2005 79548 Sourcebook for Land Use, Land-Use Change and Forestry Projects Timothy Pearson, Sarah Walker and Sandra Brown With input from Bernhard Schlamadinger (Joanneum Research), Igino Emmer (Face Foundation), Wolfram Kägi (BSS) and Ian Noble, Benoit Bosquet and Lasse Ringius (World Bank) Sourcebook for Land Use, Land-Use Change and Forestry Projects Timothy Pearson, Sarah Walker and Sandra Brown With input from Bernhard Schlamadinger, Igino Emmer, Wolfram Kägi, Ian Noble, Benoit Bosquet and Lasse Ringius i SourceBook for Land use, land-use change and forestry Projects CONTENT S 1. Purpose and Scope. ........................................................................................................ 1 2. Introduction to the Kyoto Protocol and the Clean Development Mechanism Project Cycle............................................................................................ 2 2.1. The Clean Development Mechanism....................................................................................... 2 3. Introduction to the BioCarbon Fund and the BioCarbon Fund Cycle....... 3 4. Concepts of Additionality, Baseline, Leakage and Permanence..................... 4 4.1. Additionality. ....................................................................................................................... 4 4.2. Baseline................................................................................................................................ 4 4.3. Leakage................................................................................................................................ 5 4.4. Permanence.......................................................................................................................... 5 5. Specific Considerations for the kyoto protocol............................................... 6 5.1. Currently Acceptable LULUCF Projects. ................................................................................. 6 5.2. The Eligibility of Lands.......................................................................................................... 6 5.2.1. 31 December 1989 Rule............................................................................................... 6 5.2.2. Definitions of Forest..................................................................................................... 6 5.2.3. The Eligibility Tool....................................................................................................... 6 5.3. Additionality Tests. ................................................................................................................. 7 5.4. Choice of Baseline. .................................................................................................................. 8 5.5. Crediting................................................................................................................................ 9 5.6. Submission of New Afforesation/Reforestation Methodology...................................................... 10 6. DEVELOPING A MEASUREMENT PLAN................................................................................ 11 6.1. The Concepts of Accuracy, Precision and Being Conservative................................................... 11 6.2. Define the Project Boundaries............................................................................................... 12 6.3. Stratify the Project Area. ...................................................................................................... 12 6.4. Decide Which Carbon Pools to Measure................................................................................ 12 6.5. Determine Type, Number and Location of Measurement Plots................................................. 13 6.5.1. Type of Plots. ............................................................................................................ 13 6.4.2. Number of Plots. ....................................................................................................... 15 6.5.3. Location of Plots........................................................................................................ 18 6.6. Determine Measurement Frequency......................................................................................... 18 7. FIELD MEASUREMENTS. ...................................................................................................... 19 7.1. Preparation for Fieldwork..................................................................................................... 19 7.2. Trees, Palms and Lianas........................................................................................................ 20 7.2.1. Trees......................................................................................................................... 20 7.2.2. Palms........................................................................................................................ 21 7.2.3. Lianas....................................................................................................................... 21 7.3. Non-Tree Vegetation.............................................................................................................. 21 SourceBook for Land use, land-use change and forestry Projects ii 7.4. Dead Wood........................................................................................................................... 22 7.4.1. Standing Dead Wood. ................................................................................................ 22 7.4.2. Downed Dead Wood.................................................................................................. 22 7.5. Forest Floor (Litter Layer)....................................................................................................... 22 7.6. Soil. .................................................................................................................................... 23 8. Analysis. ............................................................................................................................. 24 8.1. Live Tree Biomass................................................................................................................ 24 8.2. Belowground Tree Biomass................................................................................................... 27 8.3. Non-Tree Vegetation............................................................................................................ 28 8.4. Standing Dead Wood. ......................................................................................................... 28 8.5. Downed Dead Wood........................................................................................................... 28 8.6. Forest Floor (Litter Layer).................................................................................................... 29 8.7. Soil.................................................................................................................................... 29 8.9. Estimating Net Change........................................................................................................ 30 8.9.1 Uncertainty................................................................................................................ 30 9. Non-CO2 Gases................................................................................................................... 33 9.1 Transport and Machinery..................................................................................................... 33 9.2. Fertilisation........................................................................................................................ 33 9.3. Fire.................................................................................................................................... 33 10. Quality Assurance and Quality Control............................................................. 34 10.1. QA/QC for Field Measurements........................................................................................... 