, Land-Use Change, and Forestry and shallow convective transport by cumulus, diffusion in the PBL, dry and wet deposition different types of aerosols are parametrized by the CCatt, including particu- late matter generated by burning, resuspension of dust from the soil, agricultural activities and emissions of urban/industrial origin. Its CarMa diagram (Commu- nity aerosol & radiation Model for atmospheres) enables the effects of long and short waves in aerosol and hydrometeor particles to be evaluated. This means that the model is able to conduct studies on the direct and indirect effects of aero- sols on the radiative balance, as well as calculations of rates of heat, providing an important tool for studying the interaction between aerosols and the atmosphere (Longo et al., 2006) The effects of land-use change on atmospheric circulation are studied through the use of a series of numerical simulations whose pattern of occupation of the surface area was described by reference and Low-carbon Scenarios generated by other groups. In this study, land-use and carbon equivalent emissions maps are used for the 2007-2030 period. The characteristics of each type of occupation shown by land-use maps were obtained from the literature. These include relevant biophysical properties such as albedo, rugosity, leaf area index and root depth. these properties enable the parametri- zation of the area through the simulation of sensible and latent heat flows and momen- tum. Different effects were considered either separately or together in each simulation, enabling an understanding of the individual impact and the potential existing feedback. 196 The following section shows how equivalent carbon data and land-use maps were used to estimate aerosol emissions. 4.9.1.1. Calculation of Aerosol Emissions Two types of data were used to calculate aerosol emissions: land-use maps and car- bon equivalent emissions maps for Brazil. these data include the period from 2007 to 2030 and were generated for both the reference and the Low-carbon scenarios. Both sets of data had 1x1 km resolution and were produced by the topic a team. their use will enable aerosol emissions to be estimated based on the quantity of carbon equiva- lent available in the atmosphere for both scenarios being studied. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Emissions were assumed to be coming from biomass combustion generated by the deforestation of forest areas that were later converted for other uses (such as agricul- ture and pasture). these deforested areas were located using land-use maps, which illustrate the type of biome occupation present in a given area (Map 37). Using this type of annual map, it was possible to determine the annual development of forest cover, and thus to locate regions where it is disappearing (deforestation). While land-use maps were used to locate points of aerosol emissions, carbon equiv- alent maps were used to determine the quantity of aerosol emitted at these points. only positive carbon equivalent values were used, as the objective was to estimate emissions based on the amount of carbon released into the atmosphere. The necessary calcula- tions for transforming carbon equivalent emissions into aerosol emissions follow these steps: a) estimate quantity of carbon equivalent emitted through combustion. b) transform this quantity into carbon dioxide emissions. c) obtain emissions from aerosols using the values of emissions factors available in the literature. For calculating the item, it was assumed that 85 percent of the carbon equivalent emitted was the result of combustion processes (soares Filho, personal communica- tion), thus we have (equation 48): [carbon equivalent emissions from combustion] = 0.85[carbon equivalent emissions] (48) a factor of 3.66 was adopted (soares Filho, personal comunication) for the transfor- mation described in item b for the conversion of carbon equivalent emitted in terms of carbon dioxide (equation 48b): [carbon dioxide emissions] = 3.66[carbon dioxide equivalent] (48b) Map 37: Land-use map for the year 2007 in the Reference Scenario (1x1km resolution) 197 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Item c was realized using values from emissions factors for carbon dioxide and aero- sols with which it is possible to estimate the fraction of Co2 attributed to particulate matter and then estimate aerosol emissions. table 42 shows some values from emis- sions factors associated with forest, savanna and pasture according to work done by andreae and Merlet, 2001. Table 42: Emissions factors (g/kg) for different biomes for CO2 and aerosols (particulate matter with a less than 2.5 micrometer diameter - PM2.5) Co2 PM2.5 tropical Forest 1580 9.1 Savanna 1664 4.9 Pasture 1664 4.9 As deforestation was assumed based on biomass combustion from forest remnants, Source: Andreae and Merlet, 2001 amounts associated with tropical forests, or 1580 g/kg for the carbon dioxide emis- sions factor and 9.1 g/kg for aerosols were used. Considering what was said in previous sections, the expression for the calculation of aerosol emissions may be written as fol- lows (equation 49): 198 which directly illustrates the conversion of carbon emission equivalent into aero- sols emitted by combustion. The emission units are the same as for the carbon equiva- lent, in other words tons per hectare and per year. to finalize, in order to enter emissions data into the CCatt-BraMs, it was necessary to parametrize the annual performance of these emissions. as the model generates results every 6 hours, emissions data were converted into a larger temporal resolu- tion through performance curves for the number of emissions sources generated by the 6 different regions of Brazil, as shown in Map 38. thus, annual emissions data were Technical Synthesis Report | Land Use, Land-Use Change, and Forestry converted into daily data and inserted into the model. It should be noted that emissions generally peak between August and November, depending upon the region. Map 38: Schematic map of Brazil showing the different regions in the country and their boundaries for the analysis of results (above). Below, normal performance of the number of emissions sources in the different regions obtained with data from the AVHRR sensor (Advanced Very High Resolution Radiometer) from 1998 to 2008, present in the satellites of the NOAA series (National Oceanic and Atmospheric Administration) 4.9.1.2 Aerosol Emissions in the Reference and Low-carbon Scenarios This section discusses results obtained for emissions estimates from aerosols in the reference and Low-carbon scenarios. only emissions from deforestation according to the methodology were taken into account in the calculations. The annual emissions performance may be seen in Figure 48 and table 44. Both scenarios show a drop in emissions until the year 2010. In the Low-carbon Sce- 199 nario, the drop is more sudden, reaching approximately 6000 tons per hectare in 2010. In the reference scenario, this amount is about 16,000 tons (62 percent higher). From 2010 on, emissions pratically stabilized in the Low-carbon scenario, with close to 7600 tons in 2030. on the other hand, there was an increase in the reference scenario, with maximum emissions of approximately 22,400 tons occurring in 2030, although this is below the 2007 emissions level (22,800 tons). according to table 43, emissions in the Low-carbon scenario are 62 to 66 percent less than those of the reference scenario be- tween 2010 and 2030. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Figure 48: Estimate of total annual aerosol emissions in Brazil for the reference and Low- carbon Scenarios (Table 43) Table 43: Total annual aerosol emissions (tons per hectare and per year) throughout the country for the reference (REF) and low carbon (LC) scenarios. Also shown are the figures of absolute differences (LC-REF) and differences in percentages (LC-REF (%)) between year reF LC LC-reF LC-reF (%) emissions for the two scenarios. 2007 22789 22854 65 0.3 200 2008 20505 20253 -252 -1.2 2009 18684 6397 -12286 -65.8 2010 15879 6035 -9844 -62.0 2011 16723 5975 -10748 -64.3 2012 16234 5940 -10295 -63.4 2013 17244 6242 -11002 -63.8 2014 17594 6655 -10939 -62.2 2015 17996 6411 -11585 -64.4 2016 18534 6267 -12267 -66.2 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2017 18428 6546 -11882 -64.5 2018 17960 6669 -11291 -62.9 2019 18532 6814 -11718 -63.2 2020 18929 6859 -12070 -63.8 2021 18468 6977 -11491 -62.2 2022 19063 6972 -12091 -63.4 2023 19282 6958 -12323 -63.9 2024 20224 6947 -13276 -65.6 2025 20021 7084 -12937 -64.6 2026 20729 7238 -13491 -65.1 2027 20604 7147 -13457 -65.3 2028 21002 7160 -13842 -65.9 2029 20507 7301 -13207 -64.4 2030 22373 7584 -14789 -66.1 Total 458305 191286 -267020 -58.3 Map 39 shows the spatial distribution of aerosol emissions in the country in the two scenarios, concentrating principally on the amazon rainforest and in its transition re- gion with savannas. The difference in the spatial distribution between the low-carbon and Reference Scenarios is principally due to the difference in intensity of the pace of deforestation in the regions. In the Reference Scenario, mainly in the transition areas of the amazon rainforest, conversion of forest areas into pasture is intense, resulting in an increase in areas of emissions. the area of pasture expansion in the country went from 2.6x106 to 2.8x106 km² in the reference scenario. as a visual comparison, Map 39 also shows the aerosol load simulated by the CCatt- BraMs model, represented by the optic depth of the aerosol, illustrating the result ob- tained by entering amounts of emissions from burning into the model. Map 39: Figures (A), (B), (C) and (D) show the locations of deforestation from 2007 to 2030 in the reference (REF) and low carbon (BC) scenarios. Regions with forest remnants 201 are also shown during that period (in green). Figures (E) and (F) show the optic depth of average aerosol from 2007 to 2030 in the reference (E) and low carbon (F) scenarios, where the current lines represent the average wind field over Brazil Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 4.9.2 Results one important aspect that emerges is that, although emissions from burning come principally from the amazon region, the smoke emitted may be transported to regions far from the emissions sources due to atmospheric circulation. Results shown by the CCatt-BraMs model in Map 38 demonstrate the transport of plumes from burning by low level streams. these streams flow predominantly from the northeast to the Southeast and are found at an elevation of about 2000 meters. They originate from the 202 change in the trajectory of the tradewinds when they meet the Andes mountain range and are responsible not only for the transport of plumes due to burning, but also for the humidity generated from the amazon region to the southern and southeastern parts of Brazil. this aspect of transport to distant regions prompted the analysis of results from different parts of the country. The boundaries used in the respective regions were the same as those exhibited in Map 38. region 2 covers a good part of the deforestation area and is responsible for the greater production of fire outbreak, followed by region 1 (re- membering that in Map 38 the fire outbreak are normalized). although there are also fire outbreaks in other parts of Brazil, the impact of meteorological variabilities ap- pears much greater due to the transport of plumes from the amazon region to regions 4, 5 and 6. In the following section, the different impacts on precipitation and tempera- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry ture of the reference and Low-carbon Scenarios of the study will be presented. 4.9.2.1 Precipitation This section discusses the differences in the amounts of precipitation observed between the reference and Low-carbon Scenarios. But before discussing these differ- ences, and as a justification for evaluating the model’s performance, Figure 49 shows the average monthly precipitation simulated in the CCatt-BraMs model from 2007 to 2008 in the reference and Low-carbon scenarios compared to data on precipitation obtained by the national Water agency (agência nacional de �guas - ana), which cor- responds to the climatology realized during the period between 1982 and 2005. the reason that only 2007 and 2008 are included in this analysis is that they are the initial years for the two scenarios, and it is expected that their performance is not very differ- ent from the climatology. In addition, the difference between the two scenarios with regard to emissions caused by burning and land-use change is not very perceptible. Figure 49: Average monthly precipitation in the 6 regions analyzed in the reference and Low-carbon Scenarios from 2007 to 2008 compared to data obtained from the Agência Nacional de �guas (National Water Agency - ANA), which corresponds to monthly precip- itation during the period from 1982 to 2005. The margins of error represent the standard deviation for each month 203 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry The results show that the model was able to coherently accompany the performance of the precipitation observed in the different regions of Brazil, mainly with regard to their seasonal performance. the only exception was region 6, where the model tended to underestimate the precipitation in the middle of the year compared to ANA climatol- ogy, but that didn’t impede the analysis of the impact on the difference in precipitation between the scenarios, nor its tendency to vary through the years. With regard to the difference between the two scenarios, Figure 50 shows the aver- age monthly precipitation observed in the reference and Low-carbon Scenarios during the period between 2007 and 2030 simulated in the CCat-BraMs. Figure 50: Average monthly precipitation in the 6 regions analyzed in the reference and Low-carbon Scenarios from 2007 to 2030 (bar graph left axis). Also shown is the differ- ence between the reference and Low-carbon Scenarios (line graph right axis) 204 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Results suggest that the impact on precipitation between the two scenarios is prin- cipally due to the effect of the aerosols. In the six regions analyzed, the greatest differ- ence between the two scenarios occurs during the months of september and october, during which there are more emissions of particulate matter due to burning. In regions 1 and 2, which are the most affected by these emissions, the impact on the Reference scenario was from -55 to -70 mm on average between 2007 and 2030. Particularly in region 2, the increase in the length of the dry season in the Reference Scenario is very clear. similar to the discussion on the topic in section 4.2 on liquid radiation, the im- pact on precipitation also occurred in other regions, mainly due to the transport of emissions from burning by low level streams, with little impact in region 3, which is located in the northeast (-11 mm), and moderate impact in regions 4, 5 and 6, more to the south, where the model suggested a difference of -21, -23 and -23 mm, respectively, during the month of october. the difference between the two scenarios can be seen in Figure 51, which presents the average spatial and trimestrial distribution during the period from 2007 to 2030. It also shows that the impact is not very significant between February and June, when aerosol emissions from burning are less intense. The peak occurs during the August, september and october trimester, when the impact on the precipitation in the refer- ence Scenario can be -200 mm at some points. It should be noted that the Low-carbon scenario has more precipitation because of the influence of aerosols on cloud micro- physics. environments that are more heavily loaded with particulate matter (in the case of the reference scenario) make it difficult for the cloud droplets to grow, and pre- cipitation tends to be lower in these cases. this effect was parameterized in the CCatt- 205 BraMs model. Figure 51: Difference in precipitation (mm) between the reference and Low-carbon Sce- narios for the years 2007 to 2030 during the February, March and April (A), May, April and June (B), August, September and October (C), and November, December and January (D) trimesters. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 4.9.9.2 Temperature This section discusses the differences observed in the temperatures between the reference and Low-carbon Scenarios. Similar to the section on precipitation, results from the model and climatology amounts were compared. Figure 52 shows the average monthly temperature simulated in the CCatt-BraMs model during the years 2007 and 2008 in the reference and Low-carbon scenarios, compared to the temperature data obtained by the national Institute of Meteorology (InMet) which correspond to 206 the climatology realized from 1977 to 2000. results show that the model was able to coherently accompany the performance of the temperature observed in the different regions of Brazil, underestimating the temperature only in region 5 and overestimating it in region 6, but with the seasonal performance preserved. once again, these facts do not prevent the impact on the difference in temperature between the scenarios, nor its tendency to vary through the years, from being analyzed. Figure 52: Average monthly temperature in the 6 regions analyzed in the reference and Low-carbon Scenarios in 2007 and 2008 compared to data obtained from the National Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Meteorology Institute (Instituto Nacional de Meteorologia - INMET), which corresponds to the monthly climatology of temperature during the period from 1977 to 2000. The margins of error represent the standard deviation for each month When analyzing the results and comparing them to temperatures obtained from the reference and Low-carbon Scenarios, one particular detail stands out. Although in the Reference Scenario there is a lower penetration of radiation from solar radiation due to the greater quantity of particulate matter present in this scenario, resulting in less net radiation, results showed that the temperatures were higher than in the Low- carbon scenario. this occurred due to the modification in the amount emitted via latent 207 and sensible heat by area. In this context, more energy is emitted via sensible heat in the Reference Scenario, compensating for the incident radiative loss, and resulting in higher temperatures. In compensation, the Low-carbon Scenario, even with greater net radiation available, has more energy emitted via latent heat, with no temperature changes in the process. Its temperatures were thus lower. It should be noted that the difference between the amounts of latent and sensible heat between the scenarios was mainly framed by the difference in simulated precipitation between them. Figure 53 shows the average spatial and trimestrial distribution of the tempera- ture difference between the two scenarios from 2007 to 2030. W ith precipitation, the impact is less significant from February to June. the peak occurs during the trimester corresponding to the months of august, september and october, when the temperature in the reference scenario can be almost 3 degrees higher at some points. regions 1, Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2 and 3 show a tendency for the temperature to increase regardless of the scenario in question, while regions 4, 5 and 6 exhibit little tendency in this direction. For example, in region 2, in the reference scenario in 2007, the temperature was 27 degrees. over time this amount increased to 30 degrees in 2030. the increase in temperature through the years in regions 1, 2, and 3 is related to the increase in the sensible heat simulated in these regions. In annual terms, the Reference Scenario is about 1 degree higher in re- gions 1 and 2, and tens of degrees higher in other regions. Figure 53: Difference in temperature (Celsius) between the reference and Low-carbon Scenarios for the years 2007-2030 for the February, March and April (A), April, May and June (B), August, September and October (C), and November, December and January (D) trimesters 208 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 4.9.3 Summary of the Reduction of Impacts on Rainfall and Temper- ature Regimes in the Low-carbon Scenario to summarize, the regions most affected by burning tend to have reduced aver- age annual precipitation during the period from 2007 to 2030 due to the gradual increase in emissions from particulated material during that period. It should be mentioned that this performance occurred independent of the scenario in question although the Reference Scenario presents less precipitation than the Low-carbon scenario. an example of this performance in the reduction of pre- cipitation occurred in the region of the deforestation arc. While it rained about 1800 mm in 2007 in the region, in 2029, this amount was almost 1200 mm in the reference scenario, a reduction of approximately 35 percent. In annual terms, the average percentage difference in precipitation between the reference and Low-carbon scenarios may exceed 30 percent, since this percentage was ap- proximately 15 to 20 percent most of the years between 2007 and 2030. Figure 54 shows a spatialization of the difference in the amount of accumulated rainfall during the august-september-october trimester between the reference and Low-carbon scenarios using the 2007-2030 average. 