34 10.2. QA/QC for Sample Preparation and Laboratory Measurements................................................ 34 10.3. QA/QC for Data Entry........................................................................................................ 34 10.4. QA/QC for Data Archiving.................................................................................................. 35 11. Guidance on Leakage. ................................................................................................. 36 12. References. ..................................................................................................................... 38 APPENDIX A: Glossary....................................................................................................... 39 APPENDIX B: Creating Biomass Regression Equations.......................................... 40 Method 1: Developing Biomass Equations...................................................................................... 40 Method II: Mean Tree Biomass Estimate......................................................................................... 40 APPENDIX C: Published Biomass Regression Equations.......................................... 41 Temperate Equations:................................................................................................................... 41 Tropical Equations:. ..................................................................................................................... 43 Agroforestry Equations. ................................................................................................................ 44 Appendix D: Checklist for Cdm afforestation/reforestation Projects. .... 46 SourceBook for Land use, land-use change and forestry Projects 1 1. P u r p o s e a n d S co p e This sourcebook is designed to be a guide for developing and imple- menting land use, land-use change and forestry (LULUCF) projects for the BioCarbon Fund of the World Bank that meet the requirements for the Clean Development Mechanism (CDM) of the Kyoto Protocol. Only project types and carbon pools that are eligible for credit under the CDM during the first commitment period (2008-2012) are covered. With its user-friendly format, the sourcebook introduces readers to the CDM processes and requirements, and provides methods and procedures to produce accurate and precise estimates of changes in carbon stocks. The sourcebook is not designed as a primer on field measurement tech- niques, although guidance is given. The sourcebook is intended as an addition to the IPCC Good Practice Guidance on Land Use, Land-Use Change and Forestry (2003), providing additional explanation, clarification and enhanced methodologies. It is designed to be used alongside the Good Practice Guidance. Acknowledgements Content on new methodologies and reasons for failure in the first year of consideration were largely derived from comments Ken MacDicken, David Shoch and Matt Delaney played a central by Martin Enderlin (chair of the A/R Working Group and role in developing the methods presented here. We also wish to member of Clean Development Mechanism Executive Board) thank the organisations and agencies that have funded our work over during his presentation at the Winrock International side event the past 10 years, which made possible the advances we have at COP/MOP 1 in Montreal in 2005. The title of the side achieved – in particular, The Nature Conservancy, US Agency for event was “Gaining approval for Land Use, Land-Use Change International Development, USDA Forest Service, United Nations and Forestry projects and project methodologies under the Clean Development Programme and World Bank. Finally, we would like Development Mechanism: lessons learned�. to acknowledge Ian Monroe for creating the illustration on page 21.  SourceBook for Land use, land-use change and forestry Projects  n t r o d u c t i o n to t h e K yoto P r oto co l a n d t h e Cl e a n 2. I D e v e lo p m e n t M e c h a n is m P r o j e c t C yc l e Carbon exists in everything that is living or has ever lived. There The UNFCCC established a CDM Executive Board that is is a perpetual cycle of carbon being sequestered on earth and emit- charged with approving or rejecting project designs and method- ted back into the atmosphere. Humankind increasingly influ- ologies, registering and administering project auditors (designated ences this carbon cycle through the burning of ever-greater quan- operational entities) and approving the issuance of certified emis- tities of oil, gasoline and coal and the cutting down of forests. It sion reductions. is argued that the human-induced accumulation of carbon diox- ide (CO2) and other greenhouse gases in the atmosphere is driving For each project, a Project Design Document must be submitted climate change. It is likely that current atmospheric concentra- that employs an approved methodology, including baseline and tions are at a 20-million-year high and that current rates of accu- monitoring methods. It is envisaged that, in the future, a group of mulation are unprecedented [1 ]. approved methodologies will exist that can be applied to new projects. However at the time of writing, only one methodology The Kyoto Protocol of the United Nations Framework Convention had been accepted. The Project Design Document describes the on Climate Change (UNFCCC) was developed as an attempt to project, illustrates how the methodology will be applied, estimates confront and begin to reverse the rising CO2 concentrations. In the greenhouse gases and environmental and socio-economic im- 1997, 38 industrialised nations signed the Kyoto Protocol and pacts of the project, including all baseline information, and agreed to cut their emissions of greenhouse gases between 2008 presents a monitoring plan. and 2012 to levels 5.2 per cent below 1990 levels. By June 2005, 150 countries had ratified the Kyoto Protocol, including 34 of the For the first commitment period (2008-2012), Annex I Parties are 38 industrialised nations whose emissions account for 61.6 per limited in the extent to which they can use offsets from LULUCF cent of all industrialized nations’ emissions. to meet their reduction commitments. The total additions to an Annex I Party’s assigned amount from emissions that can result Emissions of CO2 from land use and land-use change represent up from LULUCF project activities under the CDM is constrained at to 20 per cent of current CO2 emissions from burning fossil fuels one per cent of base year emissions of that country per year for the [2, 3 ]. Changes in land-use can positively impact atmospheric five years of the commitment period. CO2 concentrations by either: i) decreasing emissions that would occur without intervention, or ii) sequestering CO2 from the at- mosphere into vegetation and the associated soil. Preventing de- forestation, decreasing the impact of logging or preventing the drainage of wetlands or peat lands are practices that decrease emis- sions. In contrast, planting trees, changing agricultural tillage or cropping practices, or re-establishing grasslands sequester carbon. The Kyoto Protocol recognised the role that changes in the use of land and forests have on the global carbon cycle. Parties to the Protocol can use credits generated either by sequestering carbon or by reducing carbon emissions from land use to help them reach their reduction targets. Carbon credits can be produced within the emission-source country or in an alternative industrialised na- tion (Joint Implementation [JI], Article 6). In addition, the Pro- tocol includes a mechanism by which industrialised (Annex I) nations can offset some of their emissions by investing in projects in non-industrialised (non-Annex I) nations (CDM, Article 12). 2.1. The Clean Development Mechanism “The purpose of the clean development mechanism shall be to assist Parties not included in Annex I in achieving sustainable development and in contributing to the ultimate objective of the convention, and to assist parties included in Annex I in achieving compliance with their quantified limitation and reduction commitments.� Article 12 of the Kyoto Protocol (1997) SourceBook for Land use, land-use change and forestry Projects  I n t r o d u c t i o n to t h e Bi o C a r b o n F u n d 3.  a n d t h e Bi o C a r b o n F u n d C yc l e The World Bank’s BioCarbon Fund provides carbon finances for projects that sequester or conserve greenhouse gases in forest, agro- and other ecosystems. The BioCarbon Fund aims to “test and demonstrate how land use, land-use change and forestry ac- tivities can generate high-quality emission reductions with envi- ronmental and livelihood benefits that can be measured, moni- tored and certified and stand the test of time�. BioCarbon Fund projects have to fulfill criteria to ensure the fund meets its own targets in the areas of Climate and Environment; Poverty Alleviation; Project Management and Learning; and Port- folio Balance. Each BioCarbon Fund project is expected to deliver between 400,000 and 800,000 tons of CO2 equivalent (CO2e) over a pe- riod of 10 to 15 years. In return, a typical project will receive about US$2-3 million in payments ($3-4 per tonne CO2e). Prospective project developers submit a Project Idea Note. If both parties agree to take the proposal further, more formal documents are prepared, including an Emissions Reductions Purchase Agree- ment and a Project Design Document that is submitted to the CDM Executive Board. As of spring 2005, 140 Project Idea Notes had been submitted to the BioCarbon Fund and the window of opportunities for submission closed. However, future windows of opportunities for submissions are envisaged. For information, go to carbonfinance.org/biocarbon/home.cfm.  SourceBook for Land use, land-use change and forestry Projects 4. Co n c e p t s o f A d d i t i o n a li t y, B a s e li n e , L e a k ag e and Permanence This section introduces four core and interlinked concepts that 4.2. Baseline need to be understood to develop projects and acceptable method- ologies to deliver credits under the CDM of the Kyoto Protocol. As stated above, CDM afforestation and reforestation projects en- They are: additionality, baseline, leakage and permanence. Subse- hance greenhouse gas removals in one country to permit an equiv- quent sections of this sourcebook will draw upon these concepts alent quantity of greenhouse gas emissions in another country, in the context of the issues of developing methodologies. without changing the global emission balance. Technically, the CDM is a baseline-and-credit trade mechanism, not a cap-and- 4.1. Additionality trade mechanism. Therefore, enhancements of removals by affor- estation and reforestation projects must create real, measureable The CDM is a carbon-neutral process. It allows an Annex I Party and long-term benefits related to the mitigation of climate change and a non-Annex I Party to co-operate and carry out a project in (Kyoto Protocol, Article 12.5b), and must be additional to any the non-Annex I Party that will sequester carbon (or reduce emis- that would occur in the absence of the certified project activity sions). Certified emission reduction credits (CERs) are created (Kyoto Protocol, Article 12.5c). The “in the absence� scenario is through the project and transferred to the Annex I Party, which is also referred to as the baseline scenario. now able to emit an equivalent number of units of carbon while meeting its targets. Thus, the atmospheric concentration of green- The Marrakech Accords define a baseline scenario as one that “rea- house gases remains unchanged as a result of the transaction. The sonably represents greenhouse gas emissions that would occur in Annex I Party is assisted in meeting its commitments cost-effec- the absence of the proposed project activity� and is derived using tively while, in well-designed projects, the non-Annex I Party ben- an approved baseline method. The Marrackech Accords also state efits in meeting sustainable development goals. that the project baseline shall be established “in a transparent and conservative manner regarding the choices of approaches, assump- However, if the project that sequesters the carbon (or reduces tions� and that it shall be established “on a project-specific basis�. emissions) would have taken place without the CDM transaction, In summary, the baseline is the most likely course of action and then greenhouse gases in the atmosphere will increase as a result of development over time, in the absence of CDM financing. the transfer of CERs. For example, if an area would have been reforested, either through deliberate management action or The figure below shows the time-path of carbon stocks in the through natural processes, irrespective of the CDM transaction, project and baseline scenarios. then the CDM transaction simply allows the Annex I Party to emit more greenhouse gases and the atmosphere is worse off than it would have been without the transaction. Project scenario This is the purpose of the additionality clause in Article 12 of the (observable) Kyoto Protocol. Some confusion has arisen, however, because the agreed definition of additionality does not fully capture these core Carbon stocks concepts. The definition agreed at Ninth Conference of the Par- ties (COP9) in Milan in 2003 is: “The proposed afforestation or Additional carbon reforestation project activity under the CDM is additional if the removed from actual net greenhouse gas removals by sinks is increased above the atmosphere Baseline sum of the changes in carbon stocks in the carbon pools within scenario the project boundary that would have occurred in the absence of the registered CDM afforestation or reforestation project activi- ty…�. This definition focuses more on identifying the additional component than on project eligibility. Further guidance from the Time CDM Executive Board and recommended steps for dealing with additionality and baselines are outlined in Sections 5.3 and 5.4. However the essential question that must be asked of each project The baseline scenario can either be estimated and validated up- is: How much carbon is being sequestered as a direct result of the front and then “frozen� for the first phase of the crediting period CDM transaction? If more CERs are issued than this amount, (that is, 30 years or the first 20 years of up to 60 years), or it is also then the project increases greenhouse gases in the atmosphere. possible to monitor the baseline during the afforestation or refor- This test applies equally to LULUCF and non-LULUCF estation project. However, even in the latter case, it is still neces- projects. sary to establish a methodology upfront on how to select the con- SourceBook for Land use, land-use change and forestry Projects  trol plots and monitor them, and to provide an upfront estimation of the baseline, including the associated emissions and removals of greenhouse gases (the upfront estimation is for information only – the results of the monitored baseline would be used for calculat- ing emission reductions). The advantage of an upfront estimated and “frozen� baseline is that there is greater certainty about the emission reductions generated by the project. This is the option that has been used by most projects to date. 4.3. Leakage Some projects will be successful in sequestering more carbon with- in the project area, but the project activities may change activities or behaviours elsewhere. These changes may lead to reduced se- questration or increased emissions outside the project boundary, negating some of the benefits of the project. This is called leakage. A simple example is a project that reforests an area of poor quality grazing land, but leads to the owners of the displaced livestock to clear land outside the project boundaries to establish new pastures. The types of activities that might result in leakage vary with the type of projects, but both LULUCF and non-LULUCF projects are subject to leakage. Leakage can often be minimised by good project design – such as in the example above by including im- proved pasture management around the plantation so that dis- placed livestock can be accommodated without further clearing. Section 11 deals with leakage in more detail. 