209 Figure 54: Difference in accumulated rainfall between the reference and Low-carbon Scenarios using the 2007-2030 average. The color scale refers to amounts in millimetres of rainfall per year Technical Synthesis Report | Land Use, Land-Use Change, and Forestry The average temperature tends to increase in regions most affected by burning during the 2007-2030 period due to the gradual rise in the amounts of sensible heat related to the decrease in precipitation during that period. once again, this type of per- formance was independent of the scenario in question. analyzing the region of the de- forestation arc, while its temperature was between 26 and 27 degrees in 2007, in 2030 its average exceeded 30 degrees in the reference scenario. In annual terms, the average difference between the reference and Low-carbon Scenarios was around 1 degree in the regions most affected by burning and by tens of degrees in other regions. Figure 55 shows a spatialization of the difference in temperature in the august- september-october trimester between the reference and Low-carbon scenario, using the 2007-2030 average. Figure 55: Difference between the reference and Low-carbon Scenarios in average air temperature, taken between 2007 and 2030. The color scale reflects the amounts in celsius 210 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry The use of a low-emissions scenario from particulate material and with less conver- sion of green areas into pastures resulted in a 15 to 20 percent increase in average an- nual precipitation between 2007 and 2030 compared to the reference scenario. along similar lines, a one-degree reduction in the temperature per year, due to less sensible heat flow, was observed. 5 analysis of transition Costs from the reference sce- nario to the Low-carbon Scenario An economic analysis of the Low-carbon Scenario helps inform both the govern- 211 ment and society about the economic costs and benefits of choosing the development path with lower carbon emissions. It also helps to understand the conditions under which the proposed mitigation and carbon uptake options could be effectively imple- mented. at the same time, there is no single method for analyzing these options. a va- riety of perspectives can be used to inform a wide range of audiences and agents about the economic conditions under which a Low-carbon Scenario could be implemented. this study conducted a cost–benefit analysis that enabled comparisons to be made between individual options in the Low-carbon Scenario and between the low-carbon and reference-scenario options in general. It should be emphasized that it is not possible to conduct an exhaustive cross-sec- toral economic analysis of externalities. although the key co-benefits of certain miti- gation and carbon uptake options considered under the Low-carbon Scenario could Technical Synthesis Report | Land Use, Land-Use Change, and Forestry be measured in physical terms to study their sustainability, the shear number and diversity of the sectors involved virtually precludes a comprehensive analysis of exter- nalities. Ensuring the homogeneity of the analysis inevitably means limiting it to direct and measurable costs and revenues, thus omitting important co-benefits that may be essential for shaping the decision-making process. Making a joint assessment of the different measures considered is especially chal- lenging, since they are implemented in diverse contexts. some occur within the public economy framework and are implemented by local or federal government, while oth- ers are conducted by the private sector. Some generate revenue, others savings, and still others generate co-benefits and externalities. some are capital-intensive with a timeframe that goes beyond 2030, while others involve short-term changes in opera- tional conditions. the assessment could vary significantly, depending on whether it is from the public or private sector perspective. In order to better inform decision-mak- ers, the study team conducted the cost-benefit analysis using both social and private sector approaches. The social approach provided a basis for making a cross-sectoral comparison of the cost-effectiveness of the 40 mitigation and carbon uptake options considered in the study. a social discount rate was used to calculate the Marginal abatement Costs (MaCs). the MaCs of all proposed mitigation and carbon uptake measures were sorted by increasing value, and plotted along a single graph to facilitate a quick cross-sectoral comparison of their costs and the volume of emissions they could reduce or sequester. this graph, which combines the 40 options from the four sectors mentioned above, is presented in the main part of the low-carbon multisectoral report for Brazil. The private approach assessed the conditions under which the proposed measures could become attractive to economic agents who are deciding whether to invest in low- carbon alternatives in lieu of the more carbon-intensive options found in the Reference Scenario. The private approach adopted in the study estimated the economic incentive that would be needed in order for the proposed mitigation measure to become attrac- tive. If the incentives were provided through the carbon finance market, the private ap- proach would indicate the minimum carbon price, expressed in Us$ per tCo2e, needed to make the low-carbon option attractive enough to be implemented. This does not necessarily mean that the corresponding economic incentive must be in the form of carbon revenue through the sale of carbon credits; capital subsidies for low-carbon technologies or a combination of incentives could be used. Financing conditions and tax credits can sometimes be far more efficient in channeling the corresponding incen- 212 tive to make the low-carbon option the preferred choice of project developers. the “social approach�: Calculating the Marginal abatement Cost Curve Using the social approach, the costs and benefits of the option implemented in the reference scenario during the 2010–30 period were subtracted year by year from the costs and benefits of the proposed low-carbon option implemented during the same period. the 2009 net present value (nPV) of the annual incremental costs and benefits were then calculated to determine the average avoided tCo2e, or MaC, during that pe- riod. the nPV was calculated using a social discount rate of 8 percent. that is the value used in the Pne 2030 for Brazil’s long-term national energy Plan and is generally used for projects financed by the Brazilian development Bank (Bndes). Technical Synthesis Report | Land Use, Land-Use Change, and Forestry In this study, activity-level mitigation measures were analyzed individually. Port- folios of these measures were then developed at the sectoral level to construct a Low- carbon Scenario. The associated potential for each mitigation option was adjusted to ensure internal consistency at the sectoral level to avoid the duplicate calculation of emissions reductions. Since decision-makers may have to choose between alternatives that differ mark- edly in terms of cost-benefit distribution over time, particularly with regard to invest- ment costs, 2009 values were used for calculations and comparisons (Box 4). 213 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry For purposes of comparison, the study also conducted sensitivity analyses for dis- count rates of 4 and 12 percent. The results of calculations of marginal abatement costs using the social approach for mitigation and carbon uptake options for the four sectors covered by the general multisectoral study are presented in table 44 below. the results for options related to land use and land-use change are in boldface. Table 44: Mitigation potential and marginal abatement cost of various alternatives, based on three discount rates 214 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry The “Private approach�: determining the Break-even Carbon Price To assess the feasibility of implementing the mitigation and carbon uptake options from a private sector perspective, the study team calculated the incentives that would be required to make the proposed measures attractive to Brazil’s economic agents. the team applied a two-part method. First, it estimated the minimum internal rate of return (Irr) that Brazil’s economic agents could expect in the subsector where the pro- posed mitigation measure is implemented. Second, it estimated the required minimum incentive as the perceived revenue per avoided tCo2 that would make shifting from the reference option to the low-carbon option attractive; that is, the resulting Irr, includ- ing the incentive, would at least equal the benchmark IRR. Because the risk levels perceived by investors differ according to type of technology, investor strategies may also vary based on the market conditions observed in particu- lar subsectors – and necessary rates of return may differ depending on the technology.40 To establish such a benchmark IRR, the study team consulted both the institutions in Brazil that finance projects in the subsectors considered, as well as important players and entrepreneurs in the field. While issues of confidentiality prevented these institu- 215 tions from disclosing detailed information through this report, the consistency of the data provided gave the project team a sense of the robustness of the estimates thus established. This data was compiled to arrive at a consensus on the rates used and observed in practice, yet these benchmark IRRs remain indicative. At the same time, they differ markedly from the social discount rate used to calculate the MaC and can change from one sector or subsector to another, confirming that the MaC presented in the above sec- tion should not be used as a proxy for the market incentive to be provided at the project level. ghg mitigation projects with Irrs above benchmark Irrs are expected to attract market investors; conversely, those with Irrs below benchmark Irrs will likely re- quire added incentives, such as carbon credits or other mechanisms in order to attract Technical Synthesis Report | Land Use, Land-Use Change, and Forestry private financing. the level of such incentives is seen as the break-even carbon price because it represents the amount of incentive that will equate benefits and costs to achieve the required benchmark Irr. If the break-even carbon price for a ghg miti- gation option is negative, the implementation of such a measure is, for the most part, already attractive, and its Irr is, in most cases, even higher than the sector’s Irr bench- mark and no incentive is needed. however, if the break-even carbon price is positive, the option is not attractive and cannot generate the required benchmark IRR without incentives in the amount of the break-even cost. Interestingly, for certain mitigation options, the value of the Marginal abatement Cost (MaC), which uses the social discount rate of 8 percent, was less than zero; but the break-even carbon price, which uses private-sector discount rates, such as the indica- tive benchmark Irr, was positive (e.g., reduction of deforestation). Corresponding op- tions, which appeared economically attractive under a social approach, are no longer attractive when using a private-sector approach. other mitigation options, already considered expensive when viewed with the social discount rates, would have much higher costs when assessed from the private sector perspective. 40 It is important to note that, in practice, certain proposed mitigation options are components of projects and cannot be separately financed; thus, for these options, the Irrs for overall projects were used. Table 45: Comparison of sector benchmark IRRs and break-even carbon prices for various mitigation options (8 percent social discount rate) Abatement  cost   Carbon  incentive-­� (US$/tCO 2 )  (8%   incremental   Benchmark   Mitigation  option social  discount   approach   IRR  (%) 216 rate) (US$/tCO 2 ) Residential  lighting (120) (243) 15 Steam  recovery  systems (97) (228) 15 Heat  recovery  systems (92) (220) 15 Industrial  lighting (65) (173) 15 Solar  thermal  industrial  energy (55) (123) 15 Combustion  optimization (44) (104) 15 Recycling (35) (91) 15 Furnace  heat  recovery  system (26) (41) 15 Other  energy  efficiency  measures (14) (22) 15 Scalling  up  no  tillage  cropping 0 0 8 Optimizing  traffic (2) 4 15 Reducing  deforestation  +  livestock 0 6 10 Landfill  methane  destruction 3 7 12 Sugarcane  cogeneration (105) 8 18 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Natural  gas  displacing  other  fuels (20) 10 15 Reforestation 39 12 10 Ethanol  displacing  domestic  gasoline (8) 24 15 Investing  in  bike  lanes 1 25 15 Wastewater  treat.  +  methane  destruction  (res.  &  com.) 10 33 12 Gas  to  liquid  (GTL) (2) 34 25 Ethanol  exports  displacing  gasoline  abroad 2 48 15 Eletric  motors (50) 72 15 Existing  refineries  (energy  integration) 7 75 15 Wind (8) 93 10 Renewable  charcoal  displacing  non-­�renew.  charcoal 21 95 15 Investing  in  railroads  and  waterways  vs.  roads 29 97 17 New  refineries 19 106 15 Commercial  lighting (52) 122 15 New  industrial  processes 2 174 15 Existing  refineries  (incrustation  control) 73 209 15 Transmission  line  Brazil-­�Venezuela (31) 216 15 Refrigerators  (MEPS) (41) 223 15 Wastewater  treat.  +  methane  destruction  (ind.) 103 251 12 Investing  in  metro 106 371 17 Existing  refineries  (advanced  controls) 95 431 15 Solar  heater  -­�  residential* 4 698 15 Bullet  train:  São  Paulo  -­�  Rio  De  Janeiro 400 7787 19 *Note:  Positive  MACs  residential  solar  heater  versus  negative  costs  for  industrial  solar-­�thermal  substitution  reflect  the  lower   carbon  content  of  residential  electricity  generation  (mainly  hydropower)  versus  the  higher  carbon  content  of  industrial  thermal   Many of the mitigation options with negative MaCs would also not require incen- energy  generation  (gas,  diesel,  coal).   tives from the private sector perspective (e.g., most energy-conservation options in the industry). these would generate such great economies of energy that implementa- tion, even from a private-sector perspective, would be considered a win-win situation. In such cases, mandatory standards may be an option to harvest such “low-hanging fruits.� obviously not all mitigation options would be tackled solely from a private sector perspective; otherwise, government incentives may be provided for reasons other than ghg emissions reductions. nonetheless, this perspective is valid to demonstrate where incentives might be better placed or most required and where other tools, such as regulation and standards, may be more appropriate than carbon finance. In theory, every measure whose break-even carbon price falls below the market car- bon price would be implemented as a result of the action of market forces. however, as mentioned earlier, the corresponding economic incentive would not necessarily be in the form of carbon revenue through the sale of carbon credits; other incentives, such as financing conditions or tax credits, could be used. an estimate of the total volume of in- 217 centives needed over the study period would amount to Us$445 billion or Us$21billion per year on average. Transport mitigation options would require the greatest amount of average annual incentives at approximately $9 billion, followed by energy at $7 bil- lion, waste at $3 billion and LULUCF at $2.2 billion (table 46). almost all of the mitiga- tion options would require financial incentives, with the exception of energy efficiency measures. Table 46: Volume of incentive required (undiscounted) in order to achieve the emissions reductions considered in the Low-carbon Scenario from 2010 to 2030 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Total Incentive annual Incentive avoided emissions Required (Us$ Required (Us$ (Mt CO2e) MMs) MMs) Energy 1,721 142,892 6,804 Transport 487 185,018 8,810 Waste 1,317 70,256 3,346 LULUCF 7,481 46,769 2,227 Total $ 11,006 $ 444,935 $ 21,187 5.1 Costs of Reducing Emissions from Deforestation the two main emissions mitigation and carbon uptake options identified in this study are: (i) avoiding deforestation, estimated at 9.8 gt Co2e over the 2010–30 period and (ii) carbon uptake through the restoration of legal forest reserves, estimated at about 1.0 gt Co2e during the same period. 218 the sub-sections that follow analyze the costs of transitioning from the LULUCF ref- erence Scenario to a proposed Low-carbon Scenario, to harvest the potential of these two major mitigation and uptake measures.41 two key measures were analyzed in terms of investment and financing needs to quantify the costs involved in avoiding deforestation, which are (Chapters 2&3): • Improving livestock productivity to free up the land necessary for other activi- ties. It is estimated that this measure will lead to a 70 percent reduction in defor- estation, declining from an annual average of 19,500 km² to roughly 4,780 km² per year (a figure slightly below the government target of 5,000 km²). • Preserving forests. This complementary set of measures aims at protecting the Technical Synthesis Report | Land Use, Land-Use Change, and Forestry forest where deforestation is illegal.42 5.1.1 Improving Livestock Productivity There are four categories of livestock production systems: two of lower productivity (degraded and extensive pasture) and two of higher productivity (feedlot and mixed crop-livestock). In the reference scenario, degraded and extensive pasture account for over 90 percent of the land used for livestock activities. In the Low-carbon scenario, these lower productivity systems are gradually replaced by the feedlot and mixed crop-livestock systems, until these higher productivity systems cover approximately 60 percent of the total land required by livestock production in 2030. the increase in beef production in the higher productivity systems would reduce the need for pasture, so the land could be used for other purposes. This would in turn reduce the pressure on the forests, resulting in lower ghg emissions. as discussed in Chapter 3 (table 35), 70.4 million hectares of additional land would be made available: 16.8 million hectares for crops, production forests and pasture expansion in the reference scenario and 53.4 million for new activities for new mitiga- tion and carbon uptake activities in the Low-carbon scenario (44.3 million hectares for the restoration of environmental liability in legal forest reserves, 6.4 million hectares for additional ethanol production, and 2.7 million hectares for production forests). Compared to the lower productivity systems, high productivity systems require significantly more financial resources for investment and expenses, and offer higher returns. In terms of production costs over the 2010-2030 period, recovery of de- 41 More details are found in the LULUCF technical report and in consultants’ reports on related topics. 42 other measures on avoiding deforestation where it is still legal will not be calculated in this analysis. some of the measures currently being discusssed in Brazil as well as internationally include financial incentives, sometimes called payments for environmental services, and are offered to economic agents to compensate for opportunity costs for cancelling deforestation rights. graded pasture through the adoption of the crop-livestock system would require an additional investment of r$2,925 (Us$1,330) per hectare, as well as another r$21,300 (Us$9,682) per hectare to cover expenses. adoption of the feedlot system for cattle during the same period would require r$1,144 (Us$ 520) per hectare of additional in- vestments and r$4,869 (Us$2,213) per hectare for additional expenditures (table 47). 219 Table 47: Investments and expenditures for prototypical livestock systems (2009-30) Production r$ gross per hectare* r$ additional per hectare * System Investment expenditures Total Investment expenditures Total Degraded pasture 2,124 2,594 4,717 - - - extensive pasture 2,775 4,644 7,419 651 2,051 2,702 Feedlot 3,267 7,463 10,730 1,144 4,869 6,013 Crop-Live- 5,049 23,894 28,943 2,925 21,300 24,225 stock Technical Synthesis Report | Land Use, Land-Use Change, and Forestry * Exchange rate R$2.20 = 1US$. Based on the relative prices considered, the higher productivity systems (feedlot and crop/livestock) generate dramatically higher Irrs (7.50 percent and 15.47 per- cent, respectively) than those of low productivity systems (degraded and extensive pasture) (table 48). Table 48: Economic and financial performance of prototypical livestock systems (2009-2030) Degraded pasture (1,857.84) nC** system nPV* (R$ per hectare) IRR (%) extensive pasture (1,128.76) 0.56 Feedlot (95.19) 7.50 Crop-livestock 1,953.46 15.47 * Based on an 8 percent social discount rate. ** NC = non-calculable sufficiently negative value. As a result, the economics of the reference and Low-carbon Scenarios differ mark- edly. the per-hectare cost under the Low-carbon scenario is r$10,600, far higher than that of the reference scenario. over the 2010-30 period, the per-hectare cost differ- ence would amount to r$3,139 on average (table 49). Table 49: Investment and expenses in the reference and Low-carbon Scenarios Total investment expenditure Total investment expenditure Reference 2,688 5,020 7,708 2,688 5,020 7,708 scenario (gross R$ per hectare) (additional R$ per hectare) 220 Low Carbon 2,996 7,849 10,845 308 2,829 3,137 Source: EMBRAPA The economic performance of the livestock sector is far better in the Low-carbon scenario than in the reference scenario. Using an 8 percent social discount rate, the overall nPV of the investment and corresponding cash flows of the reference scenario over the 2010-30 period, were r$18 billion (Us$8 billion). By contrast, the nPV of the Low-carbon scenario results in r$14 billion (Us$6.5 billion). Compared to the refer- ence Scenario, the average IRR for the Low-carbon Scenario for the livestock sector increased from a negative value43 to 11.24 percent (table 50). It is important to note that the nPV and Irr calculated here refer only to new investments made from 2010 onward in both scenarios. neither investments made before that date nor related ex- penses were taken into account. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Table 50: Comparable economic and financial performance of the livestock sector nPV Irr scenario (2010–30) (%) Reference (17,782) nC* (R$ billion) Low Carbon 14,335 11.24 * NC = sufficiently negative non-calculable amount. These differences in economics are accompanied by differences in environmental performance: the Low-carbon scenario for LULUCF does not require additional land for land use, and therefore does not contribute to deforestation and, in turn, its associ- ated ghg emissions. 5.1.2 Forest Protection Although the low carbon land-use scenario offers solutions for bringing the need for additional land virtually to zero, it is expected that complementary forest protection measures would also be required for two major reasons. First, the legal limit for defor- estation (up to 20 percent of properties located in the amazon region) has not yet been reached. thus, where the complex dynamic of deforestation is powered by the financial value of the wood or cleared land (along with with the need for cropland, pasture and production plantations), deforestation would continue. second, there may be a sig- nificant delay between the time demand for cropland, pasture or production forests is 43 the illegal appropriation of land for speculative purposes may explain why an apparently unattractive activity from the economic point of view continues unabated. however, the land title question that the “terra Legal� (Legal Land) program seeks to address was not within the scope of this study. reduced and the time one could effectively observe a behavioral change among defores- tation agents at the frontier. Therefore, the Low-carbon Scenario proposes to implement additional forest protection measures in forested areas where deforestation is illegal. given the many ongoing programs and abundant literature available on this topic, including the Plan of action for the Prevention and Control of deforestation in the Legal amazon (PPCdaM), this study was limited to reviewing existing proposals (Chapter 3). We present here, in 221 order of magnitude, the results of a preliminary analysis of additional costs that could arise from the need for additional forest-protection activities. These aim to ensure that the full potential to reduce deforestation will be achieved via the release of pasture land and livestock productivity gains, as proposed in the Low-carbon Scenario. to assess investment costs and expenditures for managing and enforcing the pro- tection of conservation units where deforestation is illegal, the study used the “Inves- timento Mínimo de Conservação� (Minimum Conservation Investment – MCI [IMC]), developed by the Working group on the Financial sustainability of the national system of Conservation Units (sustentabilidade Financeira do sistema nacional de Unidades de Conservação - snUC), created by the Ministry of environment.44 Using the IMC tool, the study assessed the costs associated with four protection activities over the 2010-30 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry period: (i) protection of indigenous reserves, (ii) protection of conservation units, (iii) control along the road network and (iv) remote sensor monitoring. these activities aim to prevent intrusion into and deforestation of these areas, as well as prohibiting the transport of products resulting from from illegal deforestation. During this period, pro- tection costs would total Us$24 billion, or $1.14 billion per year on average (table 51). 44 the IMC (Minimum Conservation Investment - Investimento Mínimo de Conservação) is a tool that is based on the financial module of the sistema Mínimo de Conservação (Minimum Conservation system – MICosys), developed by d. Vreugdenhill; see d. Vreugdenhill, “MICosys, aplication honduras ‘national Parks Model’�, evaluation record in Ms excel, prepared by PProBaP, CohdeFor/PnUd/World Bank/geF Project (2002). Table 51: Projection of costs for forest protection in areas where deforestation is illegal (in millions of US$) Conservation Units Indigenous Reserves Control of the high- Monitoring Total way network by Remote annual year Invest- expendi- Invest- expendi- Invest- expendi- Cost sensing Cost 222 ment ture ment ture ment ture 2010 516 430 1,680 372 112 93 1 3,205 2011 0 430 43 381 0 93 1 949 2012 0 430 43 391 0 93 1 958 2013 0 430 43 400 0 93 1 968 2014 0 430 43 410 0 93 1 977 2015 0 430 43 419 0 93 1 987 2016 0 430 43 429 0 93 1 996 2017 0 430 43 438 0 93 1 1,006 2018 0 430 43 448 0 93 1 1,015 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 2019 0 430 43 457 0 93 1 1,025 2020 0 430 43 467 0 93 1 1,034 2021 0 430 43 476 0 93 1 1,044 2022 0 430 43 486 0 93 1 1,053 2023 0 430 43 495 0 93 1 1,063 2024 0 430 43 505 0 93 1 1,072 2025 0 430 43 514 0 93 1 1,082 2026 0 430 43 523 0 93 1 1,091 2027 0 430 43 533 0 93 1 1,101 2028 0 430 43 542 0 93 1 1,110 2029 0 430 43 552 0 93 1 1,120 2030 0 430 43 561 0 93 1 1,129 Total 516 9,035 2,539 9,797 112 1,963 21 23,983 It should be emphasized that the mitigation options considered under the Low- carbon Scenario do not include additional measures to prevent deforestation in areas where it is still legally allowed. elaboration and quantification of such proposals were beyond the scope of this study. If such additional measures, like for instance, payments for compensating landowners for forfeiting their rights to deforest, were to be added, additional costs and benefits would have to be integrated into account in analysis, lead- ing most probably to higher marginal abatement costs. Calculating the Marginal abatement Cost from the social Viewpoint three calculations are required to determine the MaC. the first is the year-over- year incremental cost of the Low-carbon Scenario for livestock in relation to the Refer- ence scenario (annual differential between the net results of the two scenarios). next, the incremental costs for each year are calculated in current 2009 values, using a social discount rate of 8 percent. Finally, the weighted average based on the annual emissions reduction volume (from deforestation) is calculated. As previously mentioned, the proportion of higher productivity systems is greater in the Low-carbon Scenario than in the Reference Scenario, which results in a positive nPV of the incremental results of r$14.3 billion, versus $18 billion nPV in the refer- 223 ence scenario. the overall Irr for the Low-carbon scenario is 11.24 percent, and the calculation is based on the incremental costs of the implementation and expansion of the higher productivity (more cost intensive) systems and their related returns. the result of the calculation indicates a marginal negative cost of Us$2.5 per tCo2 avoided. this suggests the adoption of more productive systems, versus the existing predominant extensive and degraded pasture systems, should produce economic gains for the beef sector, in addition to mitigating ghgs. While the projected productivity gains in the Low-carbon Scenario would almost certainly have positive economic out- comes, this initial “social viewpoint� analysis could prove misleading for those keen on learning what the real costs would be to get livestock breeders to adopt more produc- tive systems. In reality, the conclusions differ markedly when perceived from a private Technical Synthesis Report | Land Use, Land-Use Change, and Forestry sector point of view, as shown by the following preliminary results regarding the break- even carbon price (section 7.1.3.a.iv). When the costs of forest protection over the 2010-2030 period are included – Us$24 billion- the MaC goes up to Us$0.48 per Co2 avoided. Calculating the Break-even Carbon Price from the Private sector Viewpoint transitioning from predominantly low productivity systems, specifically feedlots and livestock systems, would require high levels of investments and operations and maintenance disbursements of over Us$ 430 billion over the 2010-30 period, or Us$ 22 billion per year. although the Low-carbon scenario results in an Irr of 11.24 percent, only production systems – especially cattle in feedlots, with an Irr of 7.5 percent – may not be remunerative enough to be implemented on a large scale initially. Thus, in the case of livestock production, it would be especially important to com- plement a social economic analysis (ex: social discount rate) with a private sector anal- ysis. the main justification is that while the social point of view doesn’t vary between the reference scenario and the Low-carbon Scenario, the private sector point of view changes dramatically because Brazil’s livestock sector has limited access to bank credit and depends heavily on its own capital resources for investing in livestock-related tech- nologies. The productivity of more traditional livestock systems, whose returns are often only about 0.5 percent or less, is generally insufficient for defraying the costs of bank credit. Promoting a transition from lower to higher productivity systems could contribute to increasing the rate of return for these businesses. however, adopting high productiv- ity systems presupposes substantially higher investments, which would require access to bank credit. Thus, the rate of return for these endeavors must at least equal the credit costs plus the expected profit to provide livestock breeders an adequate incentive. Therefore, IRRs have to be far higher in the Low-carbon Scenario than in the Reference Scenario. the sum total of the expected rate of return, plus financing costs (i.e., long-term interest rate [LtIr] + percentage of spread ~ 10 percent +) is generally higher than the rates of return that certain productive options recommended for the Low-carbon Sce- nario can achieve (i.e., about 0.56 percent for extensive systems, 7.5 percent for feedlot systems, and 4.5 percent on average for the Low-carbon scenario). 224 the social approach does not explain why high productivity systems need substan- tial incentives, while traditional production systems, which make less profit, tend to expand on their own. What at first glance seems to be a win-win situation – less land needed, and thus less pressure to clear forests and expand the agricultural frontier on one hand, and a better biological and economic performance for the livestock breeder on the other – may not be a very accurate portrayal. In short, the expected Irrs or private discount rates related to livestock breeding in the reference scenario are low (approaching 0.5 percent), while those considered in the Low-carbon scenario are significantly higher (at least 10-12 percent). If bank loans, which benefit from lower interest rates (ex: Banco da amazonia [5-8.5 percent] Technical Synthesis Report | Land Use, Land-Use Change, and Forestry or Bndes [5.75-6.75 percent]) are needed only to finance part of the overall sum re- quired, it may be considered that, under the Low-carbon Scenario, a producer would need to achieve an average IRR of at least 10 percent, which is a rather conservative val- ue. The study used this benchmark IRR to produce an initial estimate of the incentives a Low-carbon Scenario would require to generate substantial productivity gains in the livestock sector resulting in the release of needed pasture land to accommodate grow- ing alternative activities without inducing pressure on forests. It should be emphasized that this study is a first attempt to determine the level of incentives required, but more studies are needed to examine the question in greater depth. To calculate the break-even carbon price, the only incremental costs considered were those associated with the implementation and expansion of higher productivity systems. given that the feedlot system has an Irr of 7.5 percent, which is less than the benchmark Irr used in this study (12 percent), the break-even carbon incentive re- quired was calculated to ensure that this system would reach an IRR equal to the bench- mark rate. the calculation indicates that this incentive should be about Us$ 1.47 per tCo2e, or approximately Us$ 9 billion over the 2010-30 period to avoid 6 gt Co2e and ensure an IRR of 12 percent. When the costs of forest protection during the same period are taken into account– Us $24 billion – the incentive to implement the overall strategy to reduce deforestation by about 80 percent of the historical observed rates rises to Us$ 6 per tCo2e or Us$ 36.5 billion to avoid 6 gt Co2e (Figure 56). Using a higher Irr of 15 percent, the resulting break-even carbon incentives would be $1.88 and $6.64, includ- ing forest protection costs. Figure 56: Marginal abatement cost (8 percent social discount rate) and break-even car- bon price (considering an IRR of 12 percent) for deforestation avoidance measures 225 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Financing requirements To implement the higher productivity, livestock production systems in the low- carbon scenario, the required financing of investments, operations and maintenance would total r$946 billion (Us$430 billion) over the 2010-2030 period, with invest- ments representing approximately 30 percent of total expenditures, or about Us$21.5 billion per year (table 51). a smaller amount would be necessary in the reference scenario, since these higher productivity systems are expected to expand in that sce- nario, albeit at a far more limited scale. releasing another 70.9 million hectares in the Low-carbon scenario would require another r$720 billion (Us$327 billion) more for financing higher productivity systems. this would represent about Us$16 billion in ad- ditional annual costs, equivalent to 72 percent of the gross value of beef production in 2008.45 as a point of reference, financing from the Brazilian government for the sector was Us$3 billion in 2007, or approximately 10 percent of the estimated annual invest- ment required by the reference scenario in 2010 (Us$32.5 billion). Financing requirements would be significantly lower if the Low-carbon scenario were not to incorporate mitigation and carbon uptake measures that require extra land on top of expansion of agricultural land in the reference scenario (legal forest carbon uptake, ethanol for increased national consumption and for export, and production forests for the iron and steel industry). In the reference scenario, the additional land 45 the gross value of meat production in 2008 (based on figures for april 2008 by the IgP-dI) was estimated by the Confederação Brasileira de agricultura e Pecuária (Brazilian Confederation of agriculture and Livestock - Cna) at R$49,59 billion (see rural Indicators Indicadores rurais XI (90 [set.-out.]):6. for agricultural and livestock production is 16.8 million hectares, less than a third of the total volume of land released under the Low-carbon scenario (via high productivity systems of livestock production to accommodate both expansion of crops and all mea- sures considered) (table 52). Without added mitigation and carbon uptake activities, the financing required in the Low-carbon scenario for improved livestock production to release land for crop expansion would total Us$238 billion – Us$108 billion more than in the reference scenario – and Us$262 billion when estimated forest protection 226 costs are added. Table 52: Livestock-sector investments and expenses to release land to absorb additional lands needed in the reference and Low-carbon Scenarios (2010-30) Cleared Cumulative Cumulative Total investment Pasture investments expenses for in feedlots for area in feedlots for feedlots for cattle and crop- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry (millions cattle and crop- cattle and crop- livestock scenario of hectares) livestock livestock (billion R$) Reference 0 92,075 134,351 226,426 (billion R$) (billion R$) Reference (absorption of additional land needed) 16.8* 107,699 356,397 464,095 Low carbon 70.4** 225,322 721,124 946,446 * Additional lands needed for crop, pasture and forest expansion. ** Absorption of additional land needed for the expansion of crops, pasture and forests in the Reference Scenario, plus land needed for proposed mitigation and carbon uptake in the Low-carbon Scenario. 5.2 Forest recovery: Legal Forest reserves Financing measures and available lines of credit for the restoration of the deforested Cost areas of native vegetation on rural properties were listed earlier (item 3.3.2). there are obviously additional needs for funding to achieve the legal scenario, but this is not a critical barrier at the moment, as available lines of credit for forest restoration are cur- rently under-utilized due to other obstacles mentioned below, the main one being the loss of productive area on rural properties. In the legal scenario, forest restoration in legally protected areas means the dis- Implications (winners and losers) placement of agricultural crops and livestock activities that currently make up the type of land use practiced in these areas. Thus, despite the fact that compliance with the legal scenario implies different benefits for the local, regional and global climate, for biodiversity conservation and for the restoration of the quality of environmental ser- vices, such as the hydrologic cycle, greater competition for land for agricultural crops is expected, raising the opportunity cost of land and possibly causing an increase in food prices. Forest restoration costs can be divided into the following components, all of 227 which include a labor cost component: a. Fencing-off. Costs are estimated at r$1,500 to r$ 2,000 per hectare. b. ground preparation. Includes costs for fertilizers, elimination of weeds and sauba ants and digging of appropriate holes for planting saplings; total cost is estimated at r$ 1,000 to r$5,000 per hectare. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry c. Planting. Includes costs for saplings and labor; costs are estimated at r$1,200 to r$ 2,300 per hectare. d. Maintenance of restored areas. Includes regular weeding and periodic and application of fertilizer where needed. these costs could account for as much as 50 percent of total costs. Final per-hectare costs would depend on the extent to which the environment has deteriorated and the levels of intervention necessary to re-establish vegetation. Four levels of intervention correspond to four scenarios (Figure 57), as follows: • Minimum: the area to be restored has great potential for natural regeneration, only requiring fencing-off to permit the re-establishment of the plant cover. • Light: In addition to fencing-off, the area requires planting of tree species used in the forest restoration exercise. • Moderate: the ground is very compacted from years of livestock grazing and is completely colonized by gramineous plants. required interventions include fencing-off, ground preparation, elimination of weeds and ants, and extensive planting of saplings, machinery could be used to contain costs. • Major: In addition to the conditions described above, the ground is extremely degraded and eroded and thus unsuitable for machinery; owing to ecological degradation, such an area would likely continue in a low-carbon state indefi- nitely. Figure 57: Variation in forest restoration costs by intervention scenario 228 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry however, intervention costs may vary considerably, primarily due to the costs of manual labor in rural areas and the purchase of inputs for machinery, whose prices tend to vary, even within the same state. average amounts that appear in Figure 57 above are based on different estimates for forest restoration and date from articles in specialized literature. as it is impossible to geo-spatialize forest restoration costs in the legal scenario, abatement and investment costs were simulated, using a moderate intervention sce- nario. The carbon uptake rate from forest restoration used was the average absorption for the Cerrado and atlantic Forest biomes, consisting of an uptake of 98.3 tCo2/ha in 2030. The incremental cost was not calculated in the legal scenario, only the cost of for- est restoration, as the legal scenario assumes that no economic activity would occur in such areas. the average marginal cost would therefore be Us$ 41.68/tCo2, while the level of incentive expected (break even carbon price) would be Us$ 50.52/tCo2. Considering that the total volume of forest restoration would be 44 million ha, based on the total marginal cost indicated above, the total non-discounted cost would be Us$ 1.84 billion for the period considered, or an average of Us$ 92 million per year. Figure 58: MaC and equilibrium price of carbon for Co2 uptake through legal forest restoration 229 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Considering that the total volume of forest restoration would be 44 million hectares, the total non-discounted cost, based on the abovementioned marginal cost, would equal Us$54 billion over the 2010-2030 period. the average annual cost for financing over the period would be Us$2.7 billion. 5.3 Renewable Charcoal Impacts on land use: necessary area and hypothesis on the correlation with defores- tation Two main aspects were taken into account in this section with regard to the impacts of the low-carbon scenario on land use: (i) availability of agricultural lands and (ii) possible impacts on deforestation practices due to a possible additional need to convert areas with native forests into areas with planted forests. The amount of land needed for implementing the Low-carbon Scenario was estimated to be between 3,327 and 3,663 million hectares, depending on the differ- ent levels of productivity46. this amount represents approximately 0.35 percent of the national territory. Even when the higher amount is added to the additional need for land for the livestock and agriculture sectors, the entire requirement may be met by occupying areas currently used for pastures at different stages, indicating that the land required for supplying the Brazilian iron and steel industry with charcoal does not ne- cessitate the conversion of areas occupied by native forests in production areas47. 