4.4. Permanence During the negotiations leading up to the Kyoto Protocol and sub- sequently, there was considerable concern that credits issued for carbon sequestration would be subject to a risk of re-emission, due to either human action or natural events such as wildfires. This was called the permanence risk and it is unique to LULUCF projects under the Protocol. Eventually, Parties agreed that credits aris- ing from CDM afforestation and reforestation projects should be temporary, but could be re-issued or renewed every five years after an independent verification to confirm sufficient carbon was still sequestered within the project to account for all credits issued. This deals effectively with the permanence risk and guarantees that any losses of sequestered carbon for which credits have been issued will have to be made up through either additional sequestration elsewhere or through credits derived from non-LULUCF activi- ties. Two types of temporary credits were agreed: temporary CERs and long-term CERs. Some accounting issues relating to these credits are described in Section 5.5. There are additional issues in relation to pricing, restrictions on replacement, etc, that also need to be taken into account. The BioCarbon Fund has documenta- tion to guide project managers on these issues.  SourceBook for Land use, land-use change and forestry Projects S p e c i f i c Co n si d e r at i o n s f o r t h e K yoto P r oto co l 5.  5.1. Currently Acceptable LULUCF Projects 5.2.2.1. Implications During the first commitment period (2008-2012), the only There are various implications for project eligibility based on LULUCF project types that are eligible for the CDM are which forest definitions are chosen. afforestation and reforestation. Tree crown cover Afforestation is the direct human-induced conversion of land that has not been forested for a period of at least 50 years, to forested A low tree crown cover threshold when defining a forest permits land through planting, seeding and/or the human-induced pro- the inclusion of restoration of open woodland type forest as a po- motion of natural seed sources. tential afforestation/reforestation project. Agroforests are also likely to fit under this low threshold, as such systems often do not Reforestation is the direct human-induced conversion of non- attain high crown cover. forested land to forested land through planting, seeding and/or human-induced promotion of natural seed sources, on land that A high tree crown cover threshold would allow for the inclusion of was forested but has been converted to non-forest land. For the many degraded forests as the starting condition for a potential af- first commitment period, reforestation activities will be limited to forestation/reforestation project. However, such a threshold reforestation occurring on those lands that did not contain forest would likely eliminate the use of agroforestry practices unless a on 31 December, 1989. high density of trees was used. In practice, no distinction is made under the CDM between afforesta- Land area tion and reforestation. A low minimum land area threshold permits the inclusion of small Neither forest management nor forest protection/conservation are patches of forests around farms and houses that may also serve as currently eligible. The project types eligible in the second commit- woodlots. ment period have not yet been established. A high minimum land area threshold will encourage large con- tiguous areas of forest with the consequent cobenefits to biodiver- 5.2. The Eligibility of Lands sity, land stabilisation and water quality. 5.2.1. 31 December 1989 Rule Tree height The criterion that all projects must meet is for no forest to be A low tree height threshold permits the inclusion of short, woody present within the project boundaries between 31 December 1989 forest vegetation, such as those that grow on poor soils or at alti- and the start of the project activity. Proof of forest absence could tude. It would also allow for the inclusion of commercial woody take the form of aerial photographs or satellite imagery from 1989 species such as coffee and some spice trees. or before, or official government documentation confirming the lack of forests. Where proof of these types does not exist, multiple A high tree height value permits the inclusion of some degraded independent, officially witnessed statements by local community forests as the starting condition for a potential afforestation/refor- members should suffice. estation project. Tree height is based on potential, not current height, so a low definition would allow the inclusion of shrubs but 5.2.2. Definitions of Forest not immature trees. The decision of what constitutes a forest has implications for what Ideally, the Designated National Authority would consider the lands are available for afforestation and reforestation activities. ecosystems in the country and which forest definitions would best National presiding authorities in non-Annex I countries, known serve national development goals. This will be simpler for a as Designated National Authorities, have the role of deciding for country that is relatively homogenous environmentally than a their country where to lay the thresholds from a range determined highly diverse nation with varied topography, soils and climates. at COP9, namely:  Minimum tree crown cover value between 10 and 30 per cent; 5.2.3. The Eligibility Tool Minimum land area value between 0.05 and 1 hectare;  Minimum tree height value between 2 and 5 metres.  The CDM Executive Board has developed a mandatory tool to be SourceBook for Land use, land-use change and forestry Projects  used to demonstrate the eligibility of lands (Executive Board 22nd If options (a) and (b) are not available/applicable, project (c)  Meeting, Annex 16). Following this decision, eligibility criteria participants shall submit a written testimony which was are no longer required in methodology documents but the eligibil- produced by following a participatory rural appraisal meth- ity tool should be applied for the Project Design Document. odology. Procedures to define the eligibility of lands for afforestation Participatory rural appraisal is an approach to the analysis of local and reforestation project activities problems and the formulation of tentative solutions with local stakeholders. It makes use of a wide range of visualisation meth- 1.  Project participants shall provide evidence that the land within ods for group-based analysis to deal with spatial and temporal as- the planned project boundary is eligible as an afforestation/re- pects of social and environmental problems. forestation CDM project activity following the steps outlined From Executive Board 22nd Meeting, Annex 16 below. 5.3. Additionality Tests Demonstrate that the land at the moment the project starts (a)   is not a forest by providing information that: The CDM Executive Board also developed a step-wise tool to i.  The land is below the forest national thresholds test the additionality of prospective project activities (Executive (crown cover, tree height and minimum land area) for Board 16th Meeting). A refined tool, especially for afforestation/ forest definition under Decisions 11/CP.7 and 19/ reforestation, was approved at the Executive Board 21st Meet- CP.9, as communicated by the respective Designated ing. Project developers are encouraged to use the tool to show the National Authority; and project activity would not have occurred in the absence of carbon ii.  The land is not temporarily unstocked as a result of financing. human intervention such as harvesting or natural From Executive Board 21st Meeting, Annex I6 causes or is not covered by young natural stands or plantations which have yet to reach a crown density or tree height in accordance with national thresholds Step 0. Preliminary screening based on the starting date and which have the potential to revert to forest with- of the afforestation/reforestation project activity out human intervention. (b)  Demonstrate that the activity is a reforestation or afforesta- PASS tion project activity:    i. For reforestation project activities, demonstrate that on 31 December 1989, the land was below the forest Step 1. Identification of alternatives to the afforestation/ national thresholds (crown cover, tree height and reforestation project activity, consistent with minimum land area) for forest definition under Deci- current laws and regulations sion 11/CP.7, as communicated by the respective Designated National Authority. ii.  For afforestation project activities, demonstrate that  the land is below the forest national thresholds (crown cover, tree height and minimum land area) for forest Step 2. Investment   arrier Step 3. B If not passed definition under Decision 11/CP.7, as communicated Analysis s Analysi by the respective Designated National Authority, for a period of at least 50 years. PASS In order to demonstrate steps 1(a) and 1(b), project partici- 2.   mpact of CDM Registration Step 4. I pants shall provide one of the following verifiable items of in- formation: (a)  Aerial photographs or satellite imagery, complemented by ground reference data; or PASS (b)  Ground-based surveys (land-use permits, land-use plans or information from local registers such as cadastre, owners register, land use or land management register); or Afforestation/Reforestation project activity is additional  SourceBook for Land use, land-use change and forestry Projects Preliminary screening based on starting date Step 0 –  and financial hurdles (Step 2) and/or other barriers (Step 3), of afforestation/reforestation enabling the project activity to be undertaken. project activity Registration of CDM project activities is only now begin- If there is an economic or financial incentive to undertake ning to occur, but the CDM Executive Board does not want the project without the CDM, and there are no barriers to penalise project activities that were mobilised early. to the project activity, then the project activity is not Project participants must provide evidence that the start additional. date of the activity was after 31 December 1999 and that the incentive from the sale More detail on the Additionality Tool can be found in of greenhouse gas allowances was seriously considered in Annex 1 of the report on the 16th Meeting of the CDM the decision to proceed with the activity. Executive Board (http://cdm.unfccc.int/EB/Meetings/021/ eb21repan16.pdf ). Step 1 –  Identification of alternatives to the afforesta- tion/reforestation project activity, consistent with current laws and regulations 5.4. Choice of Baseline Realistic and credible alternative land uses must be identified, including continuation of the current situation. The applicable legal and regulatory requirements must be Three approaches to creating a baseline were proposed at COP9: discussed for all alternatives. If the proposed project a) Existing or historical, as applicable, changes in carbon activity is the only alternative that is legally required, and stocks in the carbon pools within the project boundary; the requirements are enforced, then the project is not b) Changes in carbon stocks in the carbon pools within the additional. project boundary from a land use that represents an eco- nomically attractive course of action, taking into account Project developer may choose Step 2 or 3 or both. barriers to investment; c) Changes in carbon stocks in the carbon pools within the  Step 2 – Investment analysis project boundary from the most likely land use at the time Is the proposed project activity economically or financially the project starts. less attractive than the other alternatives (identified in Step 1) without the revenue from the sale of carbon credits? Project developers have to select the most appropriate approach and to justify their selection. Step 3 – Barrier analysis Does the proposed project activity face barriers to prevent Will the baseline be a continua- Choose implementation? Does this barrier fail to prevent the yes tion of the current land use? option a implementation of at least one of the alternatives (identified in Step 1)? These may include include: no  Investment barriers – for example, no source of funding to overcome initial costs of establishing Will the baseline change in land the activity; Choose use be motivated by economic yes option b Technological barriers – for example, lack of considerations, e.g., agriculture, properly skilled or trained labour, or lack of plantations, roads, industry? infrastructure to implement project;  Prevailing practice barriers – for example, the no project activity is a new practice in the country or region. Is the baseline change in land use Step 4 – Impact of CDM registration mandated by law, e.g., preserva- Choose An explanation is required of how the benefits and tion, low-impact harvesting, yes option c incentives of CDM registration will alleviate economic migration? SourceBook for Land use, land-use change and forestry Projects  Option a) indicates a continuation of the current land use, b) indi- If a country fails to reach its target with domestic AAUs and cates a change in land use motivated by economic considerations RMUs it can turn to flexible mechanisms: JI for trading between (for example, development or plantations or agroforestry), and c) Annex I countries and the CDM for credits derived in non-Annex indicates a change that is not motivated by economic considerations I countries. Emission Reduction Units (ERUs) are the units for JI (for example, changing legal requirements). trading and Certified Emission Reduction units (CERs) are the For afforestation/reforestation projects, project practitioners should units for CDM trading. An Annex I country that more than choose option a) if the baseline is a continuation of the current meets its target can convert its remaining AAUs and RMUs into land-use practice. If a change in the law or in enforcement of the ERUs to trade with Annex I countries that have not achieved the law would lead to a change in land use, select option c). Any other required reductions. change in land use will be economically motivated and option b) should be chosen In the Figure below, the first Annex I country’s emissions exceed its total allowable AAUs and RMUs. In contrast, the second An- 5.5. Crediting nex I country has low emissions and a surplus of AAUs and RMUs that it can convert to ERUs and sell under the JI programme. The Central to the Kyoto Protocol process are the allocation units and first country is able to overcome its excessive emissions by pur- crediting units. All units are in metric tonnes of CO2e – that is, chasing ERUs from the second Annex I country in addition to when greenhouse gases other than CO2 are converted into an equiv- CERs generated from a project in a non-Annex I country under alent quantity of CO2 in terms of global warming potential. (One the CDM. tonne of nitrogen dioxide [N2O] is equal to 296 tonnes of CO2e and 1 tonne of methane [CH4] is equal to 21 tonnes of CO2e). For LULUCF projects under the CDM, the fear of lack of perma- nence (Section 4.4) has led to the creation of expiring CER units. Each Annex I country has Assigned Amount Units (AAUs) which Two similar forms of certified emissions reduction schemes are total to the reduction target for that country for the end of the offered – the temporary CER (tCER) and the long-term CER crediting period. Any carbon sequestration an Annex I country (lCER). For both types, there is a choice between a single credit- achieves is added to their AAU total. Sequestration is measured in ing period (maximum 30 years) or a period of 20 years with the Removal Units (RMUs). possibility of two renewals (totalling 60 years). Once a CER cred- cdm cer eru ji Excess emissions above allocated units CDM-derived sequestration or emissions reductions Available emission Emissions capacity below allocated units rmu aau rmu Emis- sions aau Annex I Annex I Non-Annex I country 1 country 2 country 10 SourceBook for Land use, land-use change and forestry Projects iting period is over, the Annex I country must replace the carbon The submission of new methodologies has been a learning process either by purchasing another CER or by replacing it with an RMU for all