230 Marginal abatement cost Considering a discount rate of 8 percent per year, the consolidated marginal abate- ment cost for the use of renewable charcoal instead of coal or non-renewable charcoal was estimated to be approximately Us$ 9.00/ tCo248. When a discount rate of 15 per- cent is applied, the consolidated marginal abatement cost is an average of Us$ 27.14 / tCo2 (table 53).49 Table 53: Marginal abatement cost Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Marginal abatement discount Rate Mitigation Measure Cost (per year) Additional use of renewable char- 8% 8.95 Us$ / tCO2 coal (instead of non-renewable char- coal or coal). 15% 27.141 Source: Adaptation of data presented in the report on industrial sector emissions as emphasized earlier, this mitigation measure requires investments in the forest sector (new planted forests) as well as in the industrial sector (carbonization and ther- mo-reduction processes in blast furnaces). however, the main difference between the costs and investments necessary for the implementation of the Low-carbon Scenario has to do with the establishment, maintenance, and harvesting of additional quantities of planted forests. For the possible substitution of non-renewable charcoal, by definition the marginal abatement cost is at least in proportion with the investment in plantation forests for seven years, since this investment is not made when non-renewable charcoal from deforestation is used. In the industrial sector, additional investments are necessary for the charcoal-making process and for the increase in the number of blast furnaces used to make pig iron with renewable charcoal. this financial input is slightly higher than the investments needed for the expansion of pig iron production using coal50. The main economic results related to this mitigation alternative are presented in table 54. 46 To see details on the calculation methods and the reasons for this variation, please see the land-use section (LULUCF) in the study. 47 Ibid. 48 according to Int estimates, specific calculations for the substitution of non-renewable charcoal and coal are practically the same, respectively: Us$8.9 / tCo2 and Us$ 9.0 / tCo2. 49 given the discount rate of 15% per year, Int estimates that the marginal abatement cost for avoiding the use of non-renewable biomass is Us19.53 / tCo2e and Us$ 34.75 Us$ / tCo2 for coal. however, of the two sub-reference scenarios presented for charcoal, it was decided to adopt an average of Us$27.14/ tCo2 in this report, given the integrated nature of this mitigation alternative, and that isolated figures are dependent on the level of legal restrictions. 50 See detailed estimates in the report on industrial sector emissions. Table 54: Summary of economic parameters for the 2010-2030 period Total Abatement Total In- Internal Avoided Cost (adjust- Number Net Income vestment Rate of Emis- ed Potential) of years of (million (VP) (Us$ Return sions (Us$/tCo2) Investment Us$) million) (%) (million (rate 8% / 231 tons Co2) year) Use of re- newable charcoal 1 21-year instead of 4,245.44 -2,678.28 None 385.07 9.0 cycle coal or non- renewable charcoal The total investment necessary for implementing the Low-carbon Scenario is Source: Adapted from the report on industrial sector emissions Technical Synthesis Report | Land Use, Land-Use Change, and Forestry estimated to be approximately Us$ 4,245 billion (current value), according to the cal- culations presented in the report on emissions from the industrial sector (subject of another summary report). Considering the adjusted potential, this amount represents approximately 12.74 percent of total investments estimated for the implementation of mitigation measures throughout the Brazilian industrial sector, which are estimated to be Us$ 33,331 billion. Table 55: Investments in the additional use of renewable charcoal compared with total mitigation measures in the Brazilian industrial sector, considering the adjusted potential Investment Mitigation Measures % of Investments Sum of mitigation measures from (Us$1,000) the entire Brazilian industrial sec- 33,330,829 100.00 tor 4,245,440 12.74 Additional use of renewable charcoal instead of coal or non- -renewable charcoal Investments necessary to ensure the viability of the additional use of renewable Source: Adaptation of the data presented in the report on industrial sector emissions charcoal for iron and steel production represent over 60 percent of total investments in biomass considered in the different mitigation measures anticipated in this study, which represent approximately 20 percent of all the investments projected, as per table 56. Figure 59: Percentages of investment distribution by group of measures 232 The table below shows the main hypotheses used for the technical-economic analy- Source: Industrial sector emissions report Technical Synthesis Report | Land Use, Land-Use Change, and Forestry sis, developed based on the industrial sector emissions report (the subject of another summary report from this study). Considering that steel and iron production using renewable charcoal is responsable for 90.2 percent51 of the emissions reductions that may be attributed to biomass substitution, the table below was adapted to the amounts presented by the report on emissions from the industrial sector, reflecting this propor- tion. Adding the amounts from investments and costs referring to biomass substitution and the elimination of non-renewable biomass, the following hypotheses were put forth: table 56: hypotheses of the technical-economic analysis Baseline Mitigation Options (amount Present in 106 Us$) (amount Present in 106 Us$) Cost en- Mitgation Invest- ergy / Cost of Cost of Use of Measures ment o&M Income Investment energy o&M Income renewable charcoal instead 4,652 341,333 0 8,897 338,630 1,135 0 of coal or non- renewable charcoal Source: Adaptation of data presented in the report on industrial sector emissions 51 CoPPe, 2009 5.4 Emissions Abatement with Zero Tillage the sociological and environmental significance of the zero tillage system within the context of the national agriculture policy seems not to be fully recognized by the government. the country’s current policy does not penalize farmers who harm the 233 environment, such as in conventional planting where tons of soil are lost due to erosion, silting up rivers and lakes, or even causing soil degradation due to the loss of organic material. From a cultural point of view, the farmer is comfortable not changing his production practices, as he doesn’t have to learn anything new, take risks or make investments. obstacles to the expansion of zero tillage in the country must be surmounted and the following public policies are recommended to achieve this: a. Create incentives for basic research and technology to continuously gener- ate information that ensures the sustainability of zero tillage in different parts of the country. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry b. restructure the rural extension system, training technicians so that they may serve as a link between research institutions, universities and differ- ent segments of the productive sector. c. Establish priority credit, facilitated and differentiated for producers who adopt the zero tillage system; ex. raise the limit for agricultural credit, with lower interest rates, geared towards producers who practice zero tillage; provide agricultural insurance, with the possibility of reducing premiums depending how long it takes to adopt the system. d. expand storage facilities and guarantee the purchase of relevant products for zero tillage, such as corn and rice. e. develop financial instruments that “hedge� the prices of essential inputs for zero tillage (e.g. herbicides). Cost calculation Variables required for estimating the marginal cost of emissions abatement through zero tillage were developed as follows: discount rate. to discount cash flows for agricultural activities, a yield curve is used rather than single discount rate. The idea behind this is that the capital used is made up of a combination of different “zero coupon bonds� with annual maturation from 2008 until 2035. a yield curve was constructed by adding risk-free rates for every year, pro- vided by the ntnB (notas do tesouro nacional - national treasury notes) with a cor- responding due date, like the risk premium, which varies from 5 percent to 4 percent during that time period, multiplied by the β of 1.5 usually used in agribusiness projects in Brazil. the risk premium for the Brazilian economy and the β from the crop-livestock sector were obtained through interviews with specialists from the Banco Nacional do 234 desenvolvimento econômico e social (Brazilian development Bank). exchange rate. over the past few years, the exchange rate between the Brazilian real and the Us dollar has been extremely volatile, which made it difficult to predict its de- velopment. It was thus decided to use market estimates for the dollar for the end of the period, based on the Central Bank’s Focus research, for the next few years until 2013, the last year available. on that basis, the hypothesis was that the dollar would adjust to inflation, which is estimated to be 4.5 percent per year until 2035. Cost of land. the cost of land was estimated based on land prices according to the agricultural economics Institute of são Paulo (Instituto de economia agrícola de são Paulo [Iea]). average land prices in the municipalities of são Paulo are available in reais per hectare and were collated in June every year from 1995 until 2008. they were Technical Synthesis Report | Land Use, Land-Use Change, and Forestry updated by the IgP-dI to the prices of March 2009. the choice of the são Paulo data base distorts the calculation somewhat, given that these are the most expensive lands in the country. they have the best infrastructure and proximity both to the consumer market and the port of santos, Brazil’s largest port, and suffer from pressure from real estate speculation. however, this is still the best option as it is the most reliable data source in Brazil. Its performance can be observed in Figure 60. Figure 60: Cost of land in the state of São Paulo between 1995 and 2008 as can be observed in the above figure, after declining from 1995 to 2000, land prices (in real terms) rose systematically the following years, possibly reflecting the commodities boom and the increase in the liquidity of financial markets. Indications are that land values should drop steadily compared to 2008 prices for an indeterminate period of time, as a reaction to the financial crisis. It should be noted that even the best data available do not enable the construction of a precise model on the evolution of land prices until 2035, as the series is not long 235 enough. It was thus decided to use the historic method of mean values in real terms to estimate future land prices. therefore, the amount of r$12,593 per hectare was used for the year 2008, which is the average for the last available prices registered. the amount of r$10,562 was used for the year 2009, which means the average price was r$10,017, with a 4.5 percent increase due to inflation. From that point on, the price of land was adjusted to inflation which is estimated to be 4.5 percent per year. Price of the main commodities. The calculation of the representative price in the study for commodities (soybean, corn, rice, beans, cotton) must be useful for calculat- ing the total “marginal� cost of carbon captured with the conversion from traditional planting to zero tillage. thus, the increase in the physical quantities of total agricultural production should result in the value of agricultural production. The most obvious way Technical Synthesis Report | Land Use, Land-Use Change, and Forestry to obtain this here is to calculate average prices for the products considered for each year for the participation of each of these products in the total amount produced. For this calculation, estimates of quantities produced were based on studies done by ICone. the prices used and the estimates for the period from 2008 to 2035 were extracted from the data base of prices paid to producers from são Paulo state, which is maintained by the Instituto de economia agrícola de são Paulo (agricultural econom- ics Institute of são Paulo). the first step in estimating prices that are as yet unknown was to consider the re- spective historical series, starting in January 1980. the reason for dispensing with the years prior to 1980 was due to the fact that there was a structural break in the series as a result of the intensification of globalization and the international commodities trade. Limited to the beginning of the series, the next step was the visual inspection of the five graphs below, which show the prices paid to farmers from the state of são Paulo, adjusted by the IgP-dI/FgV in reais of March 2009, per ton of product. Figure 61 shows that the 1980s were more volatile than subsequent years. this vol- atility, which was also observed in international markets, was exacerbated in Brazil by economic instability. as if to illustrate this point, the country witnessed five monetary reforms and great volatility in the exchange rate from 1980 until 1994. starting in the 1990s, commodities prices showed greater stability throughout the world and in Brazil as well, which seems to coincide with a new agricultural price pattern. however, such stability may be threatened by an imminent structural break resulting from a combination of three factors: the increase of food consumption in emerging market countries such as China, India, south africa, russia and Brazil; the emergence of biofuels; and the effects of the global economic crisis. these factors, which are responsible for the increased volatility observed over the last two years, have generated some skepticism as to the validity of balancing supply and demand for fore- casting prices on a distant horizon. With a view to establishing a reasonable forecast for the prices of products, but faced with the aforementioned obstacles, it was decided to use historic price averages, in real terms, for the five commodities, calculated from July 1994 onward. this choice is justified by its simplicity, as there is no guarantee of accuracy using more sophisticated models. the definition of the period of analysis aims to eliminate the effect of hyper- inflation by limiting the beginning of the series to the moment the Brazilian real was 236 introduced, when the country’s economy began to show more stability. Figure 61: Variations in crop prices used in the present study Technical Synthesis Report | Land Use, Land-Use Change, and Forestry The vector of prices of each of the commodities every year for the period considered (2008 until 2035) is composed of historical averages (1994 until 2008) adjusted by the estimated accumulated inflation. Considering that inflation is largely the effect of fiscal policy, and that there is no indication of inflation goals beyond the year 2012, the current inflation target until 2012 (4.4 percent per year) was used hypothetically to estimate future annual inflation until 2035. With these considerations, the price of the 237 physical unit (ton) that is representative for the five commodities was calculated as the average for the agricultural prices considered for the participation of each commodity in the total physical production of the five commodities. outputs for o&M, principal inputs and investments.the data base used to calculate outputs is the estimate of costs per hectare of production, developed by ConaB/MaPa (Cia nacional de abastecimento/Ministério da agricultura, Pecuária e abastecimento – national supply Company/Ministry of agriculture, Livestock and supply). the agricul- tural costs considered in the study are o&M costs (operations and management) for the most important inputs and investment expenditures. as the expenditures of ConaB are not classified according to this nomenclature, the first step was to develop this clas- sification. For calculating costs for o&M and for the main inputs, all available estimates were Technical Synthesis Report | Land Use, Land-Use Change, and Forestry selected for the five crops for the 2008/2009 harvest. estimates of the cost of zero till- age for the other crops were then separated, to be combined under the conventional planting label. The total of the estimates for the two groups, aiming at a cost estimate per hectare for zero tillage and for conventional planting, followed the logic of the cal- culation of average costs for the quantity produced for each commodity compared to the total quantity of commodities produced in each type of planting. For calculating cash flow from agricultural operations, the rigidity hypothesis for the participation of o&M expenses and inputs per crop was used. thus, based on es- timates of quantities produced by each product and planting technique, the expendi- tures were adjusted by the change in composition of the physical quantities produced for all goods in the total production of each product and adjusted according to the infla- tion estimate for each year (4.5 percent per year). outputs for investments made for land were not considered, only for improvements and equipment. Considering that ConaB estimates remuneration for improvements and equipment, or fixed costs, like 6 percent52 of half the price of new equipment, the calculation of expenditures for necessary improvements and equipment for produc- tion is simply the division of the expected remuneration resulting from the fixed capi- tal, divided by 6 percent and multiplied by 2. For calculating expenditures for invest- ments per hectare, it was still necessary to consider expenditures for the investment in each crop for its relative participation every year. Investments can obviously not be made every year with the cash flow, as a substan- tial part of the investment is made just prior to the productive activity. given the lack of information on the investment performance for the commodities considered, it was decided to launch the costs for investment the first year of the series. however, as the period considered in the study is extensive, and taking into account that agricultural equipment has a 10-year period of depreciation, it was decided to make new invest- 52 6 percent is the yield considered for the alternative use of capital. ments every 10 years. For the sake of simplification, the equipment’s residual value was not considered, given the fact that each type of equipment has a differentiated residual value and most are negligeable. another area of simplification adopted for lack of information is that there will be no distinction made between fixed capital, improvement and equipment. Improvements depreciate over 25 years, besides leaving a residual of 25 percent of the initial value. 238 Cost Costs for implementing the zero tillage system for soybean, corn, rice, beans, and cotton were obtained based on data from ConaB/MaPa (Cia. nacional de abastec- imento/Ministério da agricultura, Pecuária e abastecimento) for the 2008/2009 harvest. estimates of Brazilian agricultural production costs per hectare are given for different parts of the country. the agricultural costs considered were for o&M (opera- tional and management costs), for the most important inputs (herbicides, fertilizers, etc.) and for investments (machinery and equipment). expenditures for land acquisi- tion were not considered. Costs of the conventional planting system were thus ob- tained for the development of the Reference Scenario. Estimates for the different cost Technical Synthesis Report | Land Use, Land-Use Change, and Forestry items for each planting system were obtained by calculating the average costs for each commodity, considering the quantity produced by each one, thereby simulating what would be a “joint commodity�. Cost proportionality between the items did not vary throughout the year for each commodity. the total amount that was not discounted from investments from 2010 to 2030, expressed in 2009 reais for the Low-carbon scenario (100 percent of the planted area under zero tillage) is r$335.6 billion, or 70 percent of the total required for the refer- ence scenario (r$473.9 billion) for the five aforementioned crops. It should be consid- ered that this difference is valid for the aggregate commodity, but can be greater or less for each of the five crops in the different regions that contribute to the ConaB data base (table 57). due to the lower investments, operational costs and inputs (8 percent less), re- quired by the Low-carbon Scenario, the marginal abatement cost for emissions is nega- tive (-r$ 0.72/ton C), indicating that the alternative considered in this scenario, zero tillage, is economically superior (table 58). therefore, for the same market condition, the Irr for a scenario for 100 percent of the area under zero tillage is always greater than the IRR obtained for Reference Scenario conditions. these results confirm earlier economic evaluations for planting systems in Brazil, according to which the use of zero tillage is always more advantageous, and its cost is 6 percent lower on average. It goes against common sense that environmentally sustain- able techniques tend to be more costly than those of the market and require additional incentives in order to be adopted. Table 57: Discrimination of costs considered in the study Average Productivity: dIsCrIMInatIon Classification of expense I – on-FarM Costs 1 – operation with planes o&M 239 2 – operation with machines o&M 3 – rental of machines/services o&M 4 – temporary labor o&M 5 – Permanent labor o&M 6 – seeds Inputs 7 – Fertilizer Inputs 8 – Pesticides Inputs 9 – other expenses o&M total on-Farm expenses (a) II – Post-harVest eXPenses Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 1 – Production insurance Capital 2 – technical assistance o&M 3 – outside transport o&M 4 – Improvements o&M 5 – storage o&M total of Post-harvest expenses (B) III – FInanCIaL eXPenses 1 – Interest o&M total Financial expenses (C) VarIaBLe Cost (a+B+C = d) IV – dePreCIatIon 1 – depreciation of improvements/installa- Capital tions 2 – depreciation of tools Capital 3 – depreciation of machines Capital total depreciation (e) V – other FIXed Costs 1 – Periodic maintenance of machines/tools o&M 2 – social duties o&M 3 – Fixed-rate capital security o&M total of other Fixed costs (F) Fixed Cost (e+F = g) oPeratIonaL Cost (d+g = h) VI – InCoMe FaCtors 1 – remuneration expected on fixed capital 2 – Land total Income Factors (I) totaL Cost (h+I = J) Developed by: CONAB/DIGEM/SUINF/GECUP however, the adoption of this planting technique in Brazil has somewhat stagnated and even decreased, possibly due to the perceived risk in changing productive systems, and the limited knowledge on the system’s correct use, among other obstacles that were already discussed. In light of all this, an economic incentive program should be consid- ered a strategic motivating factor for farmers to overcome these obstacles. 240 Table 58: Emissions reduction potential in tons of CO2eq, average abatement cost during the options for the Net Reduction Average abatement Break-even Carbon period and price to be paid per ton of C to compensate the implementation of zero tillage Mitigation or Up- Potential between cost during the pe- Price take of carbon 2010-30 (tCo2e) riod (Us$/tCo2) (Us$/tCo2) discount rate (8%) Zero tillage 355,415,105 - 0.72 -0.20 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 6 Conclusion In moving toward a national Low-carbon scenario, Brazil’s main challenge is un- doubtedly to reduce deforestation. despite the government’s recent success in imple- menting aggressive forest protection policies, deforestation is expected to continue 241 being the country’s largest source of ghg emissions well into the future. Moreover, several recent studies have shown than deforestation means far more than just ghg emissions. deforestation from fires also emits aerosols that affect rainfall and tempera- ture regimes (see section 4.9) and a recent World Bank assessment on the collapse of the amazon rainforest (known as “amazon dieback� in english),53 show that there is a clear interaction between deforestation and the expected damage to the forest from global climate change, with its most severe progression following the same spatial pat- tern as deforestation. For the sake of reducing ghg emissions and halting the acceler- ated dieback of the amazon forest, fires should be eliminated from the amazon region. Brazil has gained considerable experience in forest-protection policies and projects and finding ways to generate economic activities that are compatible with the sustain- ability of native forests. Forest-protection projects and policies are used as barriers to Technical Synthesis Report | Land Use, Land-Use Change, and Forestry counter the progression of pioneer fronts. however, a more drastic reduction in forest destruction needs more than just protection. Shifting to a Low-carbon Scenario would require acting on the primary cause of deforestation: the demand for more land for agriculture and livestock. Therefore, this study proposes a strategy that acts on two complementary fronts: (i) eliminating the structural causes of deforestation and (ii) protecting the forest from attempts to cut it down. Implementing the first part would involve working with stakeholders who use already deforested land, while the second presupposes working with those with a vested interest in new deforestation efforts. With regard to the first issue, eliminating the demand for more land would require accommodating the expansion of agriculture and the meat industry—both of which are important to the Brazilian economy—on already deforested land. this would mean a drastic increase in productivity per hectare. Technically, one option available is to in- crease livestock productivity, thereby giving up large quantities of pasture. This option is technically possible since current average livestock productivity is low and would entail the scaling up of already existing productive systems in Brazil (i.e., feedlots and crop-livestock systems). The potential for releasing and recovering degraded pasture is considerable and is enough to accommodate the most ambitious growth scenario. Moreover, moving from lower- to higher-productivity production systems can trigger a net gain for the sec- tor economy since more intensive processes converge with higher economic returns (Chapter 7). But this option also presupposes four challenging points. First, productive livestock systems are far more capital intensive, both at the in- vestment stage and in terms of working capital. having farmers shift to these systems would require the offer of a large volume of attractive financing far beyond current lending levels. Commercial interest rates are usually too high to make such invest- ments attractive. Moreover, banks are often unwilling to lend to farmers, whom they 53 see “avaliando o risco de Colapso da Floresta amazônica: Uma avaliação do Banco Mundial�, by José a. Marengo, Carlos a. nobre, Walter Vergara, sebastien scholtz, alejandro deeb, Peter Cox, Wolfgang Lucht, hiroki Kondo, Lincoln alves, and Jose Pesquero. perceive as high-risk borrowers. thus, a large volume of financial incentives, along with more flexible lending criteria, would be needed to make such financing viable for both farmers and the banking system. over the past five years, the Brazilian govern- ment has developed programs to stimulate the adoption of more productive systems (e.g., ProLaPeC and ProdUsa) in order to reduce business risk, increase income in the field, and renovate degraded pasture areas. a first attempt to estimate the volume of incentives required indicates an order of magnitude of Us$21.5 billion per year. 242 second, these systems require higher qualifications than traditional extensive farm- ing, in which farmers move on to new areas as soon as the pasture is too degraded to be productive, eventually converting more native vegetation into pasture. Therefore, financing should be accompanied by the intensive development of extension services. Public policies that promote rural extension and training for cattle ranchers would be important to overcome this barrier. third, a rebound effect should be avoided. In other words, higher profitability with less land needed to produce the same volume of meat might trigger an incentive to convert more native forests into pasture. Such a risk is especially high in areas where new roads have been opened or paved. Therefore, any incentive provided should be geographically selective: It should be given only when it has been clearly established, Technical Synthesis Report | Land Use, Land-Use Change, and Forestry based on a valid and geo-referenced property title, that the project will not include the conversion of native vegetation nor areas converted in recent years (e.g., less than 5 years), legally or not. this study confirmed that such a stipulation would be technically possible, since enough pasture can be vacated nationally even without increasing the productivity of livestock in the amazon region. therefore, any subsidized financing for livestock production in the amazon region should be made on an extremely selective and stringent basis, and the area in question should be closely monitored. Fourth, a number of attractive options considered in the Low-carbon scenario to mitigate emissions or increase carbon uptake emphasize the need to liberate a con- siderable amount of pasture. For example, full compliance with the Legal reserve Law would result in the replanting of over 44 million ha currently allocated for other activi- ties. While replanting the forest would remove a large amount of carbon dioxide (Co2) from the atmosphere, this area—more than twice the expected expansion of agricul- tural and pasture land under the reference scenario—would no longer be available for such activities. to avoid a “deforestation leakage� the freeing up of the equivalent ad- ditional amount of pasture would be required; otherwise, part of the production would have to be reduced to prevent the conversion of more native vegetation in another location. the same rationale applies to the expansion of any other activity that requires land (e.g., bioenergy activities involving ethanol or renewable charcoal), though on a far smaller scale. Under the Low-carbon scenario, further expansion of all these activi- ties taken together would require less than one-fourth of the additional land required for legal forest reserves. thus, there is a difficult trade-off between (i) more efforts to increase livestock productivity in order to free up more land and (ii) full enforcement of the recovery of legal reserves and crop expansion. Less compliance with the current legal obligation regarding forest reserves would make it easier to achieve the goal of ac- commodating all activities without any need for deforestation, but it would mean less carbon uptake. The reverse is also true. To protect the forest against the remaining causes of deforestation, it is proposed that forested areas where deforestation is illegal be protected against fraudulent inter- ests. It should be noted that there may be a sizeable gap between the time that demand for land decreases and the time it takes before the behavioral change of proponents of deforestation at the frontier, whether legal or illegal, is observed. Protecting forested areas where deforestation is illegal could be achieved via an ar- ray of activities, ranging from repressive police action to sustainable use projects. In 243 recent years, the Brazilian government has made considerable efforts in this area, par- ticularly under the action Plan for the Prevention and Control of deforestation in the Legal amazon (PPCdaM). Protection measures may include activities similar to those already put into practice under the PPCdaM, such as (i) expansion and consolidation of protected areas, (ii) development of integrated projects, and (iii) promotion of the sus- tainable use of forest resources. Such efforts will need to be maintained and probably amplified. If the proposed strategy is fully implemented—that is, if the demand for additional land is phased out and the forest is protected against the remaining causes of deforesta- tion—then the contribution of Brazil’s LULUCF sector activities could be inverted from high net ghg emissions to a net ghg uptake of about 195 Mt Co2 per year by 2030. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 7annex: analysis of Low-Carbon scenarios To construct a land-use scenario that allows a Low-carbon Scenario, the study gen- erated successive intermediary scenarios that incorporate the impacts of the different mitigation and uptake options considered. Four individual scenarios in particular were analyzed, plus one that combined the options considered in the first four scenarios: 244 herd optimization scenario; scenario to increase forest production plus cattle; high ethanol export scenario plus cattle and production forests; Legal scenario (or recomposition of the Legal reserve) plus cattle; Scenario where the last four scenarios occur simultaneously. In order to focus on the need to convert native vegetation, all of the successive sce- narios were developed so that the total area occupied by agriculture and livestock activities did not increase after the beginning of the period. The largest total area oc- cupied by agriculture and livestock observed between 2006 and 2008 was chosen for Technical Synthesis Report | Land Use, Land-Use Change, and Forestry each region as the limit for the expansion of agrosilvipastoral activities until 2030. only the area in the northern amazon region was larger in 2008, while more area was occupied by agriculture and livestock in other regions in 2006. Improvement of zoo- technical indexes and pasture intensification will be key variables for ensuring that the greater need for land for sugar cane and production forests, and the reduction in the productive area for the legal scenario, do not result in additional deforestation, avoid- ing the trickle-down effect on the agricultural frontier. table 59 shows the four Low- carbon scenarios, which will form the basis for the final scenario that combines the four previous ones. Table 59: Relationship of the Low-carbon Scenarios developed for this study Action taken to Scenarios avoid the domino effect Improvements in zoo- 1. herd optimization 245 technical indicators 2. herd optimization with an increase in production forests Pasture intensifica- elimination of non-renewable charcoal by 2017 and tion participation of 46 percent of renewable charcoal in iron and steel production 3. herd optimization with an increase in production forests, greater ethanol exports and adoption of 2nd Low-carbon Sce- Pasture intensifica- generation ethanol nario: mitigation tion Mixture of 20 percent ethanol in gasoline with Brazil measures Technical Synthesis Report | Land Use, Land-Use Change, and Forestry supplying 15 percent of this market 4. herd optimization with legal scenario (forest res- toration) Pasture intensifica- Restoration of environmental liability of legal forests tion calculated as 44.34 million hectares Pasture intensifica- 5. Combined effect of all these measures tion In the first Low-carbon scenario, called the herd optimization scenario, increases Source: ICONE in the productivity of the beef cattle herd, with improved zootechnical indices (higher birth rate and lower age at time of slaughter), were considered. this scenario has the greatest impact on pasture area, as the model considers the herd a directly proportion- ate variable in determining the size of the pasture area. thus, it is hoped that the pas- ture area will undergo a more rapid intensification process than what was observed in the past and in the Reference Scenario. This is essential for accommodating the greater need for land in the other scenarios. The second scenario considered the greater need for production forests, as well as a smaller herd. The principal behind the production forest scenario is to increase the demand for charcoal for iron and steel production as a substitute for coal and charcoal from native forests. given the great demand for energy for producing pig iron, which is the main raw material used for iron and steel production, the greater contribution of charcoal from production forests would make a tremendous impact on land use. A demand for about a million hectares of production forests was considered for pig iron production in the Reference Scenario. In the second Low-carbon Scenario, this demand will increase to 3.6 million hectares, which represents an additional capture of approxi- mately 500 million tons of Co2. The third scenario, greater ethanol exports and production forests, consists of a set of four exogenous changes in relation to the reference scenario. Besides the improve- ment of zootechnical indices, some of the aspects considered are: greater ethanol exports from Brazil, progressive adoption of the technology for second generation ethanol production, and a greater area allocated for production forests. Thus, like in the other Low-carbon Scenarios, the total area for agriculture and livestock does not 246 change. In the third scenario, besides the production forest scenario described above, it is assumed that ethanol will replace 10 percent of the gasoline consumed worldwide by 2030 and that Brazilian exports will represent 15 percent of global ethanol consump- tion. such assumptions are essential for the expectations of the main consumer coun- tries with regard to gasoline consumption, mandates for ethanol use, productive capac- ity and commercial regimes (Walter et al., 2008). Brazilian ethanol exports, which were at 3.5 billion liters in 2006, are expected to reach 19, 37 and 84 billion liters in 2015, 2020 and 2030, respectively. In the ethanol scenario, internal ethanol consumption does not change in relation to the Reference Scenario. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry The adoption of second generation ethanol technology has a direct impact on the land-use model, as the use of cellulosic matter (principally sugar cane bagasse) for ethanol production reduces the demand for sugar cane for the same amount of ethanol, thus reducing the need for land. Second generation ethanol is thus gradually being ad- opted. In 2010, it is responsible for only 0.4 percent (0.13 billion liters) of all national production but this percentage increases progressively to 2.5 percent (1.3 billion li- ters) in 2015 and 6.1 percent (4.5 billion liters) in 2020, reaching 13.3 percent (17.3 billion liters) in 2030. The fourth scenario considers the gradual reforestation of the Legal reserve (Lr) until its complete restoration in 2030. there are a number of obstacles to calculating the liability of the Legal reserve (Lr) in Brazil, especially because it should be done at the rural establishment level. In addition, the restoration of Permanent Preservation areas (PPa) is required in order to be considered a legal scenario. despite these ob- stacles, since the beginning of the study, there has been some expectation that the Low- carbon scenario, in the case of LULUCF, would have to be based on conditions that are very close to a legality scenario. It was then agreed to call the restoration scenario of legal reserve the“legality� one. although the team concluded that the exact calculation would be discarded, it was decided to make an approximate calculation based on data prepared by the UFMg. a simplified method was developed to calculate the amount of area necessary to be reforested in order to be considered a Legal Reserve, given the limited data avail- able. the area legally defined as an Lr depends on the area of each rural property and biome. since there are no data on the size of the properties, the municipality was used as an approximation. thus, the percentage of Lr was calculated based on the size of the municipality, excluding areas referred to in the UFMg maps as Conservation Units (CU), Indigenous Lands (IL), main watercourses and urban areas. Percentages defined by the Forest Code were used: 80 percent in the amazon biome, 35 percent in the Cerrado in- side the Legal amazon and 20 percent in the other biomes and regions. after estimating the area to be used as an Lr, the area with existing native vegeta- tion, between secondary vegetation, savanna and forests, was eliminated. The result is the area to be reforested to fulfill the legal qualifications as an Lr (table 60). the study considered that the areas that need to be regularized will be reforested gradually, year by year. thus, as of 2009, 1/22 of the total area to be reforested would be deducted from the area available for agricultural production, until it achieves full legality in 2030. 247 Table 60: Area necessary for the reforestation of the Legal Reserve by state in Brazil (hectares) UF Area for Refores- UF Area for Reforestation tation Mato grosso do -3,398,792 Acre -721,161 Sul Mato grosso -9,465,888 amazonas -34,848 goiás -2,611,730 Roraima -46,757 distrito Federal 0 Pará -11,369,199 Maranhão -40,959 amapá 0 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Piauí 0 Tocantins -1,644,537 rio grande do -3,062 Paraná -1,711,257 Norte Paraíba -27,167 santa Catarina -398,679 Pernambuco -58,239 rio grande do sul -1,184,241 Alagoas -91,861 Minas gerais -2,682,095 Sergipe -118,800 espírito santo -205,436 Bahia -242,079 rio de Janeiro -178,087 Rondônia -4,794,589 são Paulo -3,314,927 total Brazil 44,344,389 The last scenario combines the four previous ones. It thus includes a herd with im- Source: UFMG. Developed by: ICONE proved zootechnical indices, greater demand for ethanol and production forests and the recuperation of environmental liability by reforestation. 7. 1 Herd Optimization Scenario the first Low-carbon scenario was developed in partnership with eMBraPa Cer- rados. In this scenario, the herd increases from 206 to 208 million head between 2006 and 2030. thus, there is a larger gain in the herd’s birth rate compared to the reference scenario, going from 0.77 to 0.82 bullocks for each female between 2006 and 2030. this represents a 0.35 percent increase per year between 2009 and 2030. Moreover, the reproduction rate increased in relation to the Reference Scenario, with an increase of 0.80 percent per year between 2009 and 2030, and from 23 percent to 27 percent for the total cattle herd between 2006 and 2030. despite a smaller beef cattle herd, beef production was similar to that of the Reference Scenario, which was necessary for meeting the demand for meat. Between 2006 and 2018, beef production will go from 9.9 to 11.2 million tons, and up to 13.2 in 2030 (table 61). 