involved. During the first year, the primary issues that or ERU. caused new methodologies to be rejected included improper or lacking explanation regarding: The tCERs last for just five years, at which time they can be reis- additionality; sued (if verification has occurred) or the Annex I country must methods for determining the project boundary; replace them. When a project developer retires a tCER after a description of the baseline approach, justification for this crediting period is over (after which, CDM regulations on that approach and land-use scenario determination; tCER will cease), the developer is then free to harvest the trees if consideration and selection of carbon and non-CO2 desired. The fees for issuing tCERs will likely be charged every greenhouse gas pools; five years which could significantly raise the cost of this option. At methods for determining net anthropogenic greenhouse gas the end of the crediting period, all tCERs expire. removals by sinks; as well as inadequacy in making recommended changes if the new In contrast, lCERs last for the entire length of the crediting peri- methodology was being submitted for a second time. od, but must be replaced either as soon as verification shows the carbon stock has decreased or if no verification has occurred for a Secondary issues that also caused new methodologies to fail period of five years. For a low-risk lCER, the price will approach included improper or lacking explanation regarding: that of an energy CER credit [4 ]. At the end of the crediting pe- methods for creating a baseline of net greenhouse gas removals  riod, all lCERs also expire. by sinks; methods for estimating actual net greenhouse gas removals by  The lCERs are more desirable for the project developer in that sinks; they will possess a higher value. Yet a purchaser will not invest in systems for addressing leakage;  lCERs for a project in which there is significant risk – in this situ- methods for compiling project emissions;  ation, the five-year obligation of tCERs is preferable. Addition- improper or inadequate description of models, formulas,  ally, if the price of CERs is expected to increase over time, a project algorithms and data sources used; developer may want to sell tCERs in the hope of receiving greater methods for addressing uncertainties; as well as  payment for future tCERs. the overall quality, drafting and language.  5.6. Submission of a New Afforestation/ Care should be taken to adequately address all of the above con- cerns. Due to the evolving nature of the negotiations, the CDM Reforestation Methodology website (www.unfccc.int/CDM) should be regularly consulted. All projects submitted to the CDM Executive Board must include a Project Design Document in which an approved afforestation/ reforestation methodology is applied. If the proposed project does not meet the conditions of any of the approved methodologies, a new afforestation/reforestation methodology must be submitted for approval along with the Project Design Document, illustrating how the new methodology can be applied. New methodologies are reviewed by the Afforestation/Reforestation Working Group and expert reviewers before being finally approved by the CDM Executive Board. All new methodologies should be user-oriented, concise and pro- vide step-by-step tools. The methodology must address all appli- cable issues, modalities, decisions by the COP and rules of the Executive Board. The conditions for the new methodology ap- plicability and assumptions must be clear, and explain why a new methodology is warranted. SourceBook for Land use, land-use change and forestry Projects 11 6. d e v e lo pi n g a m e a s u r e m e n t pl a n The guidance given here is intended as additional to the IPCC 6.1.  The Concepts of Accuracy, Precision and Good Practice Guidance on Land Use, Land-Use Change and For- Being Conservative estry (2003), providing elaboration, clarification and enhanced methodologies. The sourcebook should be used alongside the To estimate the carbon stock on the land, one could measure eve- Good Practice Guidance. It is also worth noting that the science rything – every single tree for example in the tens, hundreds or of forestry has been in development for hundreds of years. Many thousands of hectares of the project area. Complete enumerations textbooks exist that provide more detail than is possible to include are almost never possible, however, in terms of time or cost. Con- in this sourcebook – a good example is Forest Measurements [5 ]. sequently we must sample. The steps to preparing a robust measuring plan can be summarised Sampling is the process by which a subset is studied in order to in the following flow chart: allow generalisations to be made about the whole population or area of interest. The values attained from measuring a sample are Define project boundaries an estimation of the equivalent value for the entire area or popula- tion. We need to have some idea of how close the estimation is to reality and this is provided by statistics. Stratify project area There are two important statistical concepts that have to be under- stood: accuracy and precision. Decide which carbon pools to measure Accuracy is how close your sample measurements are to the actual value. Accuracy details the agreement between the true value and repeated measured observations or estimations of a quantity. Determine type, number and location of measurement plots Precision is how well a value is defined. In sampling, precision illustrates the level of agreement among repeated measurements of the same quantity. This is represented by how closely grouped are Determine measurement frequency the results from the various sampling points or plots. A popular analogy is a bull’s eye on a target. In this analogy, how tightly the arrows are grouped is the precision, while how close they are to the centre is the accuracy. Below in (A), the points are close to the centre and therefore accurate, but they are widely spaced and therefore imprecise. In (B), the points are closely grouped and therefore precise, but are far from the centre and so inaccurate. In (C), the points are close to the centre and tightly grouped – therefore both accurate and precise. (A) Accurate, but not precise (B) Precise, but not accurate (C) Accurate and precise 12 SourceBook for Land use, land-use change and forestry Projects When sampling for carbon, we want measurements that are both 6.3. Stratify the Project Area accurate (that is, close to the reality for the entire population) and precise (that is, closely grouped) so we can have confidence in the To facilitate fieldwork and increase the accuracy and precision of result. measuring and estimating carbon, it is useful to divide the project area into sub-populations or “strata� that form relatively homog- Sampling a subset of the land for carbon estimation involves tak- enous units. In general, stratification also decreases the costs of ing measurements in a number of locations or “plots�. The monitoring because it typically diminishes the sampling efforts number of plots is predetermined to ensure precision. The average necessary, while maintaining the same level of confidence (it does value when all the plots are combined represents the wider popu- so because there is a smaller variation in carbon stocks in each lation and we can tell how representative it is by looking at the stratum than in the whole area). Useful tools for defining strata confidence interval. A 95 per cent confidence interval is a com- include ground-truthed maps from satellite imagery, aerial photo- mon and appropriate measure which tells us that, 95 times out of graphs and maps of vegetation, soils or topography. 100, the true carbon density lies within the interval. If the inter- val is small, then the result is precise. The size and spatial distribution of the land area does not influ- ence site stratification – whether one large contiguous block of A third concept that is followed in carbon measurement work is land or many small parcels are considered the population of inter- that of being conservative. Sometimes it is just not possible to est, they can be stratified in the same manner. The stratification measure a particular pool, or a very broad estimate has to be made. should be carried out using criteria that are directly related to the In these cases, the most appropriate action is to pursue the most variables to be measured and monitored – for example, the carbon conservative options within the possible biological range. pools in trees. Note there is a trade-off between the number of strata and sampling intensity. The purpose of stratification should For example, if only an imprecise measurement were possible for be to partition natural variation in the system and so reduce mon- a project activity, then the most conservative approach would be itoring costs. If stratification leads to no, or minimal, change in to report the lower bound of the 95 per cent confidence interval. costs, then it should not be undertaken. In contrast, to be conservative on the baseline, the higher bound of the confidence interval would be used. As a result, a lower se- Potential stratification options include: questration total would be reported than if the mean had been  Land use (for example, forest, plantation, agroforestry, used, but the total will be appropriately conservative. grassland, cropland, irrigated cropland);  Vegetation species (if several); 6.2. Define the Project Boundaries  Slope (for example, steep, flat); Drainage (for example, flooded, dry); Project activities can vary in size from tens of hectares to hundreds  Age of vegetation; of thousands of hectares, and can be confined to a single or several  Proximity to settlement. geographic areas. The project area may be one contiguous block Typically, a project might have between one and six strata. of land under a single owner, or many small blocks of land spread over a wide area with a large number of small landowners or a few 6.4. Decide Which Carbon Pools to Measure large ones. The spatial boundaries of the land need to be clearly defined and properly documented from the start to aid accurate There are six carbon pools applicable to afforestation/reforestation LU- measuring, accounting and verification. LUCF project activities – aboveground trees, aboveground non-tree, belowground roots, forest floor (or litter), dead wood and soil or- ganic matter. However, not all six pools will be significantly im- STEP 1 – Obtain a map of your project area. pacted in a given project. STEP 2 – Define the boundaries using features on the At COP9, it was determined that “project participants may choose map or co-ordinates attained with a global not to account for one or more carbon pools … subject to the positioning system. provision of transparent and verifiable information that the choice will not increase the expected net anthropogenic greenhouse gas removals by sinks�. Therefore, a pool can be excluded as long as it can be reasonably shown that the pool will not decrease as as part of the project ac- SourceBook for Land use, land-use change and forestry Projects 13 tivity or will not increase as part of the baseline. as statistically more efficient in estimating changes in forest carbon stocks than temporary plots, because there is high covariance be- Beyond this stipulation, the selection of which pools to measure de- tween observations at successive sampling events [5 ]. pends on several factors, including expected rate of change, magnitude and direction of change, availability and accuracy of methods to Moreover, permanent plots permit efficient verification, if needed, quantify change and the cost to measure. All pools that are expected to at relatively low cost: a verifying organisation can find and meas- decrease as a result of activities should be measured and monitored. ure permanent plots at random to verify, in quantitative terms, the Pools that are expected to increase by only a small amount relative to design and implementation of the carbon monitoring plan. The the overall rate of change need not be measured or monitored. disadvantage of permanent plots is that their location could be known and they could be treated differently than the rest of the Clearly it makes sense to measure and estimate the carbon pool in project area – it is the responsibility of the auditing Designated live trees and their roots for all project activities – trees are simple Operational Entity to ensure that this has not occurred. to measure and contain substantial amounts of carbon. If permanent sample plots are used, marking or mapping the trees Aboveground non-tree or understory may need measuring if this to measure the growth of individuals at each time interval is criti- is a significant component, such as where trees are only present at cal so that growth of survivors, mortality and ingrowth of new low densities (for example, savanna). But non-tree vegetation is trees can be tracked. Changes in carbon stocks for each tree are generally not a significant biomass component in mature forest. estimated and summed per plot. Statistical analyses can then be performed on net carbon accumulation per plot, including in- Forest floor and dead wood also tend to only be a significant com- growth and losses due to mortality. ponent in mature forests. Dead wood is composed of standing dead trees and downed dead wood, and it is unlikely that signifi- Where measurements are only made at one point in time – such cant quantities will accumulate in the 30 to 60 years of an affores- as for baseline estimations – there is no value in marking plots and tation/reforestation project. trees. Soil organic carbon is likely to change at a slow rate and is also Shape and size of plots likely to be an expensive pool to measure. However it should at least be considered, as sequestration of carbon into the soil, or The size and shape of the sample plots is a trade-off between ac- prevention of emissions of carbon from soils, can be important – curacy, precision, time and cost for measurement. There are two especially in grazing land and cropland systems – and omission of types of plots – single plots of a fixed size or nested plots contain- soil carbon is an omission of a source of reductions in atmospher- ing smaller sub-units of various shapes and sizes. Experience has ic greenhouse gases. Potentially, where forest is planted on land shown that nested plots can be the most cost-efficient. that was previously grassland, a loss in soil carbon can occur (be- cause of the very high soil carbon stocks in perennial grasslands). Nested plots are a practical design for sampling for recording discrete size classes of stems. They are well-suited to stands with a wide range As afforestation/reforestation projects have a maximum timeframe of tree diameters or to stands with changing diameters and stem densi- of 60 years, it may make sense economically and in terms of effi- ties. Single plots may be preferred for systems with low variability, ciency to only measure live biomass in trees, given that this pool such as single species plantations. will dominate the total biomass. Nested plots are composed of several full plots (typically two to 6.5. Determine Type, Number and Location of four, depending upon forest structure), each of which should be Measurement Plots viewed as separate. The plots can take the form of nested circles or rectangles. Circles work well if you have access to distance meas- 6.5.1. Type of Plots uring equipment ([DME], for example, from Haglöf, Sweden) because then the actual boundary around the plot need not be 6.5.1.1 Tree carbon pools marked. If DME is not available, it may be more efficient to use rectangular plots that are laid out with tape measures and stakes. When estimating carbon changes in trees, permanent or tempo- rary sampling plots could be used for sampling through time. We When trees attain the minimum size (measured by diameter at recommend permanent plots for trees as we see more advantages breast height, or dbh) for a nested plot, they are measured and and fewer disadvantages. Permanent sampling plots are regarded included. When they exceed the maximum dbh size, measure- 14 SourceBook for Land use, land-use change and forestry