248 Table 61: Balance of supply and demand for selected products, herd optimization scenario Products Units 2006 2008 2018 2030 Cotton Thousand tons 3,659 5,107 7,133 9,120 Rice Thousand tons 14,344 12,800 15,529 20,611 Bean Thousand tons 3,625 3,936 4,424 5,432 Corn Thousand tons 45,362 61,598 73,663 89,351 Soybean Thousand tons 57,559 63,524 83,230 105,444 Soybean meal Thousand tons 23,684 25,655 30,708 46,097 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Soybean oil Thousand tons 5,984 6,529 7,489 11,425 Sugar Thousand tons 29,767 34,349 44,061 55,852 Ethanol Million litres 18,781 28,482 51,843 75,533 Beef Thousand tons 9,928 9,699 11,222 13,163 Milk Thousand tons 26,153 28,716 38,807 54,071 Chicken Thousand tons 9,354 10,880 12,670 15,737 Eggs Million units 23,575 23,039 25,725 29,312 Pork Thousand tons 2,864 3,102 4,382 5,606 The most important result in this scenario has to do with pasture areas. With the Source: ICONE herd growing less in size, and the hypothesis that the total area will not increase from 2009 onward, the size of the pasture area declined significantly during the period of analysis. Between 2006 and 2018, the pasture area should decrease 10.7 million hect- ares and 18.8 million hectares by 2030, arriving at 190 million hectares. this implies a gain in productivity in terms of number of animals per hectare, which will increase from 0.99 to 1.09 for the entire period, representing an increment of 0.48 percent per year (table 62). Table 62: Land use in Brazil, herd optimization scenario (1000 ha) 2006 2008 2018 2030 Cotton 844.20 1,066.37 1,350.66 1,453.43 Rice 3,017.83 2,880.70 2,910.66 3,228.70 249 Bean – 1st harvest 2,694.21 2,856.81 2,390.40 2,389.68 Bean – 2nd harvest 1,529.39 1,143.11 1,280.97 1,327.74 Corn – 1st harvest 9,632.09 9,656.20 9,693.51 10,412.82 Winter corn 3,331.81 5,052.38 5,373.23 5,638.37 Soybean 22,748.97 21,334.28 25,976.84 30,520.04 sugar Cane 6,179.26 8,234.90 10,579.43 12,631.20 Production Forests 5,269.29 5,886.76 7,740.00 8,450.00 Pastures 208,888.89 205,380.63 198,217.33 190,097.26 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Source: ICONE although there was a significant reduction in pasture area in all regions, it was more pronounced in the northern amazon, comparing the reference and herd optimization scenarios. In the latter, pasture areas in the region decrease 2.7 million hectares be- tween 2006 and 2030, contrary to the reference scenario, where an increase of 12 mil- lion hectares for the same period was observed (table 63) mainly due to the reduction in the size of the cattle herd by 18.2 million head in this scenario compared to the refer- ence Scenario. This was the region that presented the greatest impact on the reduction in the herd compared to the reference scenario, from 68 million down to 49.9 million head as shown in table 64. Table 63: Regional allocation of pasture areas, reference and herd optimization scenarios (1000 ha) Reference herd optimization Scenario Scenario 2006 2008 2030 2030 Brazil 208,889 205,381 207,060 190,097 South 18,146 17,603 13,264 12,606 Southeast 44,053 41,865 39,565 39,678 Central-West Cerrado 51,200 50,636 48,395 47,338 northern amazon 52,551 53,728 64,624 51,879 northeast Coast 10,801 10,487 10,812 10,196 MaPIto and Bahia 32,138 31,061 30,399 28,401 Souce: ICONE Table 64: Regional distribution of cattle herd, Reference Scenario and herd optimization scenario (1000 head) Reference herd optimization Scenario Scenario 2006 2008 2030 2030 Brazil 205,886 201,410 234,460 208,025 250 South 27,200 26,607 27,342 25,673 Southeast 39,209 37,525 36,266 37,548 Central-West Cerrado 56,445 55,506 63,238 58,086 northern amazon 47,391 47,149 68,064 49,901 northeast Coast 8,665 8,156 8,958 8,372 MaPIto and Bahia 26,977 26,468 30,592 28,446 Source: ICONE Technical Synthesis Report | Land Use, Land-Use Change, and Forestry as there were no significant changes in the area allocated for crops, there will also not be any marked change in the distribution of the cattle herd between the six regions of the model. Due to the reduction in average age at time of slaughter, and the need for feed supplements for the animals, an increase in corn production of 517 thousand tons was observed in 2006, reaching 5 million tons in 2030. For the reference scenario, this implies an additional demand for an area of 120 thousand hectares in 2030 for first harvest corn, and an increase of 31 thousand hectares in area for second harvest corn during the same period. this exogenous increment in demand was due to an increase in the price of corn, resulting in a reduction in net exports of 152 thousand tons, as well as a reduction in demand for other uses. Chicken and pork production decreased 319 and 61 thousand tons in 2030, respectively, in relation to the reference scenario. thus, with all factors combined, total corn production increased 808 thousand tons compared to the Reference Scenario. 7.2 Production Forest Scenario As stated earlier, the central hypothesis of this scenario is the increase in the de- mand for charcoal from production forests as a substitute for charcoal from native for- ests and coal for the production of pig iron. Moreover, its starting point is its initial herd of 208 million head, resulting from the herd optimization scenario. during the period analyzed (from 2006 to 2030) the area to be used for production forests will increase 112 percent, going from 5.3 to 11.2 million hectares. In this Low- carbon scenario, the area allocated for production forests in 2030 will be 2.8 million hectares larger than in the reference scenario, which this year is an area of 8.4 million hectares. this difference of 2.8 million hectares between the two scenarios was ac- commodated to a great extent in the pasture areas, which went from 209 to 188 million hectares during the period analyzed, due to increases in productivity. the reduction of 21 million hectares of pasture accommodated, in addition to the expansion of produc- tion forests, the increase in the different crops analyzed, which maintained their occu- pation of the area and their production, compared to the reference scenario (table 65). 251 Table 65: Regional distribution of the production forest in the reference and production forest scenarios (thousand hectares) Reference Reference Scenario Production Scenario Forests Regions 2006 2008 2030 2030 Brazil 5,269 5,887 8,450 11,174 South 1,670 1,886 2,831 2,885 Southeast 2,452 2,690 2,707 4,968 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Central-West Cerrado 319 385 910 992 northern amazon 140 154 327 491 northeast Coast - 9 310 310 Mapito and Bahia 688 762 1,365 1,528 Source: ICONE With regard to the distribution of areas allocated for production forests between the regions of the model in 2030, an increase in the participation of the southeast region from 33 to 44 percent of the area in Brazil was observed, comparing this Low-carbon scenario with the reference scenario. this can be explained principally by the high concentration of the iron and steel industry in this region, and resulted in a decrease in the participation of the southern region, dropping from 31 percent down to 25 percent. 7.3 Ethanol Scenario and Production Forests In this scenario, with its high volumes of ethanol exports, 8 billion liters of fuel were exported, 6.5 times more than the amount observed in the reference scenario. thus, the total demand consisting of exports, domestic demand and final ethanol stocks will reach 147 billion liters in 2030. such an increase in the demand for ethanol means that the need for land for sugar cane (all uses) will exceed 19 million hectares in 2030 throughout the country, or 6.5 million hectares more than in the reference scenario (table 66). an analysis of the aggregate values for Brazil show that the impact of sugar cane expansion will not significantly reduce the area occupied by other crops, as it is accom- panied by the reduction in pasture areas to a great extent. In fact, the area planted with soybean and first harvest corn, the crops that will be most affected by sugar cane ex- pansion, will be reduced less than 0.5 percent during the projected period. Pasture area in the present scenario will be 26.5 million hectares less than in the reference scenario and 9.6 million hectares less than in the herd optimization scenario in 2030 (table 62). sugar cane expansion will thus impose a more efficient use of pasture areas for the herd optimization scenario, due to an approximately 5 percent increase in carrying capacity. 252 It should be emphasized that the adoption of second generation ethanol contributes significantly to reducing pressure that the expansion of ethanol exports exerts on the demand for land. In the productivity pattern projected for 2030, 182 million tons of sugar cane will be needed to produce 17 billion liters of ethanol. assuming productivity of 100 tons of sugar cane per hectare, the production of cellulosic ethanol will reduce the demand for land for cane by approximately 1.8 million hectares54. Table 66: Land use in Brazil, ethanol scenario (in 1000 hectares) herd Ethanol and Reference optimization production forest Scenario Technical Synthesis Report | Land Use, Land-Use Change, and Forestry scenario scenario 2006 2008 2030 2030 2030 Cotton 844 1,066 1,399 1,453 1,454 Rice 3,018 2,881 3,231 3,229 3,242 Bean1 st 2,694 2,857 2,394 2,390 2,414 Bean 2 nd 1,529 1,143 1,328 1,328 1,322 Corn1 st 9,632 9,656 10,292 10,413 10,333 Corn 2 nd 3,332 5,052 5,608 5,638 5,609 Soybean 22,749 21,334 30,601 30,520 30,417 Sugar cane 6,179 8,235 12,700 12,631 19,188 Production forest - 5,887 8,450 8,450 11,174 Pasture 208,889 205,381 207,060 190,097 188,049 Source: ICONE The regional analysis indicates that, like in the Reference Scenario, a good part of sugar cane expansion occurs in the southeast, where the area dedicated to sugar cane will reach 8.1 million hectares in 2020 and 11.1 million hectares in 2030. although the percentage of the total area of the southeast’s participation decreased over time, this reduction will be less acute than in the reference scenario (table 67). 54 A more rigorous evaluation does not allow such an interpretation, as ethanol production and export would probably be different if there is no implementation of second generation ethanol. Table 67: Regional sugar cane distribution in the Reference Scenario, the herd optimi- herd optimization Ethanol zation scenario and the ethanol scenario (in thousand hectares) Reference Scenario Scenario Scenario Regions 2006 2008 2030 2030 2030 Brazil 6,179 8,235 12,700 12,631 19,188 253 South 483 694 1,292 1,297 1,605 Southeast 3,944 5,120 7,056 7,197 11,149 Central-West Cerrado 501 954 1,594 1,369 2,594 northern amazon 113 135 110 111 259 northeast Coast 979 1,150 1,214 1,217 1,435 Mapito and Bahia 160 182 1,435 1,441 2,146 Source: ICONE the greatest variations in the sugar cane area in relation to the herd optimization Technical Synthesis Report | Land Use, Land-Use Change, and Forestry scenario occur in the southeast, Central West Cerrado, MaPIto and Bahia, respectively (table 67). In these regions, a greater response to pasture reduction has been observed both from the pressure caused by cane expansion as well as its potential intensification. It should be emphasized that the scenario with a large volume of ethanol exports causes practically no expansion of the area for sugar cane or any other agricultural op- eration in the northern amazon (measured by the difference in area between the pres- ent scenario and the herd optimization scenario). thus, once technical improvements in the herd optimization scenario were implemented, the decrease in pasture areas as a result of the reduction in the cattle herd was enough to accommodate almost the entire expansion of crops generated in the scenario with the greatest ethanol exports.55 7.4 Legal Scenario (Reforestation of the Legal Reserve) the area necessary for the reforestation of the Legal reserve (Lr) is estimated to be about 44 million hectares. over half is located in the northern amazon region (table 68) principally due to the fact that greatest percentage of Lr needed, or 80 percent of the properties, is in the amazon biome. 55 the scenario analized considers the progressive adoption of second generation technology in the case of sugar cane. Table 68: Reforestation needs in order to comply with the Legal Reserve in the regions of the model (1000 ha) Region area to be reforested by 2030 South 3,294 Southeast 6,381 254 Central-West Cerrado 7,870 northern amazon 24,573 northeast Coast 299 MaPIto and Bahia 1,928 Brazil 44,344 Source: ICONE the area allocated for reforestation between 2009 and 2030 is completely accom- modated by pasture land, which will decrease approximately 60 million hectares dur- ing this time, going from 203.6 in 2009 to 143.9 million hectares in 2030 (table 69). the Technical Synthesis Report | Land Use, Land-Use Change, and Forestry reduction in pasture area is the result of the improvement in zootechnical indices due to the herd optimization scenario, and of pasture intensification due to the expansion of crop areas and forest restoration. table 69 shows that since the need for reforestation will be greater in the northern amazon region in 2030, there will also be a larger reduction in pasture area in this re- gion, decreasing 25.3 million hectares between 2009 and 2030. thus, with the reduc- tion in pasture observed during the period analyzed, it was necessary to move the herd between the different regions of the model, so that the gain in productivity in all regions was similar and compatible with the development observed in the past. Table 69: Pasture area in the regions of the model in 2009 and 2030 (in 1000 ha), in the reforestation scenario of the LR Region 2009 2030 South 17,664.65 9,281.27 Southeast 41,439.97 32,590.04 Central-West Cerrado 50,385.22 38,799.26 northern amazon 52,574.64 27,306.56 northeast Coast 10,569.62 9,896.74 MaPIto and Bahia 30,966.56 25,992.52 Brazil 203,600.67 143,866.39 Source: ICONE In conclusion, the great need for reforestation and its concentration in a few regions will require a considerable reduction in pasture areas, livestock intensification and herd relocation, with a greater need for investments in livestock in this scenario than in the others. It will also imply a change in the geographic distribution of the country’s production-related operations, giving rise to new slaughterhouses and processing plants, logistic system for production distribution and other processes that are part of the production chain. The production of grain and other crops in this scenario was stable compared to the Reference Scenario, made possible due to a decrease in pasture areas to accommodate 255 all the reforestation needs, with no need to reduce other crops. It also meant that it was technically possible to comply with environmental restrictions without impacting agricultural production. however, such an arrangement requires the adoption of new techniques for livestock, the breeder’s adaptation and leads to higher production costs. Cerrado and atlantic Forest A separate analysis was done for the legal scenario in these two biomes, where areas of Legal reserve without plant cover in the Cerrado and atlantic Forest biomes have re- stored the vegetation at a rate of 1/22 per year of the total available until 2030. table 70 presents the results summarized by state. the Co2 uptake potential per hectare ranges between 68.9 and 149.6 tons, as observed in the states of rio grande do Technical Synthesis Report | Land Use, Land-Use Change, and Forestry norte and Paraná, respectively. For each federal unit, the total potential per state was between rio grande do norte (190,532 tCo2) and Mato grosso (261,904,694 tCo2), while the total for Brazil was 1,053,723,278 tCo2, at an average abatement cost of Us$ 40.42. Table 70: Presentation of quantitative results by state for the Atlantic Forest and Cerrado Abatement Break-even total tCo2 Nominal value of State tCo2/ha cost (Us$/ carbon price (2030) investment tCo2) (Us$) Alagoas 4,074,612 108.08 928,586,767 37.91 43.95 Bahia 10,896,586 108.06 2,483,712,919 37.91 43.96 256 espírito santo 7,820,631 91.39 2,107,755,900 44.83 51.97 goiás 96,268,673 88.49 26,796,150,060 46.3 53.68 Maranhão 1,484,927 87.03 420,235,831 47.07 54.58 Mato grosso 261,904,694 92.96 69,393,439,992 44.07 51.10 Mato grosso do 124,773,664 88.13 34,871,346,416 46.49 53.90 Sul Minas gerais 94,409,049 84.50 27,518,095,595 48.48 56.21 Paraíba 787,757 88.88 218,299,926 46.1 53.44 Paraná 99,308,662 149.60 16,350,124,781 27.39 31.75 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Pernambuco 2,048,635 84.44 597,524,371 48.52 56.25 rio de Janeiro 6,534,600 88.09 1,827,161,315 46.51 53.92 rio grande do 190,532 68.90 68,113,435 59.47 68.94 Norte rio grande do 72,712,093 147.40 12,150,222,069 27.8 32.22 Sul Rondônia 19,874,197 102.36 4,782,298,785 40.03 46.40 santa Catarina 22,361,522 134.65 4,090,412,294 30.43 35.28 são Paulo 158,714,178 114.94 34,010,904,452 35.65 41.33 Sergipe 4,672,513 94.42 1,218,874,295 43.39 50.31 Tocantins 64,885,751 98.31 16,256,623,392 41.68 48.32 Total 1,053,723,278 256,089,882,598 40.42 46.87 7.5 Aggregate Scenario: Herd, Production Forests, Ethanol, Forest Restoration the last Low-carbon scenario combines all of the previous ones: herd optimization, large-scale ethanol exports, increase of production forests and legality or reforestation of the Legal reserve. thus, considering a zero increase for the total area, impacts on land-use change appear greater compared to earlier scenarios. this section analyzes impacts on land use, agriculture and livestock production for Brazil and the six regions considered in the model. table 71 shows a comparison of land-use results in the different scenarios analyzed for Brazil for 2006, 2008, and 2030. Compared to the reference scenario, the main impact on land use in the aggregate scenario occurs on pasture lands, which went from 207 to 138 million hectares in 2030, respectively, for each scenario, with a herd of 208 million head (table 73). Compared to a herd optimization scenario, where the herd also has 208 million head, pasture areas amounted to less than 52.3 million hectares in 2030, which can be explained by the greater demand for land for the ethanol, pro- duction forest and legal scenarios. This implies a greater increase in productivity on 257 pastures in order to accommodate such a demand; in other words, the number of head per hectare, or carrying capacity, increased significantly compared to the reference and herd optimization scenarios. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 258 Table 71: Comparison of land use results in all scenarios for Brazil Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Reference Low Carbon herd-Forest-etha- herd with herd herd with Forest herd with ethanol nol-Restoration Restoration Var. Var. Var. Var. Var. Var. 2006 2008 2030 2030- 2030 2030- 2030 2030- 2030 2030- 2030 2030- 2030 2030- 2006 2006 2006 2006 2006 2006 grains (1st 38,937 37,794 47,917 8,980 47,860 8,923 48,005 9,068 48,005 9,068 47,860 8,923 48,005 9,068 harvest) sugar Cane 6,179 8,235 12,700 6,521 19,188 13,009 12,631 6,452 12,631 6,452 19,188 13,009 12,631 6,452 Production 5,269 5,874 8,450 3,181 11,174 5,905 8,450 3,181 11,174 5,905 8,450 3,181 8,450 3,181 Forest Pasture 208,889 205,381 207,060 -1,829 137,820 -71,069 190,097 -18,792 188,049 -20,840 180,521 -28,368 143,866 -65,023 Total Resto- 0 0 0 0 44,344 44,344 0 0 0 0 0 0 44,344 44,344 ration Agriculture and Live- 259,275 257,284 276,127 16,852 216,042 -43,233 259,183 -92 259,859 584 256,019 -3,256 212,952 -46,323 stock Area herd (thou- 205,886 201,410 234,460 28,573 208,002 2,116 208,025 2,139 208,025 2,139 208,099 2,213 208,024 2,138 sand head) Source: ICONE Table 72: Comparison of results for pasture area in all scenarios for Brazil and regions Reference Low Carbon herd-Forest-etha- herd with resto- herd herd with Forest herd with ethanol nol-Restoration ration Var. Var. Var. Var. Var. Var. Regions 2006 2008 2030 2030- 2030 2030- 2030 2030- 2030 2030- 2030 2030- 2030 2030- 2006 2006 2006 2006 2006 2006 Brazil 208,889 205,381 207,060 -1,829 137,820 -71,068 190,097 -18,792 188,049 -20,840 180,521 -28,368 143,866 -65,023 South 18,146 17,603 13,264 -4,881 9,564 -8,581 12,606 -5,540 12,606 -5,540 11,480 -6,666 9,281 -8,864 Southeast 44,053 41,865 39,565 -4,488 27,718 -16,335 39,678 -4,375 38,038 -6,015 36,138 -7,915 32,590 -11,463 Central-West 51,200 50,636 48,395 -2,806 38,285 -12,916 47,338 -3,863 47,256 -3,944 43,525 -7,676 38,799 -12,401 Cerrado Northern 52,551 53,728 64,624 12,074 26,981 -25,569 51,879 -671 51,716 -834 51,718 -833 27,307 -25,244 amazon Northeast 10,801 10,487 10,812 11 9,682 -1,120 10,196 -605 10,196 -605 9,981 -820 9,897 -904 Coast Mapito and 32,138 31,061 30,399 -1,739 25,590 -6,548 28,401 -3,738 28,237 -3,901 27,681 -4,457 25,993 -6,146 Bahia Source: ICONE 259 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry an 11 percent increase in carrying capacity in the herd optimization scenario, and an annual growth rate of 0.46 percent, was observed in Brazil between 2006 and 2030. this increment will already be significantly higher in the last Low-carbon scenario, with an annual growth rate of 2 percent per year and a 53 percent increase in the carry- ing capacity during the same period. one of the main topics to be analyzed in this scenario is the impact on land use in the regions considered in the model. In the northern amazon, southeast and Central-West, 260 pasture areas will be reduced to 25, 12 and 9 million hectares, respectively, during the period analyzed, based on data from table 73, although the reasons for this vary greatly between the regions. table 74 shows the regional land-use results for some select products. Table 73: Results for the cattle herd in the reference, herd optimization and aggregate scenarios (1000 head) herd opti- herd-Forest-etha- Reference mization nol-Restoration 2006 2008 2030 2030 2030 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Brazil 205,886 201,410 234,460 208,025 208,002 South 27,200 26,607 27,342 25,673 27,590 Southeast 39,209 37,525 36,266 37,548 39,944 Central-West 56,445 55,506 63,238 58,086 70,644 Cerrado Northern 47,391 47,149 68,064 49,901 27,951 amazon Northeast 8,665 8,156 8,958 8,372 8,587 Coast MaPIto and 26,977 26,468 30,592 28,446 33,287 Bahia Source: ICONE Table 74: Results for land use and herd for selected products in the aggregate scenario 2006 2018 2030 sugar cane (thousand hectares) South 483.25 1,033.71 1,604.67 Southeast 3,944.35 7,524.12 11,146.98 Central-West 500.59 1,586.56 2,594.07 261 northern amazon 112.63 176.25 259.31 northeast Coast 978.68 1,217.86 1,435.27 MaPIto and Bahia 159.77 921.01 2,145.76 reforestation (thousand hectares) South 1,497.35 3,294.18 Southeast 2,900.25 6,380.55 Central-West 3,577.29 7,870.04 northern amazon 11,169.51 24,572.92 northeast Coast 135.97 299.13 MaPIto and Bahia 876.17 1,927.58 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Pasture (thousand hectares) South 18,145.56 14,26407 9,564.48 Southeast 44,052.98 36,552.75 27,718.43 Central-West 51,200.45 45,065.89 38,284.66 northern amazon 52,550.55 41,681.69 26,981.17 northeast Coast 10,801.06 10,276.62 9,681.55 MaPIto and Bahia 32,138.30 29,306.12 25,590.19 herd (thousand head) South 27,200 25,911 27,590 Southeast 39,209 36,574 39,944 Central-West 56,445 61,847 70,644 northern amazon 47,391 43,174 27,951 northeast Coast 8,665 8,772 8,587 MaPIto and Bahia 26,977 29,550 33,287 Source: ICONE almost all of the reduction in pasture area in the northern amazon is due to the amount of area needed to recompose the deforested areas from Legal reserve, 24.6 million hectares. this also implies a significant reduction in the cattle herd by 22 mil- lion head in 2030 within the framework of the herd optimization scenario, although there will still be a 15 percent gain in the livestock carrying capacity in this region be- tween 2006 and 2030, representing a 1 percent annual growth rate during that period. there will be significant land-use change in the reference scenario, particularly due to the zero growth hypothesis in the total area of the Low-carbon scenarios, as well as the recuperation of the environmental liability of the Legal Reserve. Thus, it may be said that avoided deforestation in this scenario will be 37 million hectares in relation to the Reference Scenario at the end of the period in question. despite the significant reduction in pasture areas, the herd remains stable in the southeast and will increase by 12.6 million head in the Central-West in 2030 as part of the herd optimization scenario. herd expansion in the Central-West is largely due to the significant reduction in the herd in the northern amazon. Productivity gains are 62 percent in the southeast and 67 percent in the Central-West between 2006 and 2030 in the last Low-carbon Scenario. The annual growth rate for carrying capacity during this period for the two regions will be 2.22 percent and 2.20 percent, respectively, reaching 262 1.44 and 1.85 head per hectare in 2030. Livestock intensification in the southeast is due to the expansion of sugar cane (11.2 million hectares in 2030), production forests (5 million hectares) and reforestation (6.4 million hectares). all of these products combined will result in a 16 million hect- are expansion between 2006 and 2030. despite the growth in the sugar cane area, the southeast will reduce some of its contribution to this product in Brazil, going from 68 percent in 2006 to 62 percent in 2030. on the other hand, the Central West and MaPIto region and Bahia will increase their contributions independently by five percentage points during the same period, absorbing almost the entire loss of the Southeast. Compared to the herd optimization scenario, the southeast will also lose participa- tion in sugar cane production. With regard to production forests, the region absorbed Technical Synthesis Report | Land Use, Land-Use Change, and Forestry 43 percent of total Brazilian expansion due to the sizeable concentration of the iron and steel industry in this region. In addition to sugar cane expansion, the livestock intensification observed in the Central-West will principally be the result of reforestation aimed at the recomposition of the Legal reserve, which had a deforested area of 7.9 million hectares. In conclusion, it can be said that the aggregate Low-carbon Scenario implies, besides a zero increase in the total area utilized, a more intense reduction process for pasture areas in order to absorb the expansion of agricultural areas, production forests and re- forestation in the different parts of the country, especially in the southeast and Central West. Figure 62 compares land-use development for grains (first harvest), sugar cane, pastures, production forests and forest restoration (the latter only for the combined scenario) for the reference and aggregate scenarios (herd-ethanol-forests-restoration) from 2006 to 2030. to summarize, it may be observed that, whereas in the reference scenario there is expansion in the total area used for livestock-agriculture, in the aggre- gate Low-carbon Scenario, the total area remains static. Thus, all crops and reforesta- tion expand across pasture areas, necessitating a more sizeable livestock intensifica- tion process. It is worth noting that forest restoration represents 21 percent of the total area used for livestock/agriculture (areas with crops plus pastures). Figure 62: Results of the Reference Scenario (left) and aggregate Low-carbon Scenario (right). 263 Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Source: ICONE 8 references aBraF – anU�rIo estat�stICo da aBraF: ano base 2008 / aBraF – Brasília, 2009 alcides, Felipe rodrigues. Considerações ecológicas sobre Plantios de eucalipto – anais do VIII Congresso de ecologia do Brasil, 23 a 28 de setembro, 2007, Caxambu, 264 Mg. alencar, a.; et al. desmatamento na amazônia: indo além da emergência crônica. Manaus, IPaM, 2004. aliceWeb - sistema de análise das Informações de Comércio exterior via Internet. Available at: http://aliceweb.desenvolvimento.gov.br/. amend, M; C. gascon; J. reid; J.M.C. da silva. Parks Produce Local economic Benefits in amazônia. Conservation strategy Fund report, 2005. disponível em . Last access on 12/05/2009. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry aMs – associacao Mineira de silvicultura. available at: http://www.silviminas.com. br/. aMs – associacao Mineira de silvicultura. Perspectivas e tendências do abastec- imento de madeira para a indústria de base florestal no Brasil – Uma contribuição à construção e companhamento dos cenários futuros. Minas gerais. 2004. disponível em:www.silviminas.com.br amthor, J.s. and M. hustoun. 1998. terrestrial ecosystem responses to global change: a research strategy. ornL/tM-1998/27. oak ridge national Laboratory. andreae, M. o. Biomass burning: Its history, use and distribution and its impact on environmental quality and global climate, global Biomass Burning: atmospheric, Climatic and Biospheric Implications, edited by J. s. Levine, pp. 3-21, MIt Press, Cam- bridge, Mass., 1991. andreae, M. o. Correlation between cloud condensation nuclei concentration and aerosol optical thickness in remote and polluted regions, atmos. Chem. Phys., 9, 543- 556, 2009. andreae, M. o.; d. rosenfeld; P. artaxo; a.a. Costa; g.P. Frank; K.M. Longo; M.a.F. silva-dias. smoking rain clouds over the amazon, 2004. science, 303, 1337-1342. andreae, M. o. and P. Merlet. emission of trace gases and aerosols from biomass burning. global Biogeochemical Cycles, 2001, vol.15, no 4, 955-966 p. artaxo, P.; J.V. Martins; M.a. yamasoe; a.s. Procópio; t.M. Pauliquevis; M.o. andreae; P. guyon; L.V. gatti; a.M.Cordova. Physical and chemical properties of aerosols in the wet and dry season in rondônia, amazonia. Journal of geophysical research, 2002, 107 (d20): 49.1-49.14 assad, e.; Pinto, h. aquecimento global e a nova geografia da produção agrícola no Brasil. eMBraPa/ UnICaMP: Campinas, 2008. atLas solarimétrico do Brasil. recife: editora Universitária da UFPe, 2000. azevedo, t.; tocantins, M.a. Instrumentos econômicos da nova proposta para a gestão de florestas públicas no Brasil. In: Megadiversidade. v.2.n.1-2.2006 BeM 2006, Balanço energético nacional, Ministério de Minas e energia. Berry, e. X., Modification of the warm rain process, Proc. First. natl. Conf. Weather 265 Modification, ed. american Meteorological society, state University of new york, al- bany, pp. 81- 88, 1968. Betiol, L. s. 2008. Potenciais e Limites da responsabilidade Civil como mecanismo econômico de proteção ao meio ambiente. 2008. 258f. dissertação (Mestrado em di- reito Civil) – Pontifícia Universidade Católica de são Paulo, são Paulo. Boddey, r. M.; soares, L.h.B.; alves, B.J.r.; Urquiaga, s.. Bio-ethanol Production in Brazil. In: Pimentel, d. (org.). Biofuels, solar and Wind as renewable energy systems. Berkeley: springer, 2008, v. p. 321-356. Boddey, r.M.; Jantalia, C.P.; Macedo, M.o.; oliveira, o.C.; resende, a.s.; alves, B.J.r.; Uruguiaga, s. Potential of carbon sequestration in soils of the atlantic Forest region of Brazil. In: Carbon sequestration in soils of Latin america. haworth Press, Binghamil- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry ton. 2006. pp. 305-347. Bonezzi, Cristiane Belize; Caldeira-Pires, armando; Brasil Jr., antônio C. P. “avalia- ção do Ciclo de Vida e a Competitividade ambiental da siderurgia no Brasil�. available at: http://www.intempres.pco.cu/Intempres2000-2004/Intempres2004/Sitio/Ponencias/36. doc acessado em 17/06/2009. Bonham-Carter, g. geographic information systems for geoscientists: modelling with gIs, Pergamon, new york, 1994. 414p. Brasil. Banco de desenvolvimento de Minas gerais (BdMg). demonstrativo oficial 2008. disponível em: http://www.bdmg.mg.gov.br/ Brito, José otávio. Carvão vegetal no Brasil: gestões econômicas e ambientais. Estud. av. vol.4 no.9 -são Paulo May/aug. 1990. Brazil. Brazil’s contribution to prevent climate change. Brasília: Ministério das re- lações exteriores; Ministério da Ciência e tecnologia; Ministério do Meio ambiente; Ministério das Minas e energia. Ministério do desenvolvimento, Indústria e Comércio exterior. 2007. Brasil, empresa de Pesquisa energetica – ePe. Plano nacional de energia 20230 - Pne/2030. 408 p. IsBn: 978-85-60025-02-2 rio de Janeiro. 2007. disponível em . acesso em 5 dezembro 2007. Brasil. Ministério do Meio ambiente. Plano de metas do PnF 2004-2007. disponível em: www.mma.gov.br. Brasil, Ministerio do Meio ambiente - MMa. Projeto de Conservação e Utilização sustentável da diversidade Biológica Brasileira – ProBIo. disponível em . acesso em 23 de março de 2008. Brown, s. and a.e. Lugo. 1982. the storage and production of organic matter in trop- ical forests and their role in the global carbon cycle, Biotropica 14:161–187. Brown, s. and a.e. Lugo. 1984. Biomass of tropical forests: a new estimate based on forest volumes, science. 223:1290-1293. Brown, s. and a.e. Lugo. 1992. aboveground biomass estimates for tropical moist 266 forests of the Brazilian amazon, Interciencia. 17:8–18. Brown, s.; a.J.r. gillespie and a.e.Lugo. 1989. Biomass estimation methods for tropical forests with applications to forest inventory data. Forest science. 35:881-902. Brown, s.; L.r.Iverson; a. Prasad and d. Liu. 1993. geographical distributions of car- bon in biomass and soils of tropical asian forests. geocarto International. 4:45–59. CaIt-WrI, World resource Institute. 2007. available at Canalli, B.L. desafios para adoção e permanência em sPd. anais do 8º. encontro de Plantio direto no Cerrado, 28/06 a 01/07 2005, tangará da serra, Mt. editora sanches: tangará da serra, Mt. P. 101-107. 2005. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Castro, Luiz Fernando andrade. Produção de gusa em altos-Fornos a Carvão Vegetal e a Coque. relatório técnico. UFMg/ departamento de engenharia Metalúrgica e de Materiais. Belo horizonte, Julho de 2000. Catharino, e.L.M.; L.C.Bernacci; g.a.d.C. Franco; g. durigan and J.P. Metzger. 2005. aspectos da composição e diversidade do componente arbóreo das florestas da reser- va Florestal do Morro grande, Cotia, sP. Biota neotropica. 6(2). CCaP. 2006. greenhouse gas mitigation in Brazil: scenarios and opportunities through 2025. Center for Clean air Policy – CoPPe. Cerri, C. e. P.; g. sparovek; Martial Bernoux; W.e. easterling; Jerry M. Mellilo; Carlos Clemente Cerri. tropical agriculture and global warming: impacts and mitigation op- tions. scientia agricola, v. 64, p. 83-99, 2007. Coe, M.; M. Costa and B.s. soares-Filho. the influence of historical and potential future deforestation on the stream flow of the amazon river – Land surface processes and atmospheric feedbacks. Journal of hidrology, 2009. Coleman, K.; d.s. Jenkinson; g.J. Crocker; P.r.grace; J.Klir; M. Korschens; P.r.Poulton and d.d.richter. simulating trends in soil organic carbon in long-term experiments us- ing rothC-26.3. geoderma, v. 81, p. 29-44, 1997. Companhia nacional de abastecimento – ConaB. Central de informações agro- pecuárias – safras. disponível em . acesso em 3 de junho de 2008. ConaB. Companhia nacional de abastecimento. www.conab.org.br. ConCaWe. 2004. Well to wheels analysis of future automotive fuels and Pow- ertrains in the european context, January, 2004. 60 pg. Cotton, W. and r. Pielke. human impacts on weather and climate, Cambridge Univer- sity Press, new york, 1996. de Maria I.C. erosão e terraços em plantio direto. Boletim Informativo da sociedade Brasileira de Ciência do solo. Viçosa, volume 24, 1999, p 17-21. derpsch, r. agricultura sustentável. In: saturnino, h.M.; Landers, J.n. (ed.). o meio ambiente e o plantio direto. goiânia : eMBraPa, sPI, 1997. p.29-48. diekow, Jeferson; J. Mielniczuk; heike Knicker; Cimelio Bayer; deborah Pinheiro 267 dick and Ingrid Kogel Knabner. soil C and n stocks as affected by cropping systems and nitrogen fertilization in a southern Brazil acrisol managed under no-tillage for 17 years. soil & tillage research, amsterdam, v. 81, p. 87-95, 2005. dislich, r.; L. Cersósimo and W. Mantovani. 2001. análise da estrutura de fragmen- tos florestais no Planalto Paulistano – sP. revista Brasileira de Botânica. 24(3):321- 332. eBaMM release 1.0, 2005, the erg biofuels analysis meta-model. eC, 2008, european Commission Proposal for a directive of the european Parlia- ment and of the Council on the Promotion of the Use of energy from renewable sourc- es, 2008/0016 (Cod). eMBraPa. 1999. Centro nacional de Pesquisa de solos. sistema Brasileiro de Clas- Technical Synthesis Report | Land Use, Land-Use Change, and Forestry sificação de solos. rio de Janeiro. 412 p. empresa Brasileira de Pesquisa agropecuaria – eMBraPa. sistema de monitora- mento agrometereológico - agritempo. disponível em . acesso em fevereiro de 2008. eMPresa BrasILeIra de PesQUIsa agroPeCU�rIa – eMBraPa. aquecimento global e a nova geografia da produção agrícola no Brasil. 83p.Posigraf. agosto 2008. ePe, 2008, empresa de Pesquisa energética, Plano nacional de energia 2030 Pne 2030. eva, h. d.; a.s Belward; e.e. de Miranda; C.M. di Bella; V. gond; o. huber; s. Jones; M. sgrnzaroli and s.Fritz (2004). a land cover map of south america. global Change Biol- ogy 10, 731-744. Fao 2001. FaoCLIM 2.0 a World-Wide agroclimatic database. Food and agriculture organization of the United nations: roma, Itália. Farr, t. g., et al. 2007. the shuttle radar topography Mission. rev geophys. 45, rg2004, doi:10.1029/2005rg000183. Fearnside, P.M.and r.I Barbosa. soil carbon changes from conversion of forest to pasture in Brazilian amazonia. Forest ecology and Management, amsterdam, v. 108, n. 1-2, p. 147-166, 1998 Federal register, 2009, 40 CFr. Part 80, Part II, Vol. 74, no. 99, May 26, 2009. Ferreira, n. C.; L.g. Ferreira; a.r. huete and M.e. Ferreira (2007b). an operational deforestation mapping system using ModIs data and spatial context analysis. Interna- tional Journal of remote sensing 28, 47-62. Freitas, s. r.; K.M. Longo; M. silva dias; P. silva dias; r. Chatfield; e. Prins; P. artaxo; g. grell and F. recuero. Monitoring the transport of biomass burning emissions in south america. environmental Fluid Mechanics, 2005, 5 (1-2), 135 – 167 p.doi: 10.1007/ s10652-005-0243-7 Freitas, s. r.; K.M. Longo and M. andreae. Impact of including the plume rise of vege- tation fires in numerical simulations of associated atmospheric pollutants. geophysical research. Letters, 2006, 33, L17808, doi:10.1029/2006gL026608. Freitas, s. r.; K. Longo; M. dias; r. Chatfield; P. dias; P. artaxo; M. andreae; g. grell; L. rodrigues; a. Fazenda and J. Panetta. the Coupled aerosol and tracer transport model 268 to the Brazilian developments on the regional atmospheric Modeling system (Catt- BraMs). Part 1: Model description and evaluation. atmospheric Chemistry and Phys- ics. discussion Paper, 2007, 8525-8569 p. Fisher, M.J.; s.P. Braz; r.s.M. santos; s. Urquiaga; B.J.r. alves and r.M. Boddey. anoth- er dimension to grazing systems: soil carbon. tropical grasslands, v. 41, p. 65-83, 2007. Food and agricultural Policy research Institute – FaPrI. FaPrI Models. available at . accessed on January 8 2007. Fundação sos Mata atlântica – sosMa: atlas de remanescentes. 2005. available at . Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Furriela, r. B. 2005 Introdução às Mudanças Climáticas globais. são Paulo. gassen, d. and F. gassen. Plantio direto, o caminho do futuro. Passo Fundo: aldeia sul, 1996, 207p. gomes, B.Z.; F.r. Martins and J.y. tamashiro. 2004. estrutura do cerradão e da tran- sição entre cerradão e floresta paludícola num fragmento da International Paper do Brasil Ltda., em Brotas, sP. revista Brasileira de Botânica. 27(2):249-262. greenland, d.J.; a. Wild and d. adams. organic matter dynamics in soils of the trop- ics-from myth to complex reality. In: Lal, r. and P.a. sanchez. (eds.) Myths and science of soils of the tropics. Madison: sssa spec., 1992. p.17-34. (Pub. 29). greet 1.6 greenhouse gases, regulated emissions and energy use in transportation, Argonne National Laboratory, Argonne, Illinois, USA. grell, g. and d. devenyi. a generalized approach to parameterizing convection combining ensemble and data assimilation techniques. geophysical resource Letters, 2002, 29, 14, 1693. gruberruber, n.; P. Friedlingstein; C.B. Field; r. Valentini; M. heinmann; J.e. richey; P.r. Lankao; e.d. schulze and C.t.a. Chen. the vulnerability of the carbon cycle in the 21st century: an assessment of carbon-climate-human interactions. In: the global Car- bon Cycle (C.B. Field and M.r.raupach, eds). scope, no. 62. Island Press, Washington d.C. 2004, p. 45-76. hahn, C. M. et al. 2004. recuperação florestal: da muda à floresta. são Paulo: sMa. hanaki, M.t.; t. Ito and M. saigusa. effect of no-tillage rice (oryza sativa L.) cultiva- tion on methane emission in three paddy fields of different soil types with rice straw application. Jpn. J. soil sci. Plant nutr. 73: 135-143. 2002. harada, n.; s. otsuka; M. nishimura and s.Matsumoto. Influences of indigenous pho- totrophs on methane emissions from a straw-amended paddy soil. Biol. Fertil. soils 41: 46-51. 2005 hijmans r.; s. Cameron; J. Parra; P. Jones and a. Jarvisc. 2005. avery high resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 25: 1965–1978. holdridge, L. 1967. Life Zone ecology. tropical science Center, san Jose, Costa rica. houghton, r. a.; d.L. skole; C.a. nobre; J.L. hackler; K.t. Lawrence, and W.h. Cho- 269 mentowski. 2000. “annual fluxes of carbon from deforestation and regrowth in the Bra- zilian amazon�. nature 403: 301–304. IBaMa – Instituto Brasileiro do Meio ambiente e dos recursos naturais renováveis. Available at: http://www.ibama.gov.br/ IBge – Instituto Brasileiro de geografia e estatistica. Censo agropecuário: 1995- 1996. rio de Janeiro, 1996. available at: www.ibge.gov.br IBge – Instituto Brasileiro de geografia e estatistica. Censo agropecuário: 2006. rio de Janeiro, 2007. available at: www.ibge.gov.br IBge – Instituto Brasileiro de geografia e estatistica. Produção da extração vegetal e da silvicultura 2003 volume 18. rio de Janeiro. available at: www.ibge.gov.br Technical Synthesis Report | Land Use, Land-Use Change, and Forestry IBge 2006. Mapa de fertilidade de solos do Brasil. 1:5.000.000. IBge. IBge. Instituto Brasileiro de geografia e estatística. www.ibge.com.br. IBS - Instituto Brasileiro de Siderurgia. Available at: http://www.ibs.org.br/index. asp. Iea, 2008, International energy agency, energy technology Perspectives 2008 Iea, 2009, International energy agency, http://www.iea.org IISI - International Iron and Steel Institute. Available at: http://www.worldsteel. org/. Instituto Brasileiro de geografia e estatistica – IBge. Pesquisa agrícola Municipal 2007 - PaM. available at . accessed november 9 2008a. Instituto Brasileiro de geografia e estatistica – IBge. Pesquisa Pecuária Municipal 2007 – PPM. disponível em http://www.sidra.ibge.gov.br/bda/pesquisas/ppm/de- fault.asp. accessed november 8 2008b. Instituto Brasileiro de geografia e estatistica – IBge. Produção da extração Vegetal e da silvicultura 2007. available at . acesso em 9 de setembro de 2008c. Instituto nacional de Pesquisas espaciais – InPe. Projeto Prodes - monitoramento da floresta amazônica brasileira por satélite. available at . accessed october 20 2008. Instituto nacional de Pesquisas espaciais – InPe. relatório de gestão. são José dos Campos, 2009. available at . Last access 05/12/2009. Instituto nacional de Pesquisas espaciais – InPe. (2009). Projeto degrad (Mapea- mento da degradação Florestal na amazônia). available at < http://www.obt.inpe.br/ degrad/>. Last access 05/12/2009. Instituto nacional de Pesquisas espaciais – InPe. (2009). Projeto degrad (Mapea- mento da degradação Florestal na amazônia). available at < http://www.obt.inpe.br/ degrad/>. Last access 05/12/2009. 270 Instituto nacional de Pesquisas espaciais – InPe. (2009). Prodes (Programa de Monitoramento do desmatamento na amazônia). available at < http://www.obt.inpe. br/prodes/> . Last access 05/12/2009. Intergovernmental Panel on Climate Change (IPCC). good practice guidance for land use, land use change and forestry. (2003) available at Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Intergovernmental Panel on Climate Change – IPCC. IPCC special report: Land Use, Land-Use Change and Forestry.(2000). available at: < http://www.ipcc.ch/pdf/spe- cial-reports/spm/srl-en.pdf> accessed 05/11/2009. IsBn:92-3169-114-3. IPCC. International Panel on Climate Change. guidlines for national greenhouse gas Inventories. reference Manual, < http://www.ipcc-nggip.iges.or.jp/public/2006gl>. 2006. IPCC. International Panel on Climate Change. guidlines for national greenhouse gas Inventories. reference Manual, 1996. IPCC, 2001, Climate Change 2001: the scientific basis. Contribution of the Work- ing group I to the third assessment report of the Intergovernmental Panel on Climate Change. Ivaunaskas, n.M.; r.r. rodrigues and a.g. nave. 1999. Fitossociologia de um trecho de floresta estacional semidecidual em Itatinga – sP, Brasil. scientia Florestalis. 56:83- 99. Iverson, L. r.; s. Brown; a. Prasad; h. Mitasova; a.J.e. gillespie and a.e. Lugo. 1994. Use of gIs for estimating Potential and actual Forest Biomass for Continental south and southeast asia. effects of Land-Use Change on atmospheric Co2 Concentrations. springer-Verlag. Jacobson, M. Z. strong radiative heating due to the Mixing state of Black Carbon in atmospheric aerosols. nature, vol. 409, 695-697, 2001. Jank, M.s. 2008. an overview of the Brazilian sugarcane industry. Presentation given at the Better sugarcane Initiative general assembly, são Paulo, sP, november 13, 2008. Jantalia, C; d. resck, d; B. alves; L. Zotarelli; s. Urquiaga and r. Boddey. tillage effect on C stocks of a clayey oxisol under a soybean-based crop rotation in the Brazilian Cer- rado region. soil & tillage research, v. 95, p. 97-109, 2007 Jantalia, C.P.; L. Vilela; B.J.r. alves; r.M. Boddey and s. Urquaia. Influencia de pasta- gens e sistemas de produção de grãos no estoque de carbono e nitrogênio em um Latos- solo Vermelho. seropédica: eMBraPa agrobiologia. 2006a, 50 p. (Boletim de Pesquisa e desenvolvimento, 11). Jantalia, C.P.; L. Zotarelli; h.P. dos santos; e. torres; s. Urquiaga; r.M. Boddey and B.J.r. alves. em Busca da Mitigação da Produção de Óxido nitroso em sistemas agrí- colas: avaliação de Práticas Usadas na Produção de grãos no suldo País. In: Bruno J. r. 271 alves; segundo Urguiaga; Celso aita; robert Michael Boddey; Claudia P. Jantalia; Flávio a. o. Camargo. (org.). Manejo de sistemas agrícolas: Impacto no sequestro de C e nas emissões de gases de efeito estufa. 1 ed. Porto alegre: genesis, 2006b, v. , p. 81-108. Jantalia, C.P.; d.V.s. resck; B.J.r. alves; L. Zotarelli; s. Urquiaga; r.M. Boddey. tillage effect on C stocks of a clayey oxisol under a soybean-based crop rotation in the Brazil- ian Cerrado region. soil & tillage research 95, 97–109. 2007. LægreId, M.; BøCKMan, o.C.; Kaarstad, o. agriculture, fertilizers and the envi- ronment. Wallingford: CaBI, 1999, 320p. José, F. J. a., 2009, comunicação pessoal Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Jones, P.g. and a.gladkov. 2003. FloraMap. a Computer tool for Predicting the distri- bution of Plants and other organisms in the Wild.Version 1.02. Centro Internacional de agricultura tropical: Cali, Colômbia. Juvenal, thais Linhares and rené Luiz grion Mattos. o setor florestal no Brasil e a importância do reflorestamento. rio de Janeiro: Bndes, 2002. disponível em: www. Kaufman, y. J. remote sensing of direct and Indirect aerosol Forcing. aerosol Forc- bndes.gov.br ing of Climate. ed. by r. J. Charlson and J. heintzenberg, John Wiley & sons Ltd., 1995. Koren, I.; y. Kaufman; L.a. remer and J.V.Martins. Measurement of the effect of ama- zon smoke on Inhibition of Cloud Formation, science, 2004, 303, 1342-1345. Lamas, M. P. Manejo cultural do algodoeiro nas condições do Cerrado. In: o agronegócio do algodoeiro no Brasil. (n.e.M. Beltrão & d.M.P. azevedo, eds). Volume 2, 2ª. edição. Brasília, dF, eMBraPa Informação tecnológica, 2008. pp. 621 - 648. Laurance, W.F.; P.M. Fearnside; s.g. Laurance; P. delamonica; t.e. Lovejoy; J.M.r. Merona; J.Q. Chambers and C. gascona. 1999. relationship between soils and amazon forest biomass: a landscape-scale study. Forest ecology and Management 118:127-138 Leal, M.r.L.V, Uso do bagaço e palha para produção de etanol e energia, F.o. Licht ethanol Production Workshop, são Paulo, sP, 23 de março de 2009, Leite, nelson Barbosa. o setor florestal no Brasil. In: a QUestÃo FLorestaL e o desenvolvimento, 8-9 jul. 2003, rio de Janeiro. Bndes, rio de Janeiro, 2003. semi- nário. available at: www.bndes.gov.br. Lima, a.M.n.; I.r. silva; J.C.L. neves, r.F.novais; n.F.Barros; e.s. Mendonca; M.s.M.demolinari and F.P. Leite. Frações da matéria orgânica do solo após três décadas de cultivo de eucalipto no Vale do rio doce-Mg. revista Brasileira de Ciência do solo. vol.32, n.3, pp. 1053-1063. 2008. Lima, M.a. relatório técnico Final do Projeto “emissão de metano em sistemas de produção de arroz irrigado: quantificação e análise�, período 2000 a 2004. eMBraPa Meio-ambiente: Jaguariúna. 2009. Longo, K. M.; s.r. Freitas; M. silva dias; P. silva dias. numerical modeling of the bio- massburning aerosol direct radiative effects on the thermodynamics structure of the atmosphere and convective precipitation. In: International Conference on southern hemisphere Meteorology and oceanography (ICshMo), 8., 2006, Foz do Iguaçu. Pro- 272 ceedings... são José dos Campos: InPe, 2006. p. 283-289. Cd-roM. IsBn 85-17-00023- 4. Longo, K. M.; s.r. Freitas, et al. the Coupled aerosol and tracer transport model to the Brazilian developments on the regional atmospheric Modeling system. Part 2: Model sensitivity to the biomass burning inventories. atmospheric Chemistry and Physics. discussion Paper, 2007, 8571-8595 p. Macedo, I.C., M.r.L.V. Leal and J.e.a.r da silva. 2004, Balanço das emissões de gases do efeito estufa na produção e no uso do etanol no Brasil, relatório submetido à secre- taria estadual do Meio ambiente do estado de são Paulo, abril de 2004, 37 pg. Macedo, I.C., J.e.a. seabra and J.e.a.r. da silva. 2008, greenhouse gas emissions in Technical Synthesis Report | Land Use, Land-Use Change, and Forestry the production and use of ethanol from sugarcane in Brazil: the 2005/2006 averages and the prediction for 2020, Biomass & Bioenergy, 2008. Macedo, M.C.M. 1995. Pastagens no ecossistema cerrados. In: sIMPÓsIo soBre Pastagens nos eCossIsteMas BrasILeIros: pesquisas para o desenvolvimento sustentável, 1995, Brasília. anais. Brasília: sBZ, p.28-62. Macedo, M.C.M., and a.h.Zimmer. sistema de Pasto-Lavoura e seus efeitos na produ- tividade agropecuária. In: sIMPÓsIo soBre eCossIsteMa de Pastagens. 1993. Jaboticabal,sP. anais. Jaboticabal,1993.p.217-245. Machado, r. B., M.B. ramos neto; P.g.P. Pereira; e.F. Caldas, d.a. gonçalves; n.s. santos; K. tabor and M. steininger (2004). estimativas de perda da área do Cerrado brasileiro. [technical report]. Brasília-dF, Brasil: Conservation International. Maddock, J.; M. dos santos, M. and K. Prata. nitrous oxide emission from soil of the Mata atlantica, rio de Janeiro state, Brazil, J. geophys. res., 106(d19), 23055-23060. 2001. Mattoso, M.J.; J.C. Cruz and I.a. Pereira Filho. Custo de produção em plantio direto. Informe agropecuário, v. 22, p. 109-116. 2001. Malhi, y.J., et al. Climate Change deforestation and the faith of the amazon. science, 319: 169-172, 2008. MaPa – Ministério da agricultura, Pecuária e abastecimento. disponível em: http:// Marengo, J.a., et al. the drought of amazonia in 2005. J. Clim., 21, 495-516, 2008. www.agricultura.gov.br Margulis, s. Causas do desmatamento na amazônia brasileira. 1ªed. Brasília: Banco Mundial, 2003. Martins, o.s. 2004. determinação do potencial de remoção de carbono na recupera- ção de matas ciliares na região de são Carlos, sP. tese de doutorado. Programa de Pós- graduação em ecologia e recursos naturais. Universidade Federal de são Carlos. 136 p. Masera, o.r. et al. 2003. Modeling carbon sequestration in afforestation, agrofor- estry and forest management projects: the Co2FIX V.2 approach. ecological Modelling 164:177–199. McKinsey and Company, 2009, Pathway to a low-carbon economy for Brazil. report. 273 46 p. Ministério de Ciência e tecnologia - MCt. Primeiro inventário brasileiro de emissões antrópicas de gee: emissões e remoções de dióxido de carbono por conversão de flo- restas e abandono de terras cultivadas. Brasília, 2004. Ministério do Meio ambiente - MMa. Plana amazônia sustentável. Brasília, 2008. Miranda, e. e. (Coord.). 2005 . Brasil em relevo. Campinas: eMBraPa Monitora- mento por Satélite, Moraes, r.J.; MM.g.L. azevedo; F.M.a. delmanto. 2005. as leis federais mais impor- tantes de proteção ao meio ambiente comentadas. rio de Janeiro: renovar. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry nepstad, d. et al. 2007. Manejo e recuperação de mata ciliar em regiões florestais da amazônia. série Boas Práticas. Livro 1. Instituto de Pesquisa ambiental da amazônia. Mosier, a. r.; e. Wassmann; L. Verchot; J. King and C. Palm. Methane and nitrogen oxide fluxes in tropical agricultural soils:sources, sinks and mechanisms. environment, development and sustainability, v. 6, p. 11-49, 2004. nakicenovic, n.; a. grübler and a. Mcdonald, 1998, global energy perspectives, Cambridge University Press, Cambridge, UK, 1998. nepstad, d. et al. the Costs and Benefits of reducing Carbon emissions from defor- estation and Forest degradation in the Brazilian amazon. Woods hole: WhrC, 2007. v. 1. 26 p. nepstad, d. et al. globalization of the amazon soy and Beef Industries: opportuni- ties for Conservation. In: Conservation Biology, v. 6 n. 20, p. 1595-1603, 2006. nIPe, 2005, núcleo Interdisciplinar de Planejamento energético da UnICaMP, es- tudo sobre as possibilidades e impactos da produção de grandes quantidades de etanol visando a substituição parcial de gasolina no mundo – Fase 1, relatório para o Centro de gestão e estudos estratégicos (Cgee), Coordenado pelo dr. rogério Cezar de Cer- queira Leite, dezembro de 2005. nIPe, 2007, núcleo Interdisciplinar de Planejamento energético da UnICaMP, es- tudo sobre as possibilidades e impactos da produção de grandes quantidades de etanol visando a substituição parcial de gasolina no mundo – Fase 2, relatório para o Centro de gestão e estudos estratégicos (Cgee), Coordenado pelo dr. rogério Cezar de Cer- queira Leite, Março de 2007. nobre, C. a., P. J. sellers and J. shukla. 1991. amazonian deforestation and regional climate change. J. Climate 4, 957-988. nobre, et al. amazônia: Fonte ou sumidouro de Carbono? Causas e dinâmica do des- matamento na amazônia. Brasília: Ministério do Meio ambiente, 2001, 436 p. NREL, 2002, Lignocellulosic Biomass to Ethanol Process Design and Economics Utilizing Co-Current dilute acis Prehydrolysis and ezymatic hydrolysis for Corn sto- ver, report nreL/tP-510-32438, prepared by adem, a., ruth, M., Ibsen, K., Jechura, J., neeves, sheeran, J. and Wallace from nreL and Montague, L. slayton, a. and Lukas, J.from harris group. 274 nucleo Interdisciplinar de Planejamento energetico nIPe; Centro de gestao e es- tudos estrategicos - Cgee. estudo sobre as possibilidades e impactos da produção de grandes quantidades de etanol visando à substituição parcial de gasolina no mundo - relatório final. 320p. dezembro 2005. olivério, J.L. 2007. as Usinas de açúcar e �lcool – estado da arte da tecnologia In- dustrial, Congresso Internacional de agroenergia e Biocombustíveis, teresina, PI, 13 d3 Julho de 2007. Pimentel, d. CrC handbook of energy utilization in agriculture. Boca raton: CrC Press. 1980, 475p. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Plano anual de outorga Florestal – PaoF 2009. Brasília: MMa/ serviço Florestal Brasileiro, 2008.111f. Presidencia da republica. decreto nº 6.527 de 1º de agosto de 2008. available at . Last access on 05/09/2009. PnMC - Plano nacional de Mudanças sobre o Clima (versão para consulta pública). available at: . Last access on 05/13/2009. Rochette, P. No-till only increases N2o emissions in poorly-aerated soils. soil & till- age research, v. 101, p. 97–100, 2008. rodrigues, J.e.r. 2005. sistema nacional de Unidades de Conservação. são Paulo: Ed. Revista dos Tribunais. rodrigues, et al. dinamica ego, uma plataforma para modelagem de sistemas am- bientais. In: simpósio Brasileiro de sensoriamento remoto, 13. Instituto nacional de Pesquisas espaciais. 2007. saatchi, s.s.; r.a. houghton; r.C.s. alvalá; J.V. soares and y. yu 2007. distribution of aboveground Live Biomass in the amazon Basin. global Change Biology. accepted ar- ticle online: 24-Jan-2007 saatchi, s.s; r.a. houghton; r.C. dos santos; J.V. soares; y. yu. 2007. distribution of aboveground live biomass in the amazon Basin. global Change Biology, 13, 816–837. Sampaio, g.; C. Nobre; M. Costa; P. Satyamurty; B.s. soares-Filho; M. Cardoso. 2007. regional climate change over eastern amazonai caused by pasture and soybean crop- land expansion. geophysical research Letters 34. doi:10.102. sanchez, P.a. Properties and management of soils in the tropics. new york: John Wi- ley & sons, 1976. 618p. saturnino, h. M.; J.n. Landers. o Meio ambiente e o Plantio direto. goiânia: aPdC, 1997. 116 p. seMad – secretaria de estado de Meio ambiente e desenvolvimento sustentável. disponível em: http://www.semad.mg.gov.br/ silva, r.r., et al. regional impacts of future land-cover changes on the amazon Basin 275 wet season climate. J. Clim., 21 (6), 1153-1170. sIndIFer - sindicato das Indústrias de Ferro de Minas gerais. disponível em: http://www.sindifer.com.br/inst.html. sisti, C. P. J.; h.P. dos santos; B.J.r. alves; s. Urquiaga and r.M Boddey. Change in car- bon and nitrogen stocks in soil under 13 years of conventional or zero tillage in south- ern Brazil. Biology and Fertility of soils, amsterdam, v. 76, p. 39-58, 2004. smith, K.a. and F. Conen. Impacts of land management of fluxes of trace greenhouse gases. soil Use and Management, 20:255-263. 2004. smith, P.; d. Martino; Z. Cai; d. gwary; h. Janzen; P. Kumar; B. McCarl; s. ogle. F. o’Mara; C. rice; B. scholes and o. sirotenko. 2007: agriculture. In Climate Change 2007: Mitigation. Contribution of Working group III to the Fourth assessment report of the Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Intergovernmental Panel on Climate Change [B. Metz, o.r. davidson, P.r. Bosch, r. dave, L.a. Meyer (eds)], Cambridge University Press, Cambridge, United Kingdom and new york, ny, Usa. soares, J.J.; d.W. silva and M.I.s. Lima. 2003. Current state and projection of the prob- able original vegetation of the são Carlos region of são Paulo state, Brazil. Brazilian Journal of Biology. 63(3):527-536. soares, L.h.B.; B.J.r. alves; s. Urquiaga; r.M. Boddey. Mitigação dos gases de efeito estufa pelo uso de etanol de cana-de-açúcar produzido no Brasil. eMBraPa agrobiolo- gia. Circulat técnica no. 27. seropedica, rJ. 14 p. 2009. soares-Filho, et al. Cenários de desmatamento para a amazônia. estudos avança- dos, são Paulo, v.19, n.54, p.138-152, 2005. soares-Filho, et al. nexos entre as dimensões socioeconômicas e o desmatamento: a caminho de um modelo integrado. In: Batistela, M. et al (org.): amazônia. natureza e sociedade em transformação. edusp. são Paulo, 2008. soares-Filho, et al. redução das emissões de carbono do desmatamento no Brasil: o papel do programa �reas Protegidas da amazônia (arPa). (em revisão), 2008b. soares-Filho, et al. reduction of carbon emissions associated with deforestation in brazil: the role of the amazon region protected areas program (arPa). Brasília: IPaM, 2008. v. 1. 32 p. soares-Filho, et al. (2009). Projeto dinamica ego. disponível em último acesso em 13/05/2009. sousa, e.L., 2008, a importância do etanol e da cogeração na atual matriz energética brasileira, apresentação feita em reunião do tribunal de Contas da União, Brasília, df, 12 de novembro de 2008. teixeira, a. M. et al. Modeling Landscape dynamics in the atlantic rainforest do- main: Implications for conservation. Forest ecology and Management 257, 1219– 1230, 2009. Unh, 2007. r-hydronet: a regional, electronic hydrometeorological data net- work for south america, Central america, and the Caribbean. Water systems analysis group, Universidade de new hampshire. Unruh, g. C. Understanding carbon lock-in. energy Policy, 28 (12): 817-830, 2000. 276 Urquiaga, s.; B.J.r. alves; r.M. Boddey. Produção de biocombustíveis - a questão do balanço energético. revista de Política agrícola, Brasília, dF, v. anoXIV, n. 5, p. 42-46, 2005. Usdoe, 2005, Us department of energy, Clean Cities Initiative Fact sheet, april 2005, 2 pg. Valverde, sebastião renato. Plantações de eucalipto no Brasil. revista da Madeira, no 07, 18 de setembro de 2007. van den Wall Bake, J.d. 2006. Cane as key in Brazilian ethanol industry, Master the- sis, Utrecht University, 2006. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry Veloso, h.P.; a.L.r. rangel Filho and J.C.a. Lima. 1991. Classificação da vegetação brasileira adaptada a um sistema universal. são Paulo: Fundação Instituto Brasileiro de geografia e estatística – IBge. 123 p. Vose, r.s., r.L. schmoyer; P.M. steurer; t.C. Peterson; r. heim; t.r. Karl and J.K. eis- cheid. 1992. the global historical Climatology network: Long-term monthly tempera- ture, precipitation, sea level pressure, and station pressure data. oak ridge national Laboratory, Carbon dioxide Information analysis Center. Walko, r.; L. Band; J. Baron; F. Kittel; r. Lammers; t. Lee; d. ojima; r. Pielke; C. taylor; C.tague; C.tremback and P.Vidale. Coupled atmosphere-Biophysics-hydrology Models for environmental Modeling. Journal applied Meteorology, 2000. 39: (6) 931-944 p. Walter, arnaldo; Frank rosillo-Calle; Paulo dolian; eric Piacente and Kamyla Borges Cunha. Perspectives on fuel ethanol consumption and trade. Biomass and bioenegy, 23 (2008) p.730-748. available at www.elsevier.com/locate/biombioe. accessed on June 8, 2008. Walter, a.s.; F. rosillo-Calle; P. dolzan; e. Piacente and K.B. da Cunha. 2008, Perspec- tives on ethanol consumption and trade, Biomass & Bioenergy, 2008. Walter, a.s.; P. dolzan; o. Quilodrán; J. garcia; C. da silva, F. Piacente and a. segerstedt. 2008, a sustainability analysis of the Brazilian ethanol, report to the UK Embassy in Brasilia. Wassmann, r.; r.s. Lantin; h.U. neue; L.V. Buendia; t.M. Corton and y. Lu. Charac- terization of Methane emissions from rice Fields in asia. III. Mitigation options and Future research needs nutrient Cycling in agroecosystems 58: 23–36, 2000. Weck, J. 1970. an improved CVP-index for the delimitation of the potential produc- tivity zones of forest lands of India, Indian Forester. 96:565-572. Whitmore, t.C. 1984. tropical rain Forests of the Far east, 2nd edition, oxford sci- ence Publications, Clarendon Press, oxford. WMo 1996. Climatological normals (Clino) for the Period 1961-1990. Xavier, R.P. Contribuição da fixação biológica de nitrogênio na produção sustentável de cana-de-açúcar. tese de doutorado, Programa de Pós-graduação em Fitotecnia, Uni- 277 versidade Federal rural do rio de Janeiro. seropédica: UFrrJ, 114 pp., 2006. Xiang P; h. huang; M. huang; d.X. gan; y. Zhou and Z. Fu. studies on technique of reducing Methane emission in a rice-duck ecological system and the evaluation of Its economic significance agricultural sciences in China, 5: 758-766 2006. Zinn, y.L.; r. Lal; d.V.s. resck. Changes in soil organic carbon stocks under agricul- ture in Brazil. soil & tillage research, v. 84, p. 28-40, 2005. Legislation: BrasIL. Presidência da república. Lei nº 4771, 15 de setembro, 1965. disponível em: www.planalto.gov.br. acessado em: 26 de agosto, 2005. Technical Synthesis Report | Land Use, Land-Use Change, and Forestry BrasIL. Presidência da república. Lei nº 5106, 2 de setembro, 1966. disponível em: www.planalto.gov.br. acessado em: 26 de agosto, 2005. BrasIL. assembléia nacional Constituinte. Constituição da república Federativa do Brasil, 5 de outubro de, 1988. disponível em: www.planalto.gov.br. acessado em: 29 de agosto, 2005. BrasIL. Presidência da república. decreto nº 97.628, 10 de abril, 1989. BrasIL. governo do estado de Minas gerais. Lei nº 10.561, 27 de dezembro, 1991. disponível em: www.ief.mg.gov.br. acessado em: 09 de setembro, 2005. Constituição Federal do Brasil http://www.planalto.gov.br/ccivil_03/constituicao/constitui%C3%a7ao.htm Lei Federal nº 4.771/1965 – Código Florestal http://www.planalto.gov.br/ccivil_03/LeIs/L4771.htm Lei estadual de Mudanças Climáticas nº. 3.135 de 04 de junho de 2007. Manaus: assembléia Legislativa. disponível em < http://www.florestavivaamazonas.org.br/download/ Lei_est_n_3135_de_050607.pdf > Lei Federal nº 6.938/1981 – Lei da Política nacional do Meio ambiente http://www.planalto.gov.br/ccivil_03/Leis/L6938.htm Lei Federal nº 9.985/2000 – sistema nacional de Unidades de Conservação (snUC) http://www.planalto.gov.br/ccivil_03/Leis/L9985.htm Lei Federal nº 11.284/2006 – Lei de gestão de Florestas Públicas http://www.planalto.gov.br/CCIVIL/_ato2004-2006/2006/Lei/L11284.htm 278 Lei Federal nº 11.428/2006 – Lei da Mata atlântica http://www.planalto.gov.br/ccivil_03/_ato2004-2006/2006/Lei/L11428.htm Websites and electronic documents consulted: http://www.cgee.org.br/publicacoes/MudancadoClima.php http://www.acr.org.br/arquivos2/14.pdf http://www.cepea.esalq.usp.br/economiaambiental/files/2005/04abr.pdf http://www.mct.gov.br Technical Synthesis Report | Land Use, Land-Use Change, and Forestry http://www.sectma.pb.gov.br/comites/rbma/pdf/cad3.pdf http://www.cnpma.embrapa.br http://www.mct.gov.br/upd_blob/0012/12919.pdf http://unfccc.int/meetings/cop_13/items/4049.php http://www.bndes.gov.br/programas/agropecuarios/pronaf.asp http://www.bndes.gov.br/programas/agropecuarios/propflora.asp http://www.integracao.gov.br/fundos/fundos_constitucionais/index.asp http://www.mma.gov.br/sitio/index.php?ido=conteudo.monta&idEstrutura=51 http://www.mma.gov.br/sitio/index.php?ido=conteudo.monta&idEstrutura=5 WORLD BANK