Projects ment of the tree in that nest stops and begins in the next larger A single plot can be used just as effectively as a nested design and nest. How to track and analyse data from nested plots is described, may be preferred for systems with low variation, such as single with examples, in Section 8.1. species plantations. If a single plot is used, then the plot size should be large enough that at least eight to 10 trees will be meas- It is possible to calculate the appropriate plot size specifically for ured within the plot boundaries at the end of the project activity. each project; however, this adds an additional complication and (Therefore, substantially more than eight to 10 trees will be meas- an additional effort to the process. For simplicity, plot-size rules ured per plot at the start of the project activity.) are presented in the table below that can be applied to any project. Experience has shown these plot sizes will represent a reasonable Data and analyses at the plot level are extrapolated to the area of a balance of effort and precision. full hectare to produce carbon stock estimates. Extrapolation occurs by calculating the proportion of a hectare (10,000 m2) that is occupied by a given plot using expansion factors. As an example, Stem Diameter Circular Plot Square Plot if a series of nested circles measuring 4m, 14m and 20m in radius is used, their areas are equal to 50m2, 616m2 and 1,257m2 † < 5cm dbh 1m 2m x 2m respectively (using expansion factors of 198.9 for the smallest plot, 16.2 for the intermediate and 8.0 for the largest to convert the 5–20cm dbh 4m 7m x 7m plot data to a hectare basis). Expansion factors are described further in Section 8. 20–50cm dbh 14m 25m x 25m Because all carbon measurements are reported on a horizontal- > 50cm dbh 20m 35m x 35m projection basis, plots on sloping lands must use a correction fac- † stems < 5cm dbh would only be measured in very young forest. tor. This correction factor accounts for the fact that when dis- tances measured along a slope are projected to the horizontal The schematic diagram below represents a three-nest sampling plot in both circular and rectangular forms: Large plot: Intermediate plot: Small plot: radius 20m radius 14m radius 4m trees > 50cm dbh trees 20–50cm dbh trees 5–20cm dbh Large plot: 20m x 50m trees > 50cm dbh Intermediate plot: 17m x 35m trees 20–50cm dbh Small plot: 5m x 10m trees 5–20cm dbh SourceBook for Land use, land-use change and forestry Projects 15 plane, they are smaller. If the plot is split between level and sloping STEP 1 – Identify the desired precision level. ground, it is simpler to move the plot so that it is either entirely level or sloping. If the plot falls on a slope, then the slope angle The level of precision required for a carbon inventory has a direct should be measured using a clinometer. Where the plot is located effect on inventory costs as described above. Accurate estimates of on a slope that is greater than 10 per cent, the slope should be the net change in carbon stocks can be achieved at a reasonable quantified so that an adjustment can be made to the plot area at cost to within 10 per cent of the true value of the mean at the 95 the time of analysis. Details on this calculation are given in Sec- per cent confidence level [6]. The level of precision should be de- tion 8. termined at the outset – ±10 per cent of the mean is frequently employed, although a precision as low as ±20 per cent of the mean 6.5.1.2. Non-tree carbon pools could be used. There are no hard and fast rules for setting the precision level, but the lower the precision, the more difficult it will Non-tree carbon pools differ from trees in that it is not physically be to say with confidence that a change in carbon stocks has oc- possible to measure the identical sample at two periods in time. curred between two time periods. With non-tree vegetation, forest floor and soil, this is because the process of measuring the sample destroys the sample – it is col- Once the level of precision has been decided upon, sample sizes lected, weighed and dried in an oven. With downed dead wood can be determined for each stratum in the project area. Each car- the sample is not necessarily destroyed, but tracking pieces of dead bon pool will have a different variance (that is, amount of varia- wood between two periods of time is logistically very challenging. tion around the mean). However, experience has shown that fo- Consequently, for each of these pools, the samples are temporary. cusing on the variance of the dominant carbon pool (for example, To maintain statistical independence (an abstract concept that is trees for forestry activities) captures most of the variance. Even important to guarantee representative results), the sampling loca- though variation in the other components may be higher, if a high tion should be moved at each census. precision is attained in the dominant component, a lack of preci- sion in the other components will not harm the overall results. For the destructively sampled components, the size of the plot should be large enough to capture a sufficiently large sample while still maintaining a high level of sampling efficiency. Typically, for STEP 2 –  Identify an area to collect preliminary data. For herbaceous vegetation and forest floor, a small sub-plot of between example, if the activity is to afforest agricultural 0.25m2 and 0.5m2 is used. For shrubs, a larger plot of perhaps lands and will last for 20 years, then an estima- 1m2 could be used. For soil, typically four 30cm soil cores are tion of the carbon stocks in the trees of about six pooled to create a single sample for carbon concentration with an to 10 plots within an existing 15 to 20-year-old additional core for bulk density. Sections 7.3 to 7.6 have more forest would suffice. information on carrying out these measurements. Preliminary data are necessary in order to evaluate variance and, 6.5.2. Number of Plots from this, the required number of plots for the desired level of precision. Between six to 10 plots is usually sufficient to evaluate It is important that sampling is carried out with statistical rigour, variance. If the project consists of multiple strata, preliminary as it is likely this will be a requirement of the Designated Operat- data is required for each stratum. ing Entity. In employing this rigour, the first step is identifying the number of plots required to reach the desired precision in the results. STEP 3 – E  stimate carbon stock, standard deviation and variance from the preliminary data. An online tool for calculating number of plots is available at: STEP 4 – Calculate the required number of plots. http://www.winrock.org/Ecosystems/tools.asp. For L strata, the number of plots (n) needed = To use the tool, input the desired precision and the number, area, mean carbon density and co-efficient of variation for each strata. 2  L  With this information, the tool calculates the required number of  ∑ N h * sh  n = 2  h2=1  plots. N *E  L 2 To calculate number of plots without the tool, use the following 2 +  ∑ N h * sh  steps: t  h =1  16 SourceBook for Land use, land-use change and forestry Projects Where: E =  allowable error or the desired half-width of the confidence in- terval. Calculated by multiplying the mean carbon stock by the desired precision (that is, mean carbon stock x 0.1, for 10 per cent precision, or 0.2 for 20 per cent precision), t =  the sample statistic from the t-distribution for the 95 per cent confidence level. t is usually set at 2 as sample size is unknown at this stage, Nh = number of sampling units for stratum h (= area of stratum in hectares or area of the plot in hectares), n = number of sampling units in the population (n O ∑ Nh# sh = standard deviation of stratum h. This equation can be simplified. For a single-stratum project: @N x s#² nO N² x E2 + N x s2 t² For two strata: G@N1 x s1#+ (N2 x s2#F² nO N² x E2 + N1 x s1+ N2 x s22 t² The following two examples demonstrate the use of the formula and also illustrate the advantage of stratification. In this example, a 5,000-hectare project area requires 29 plots without stratifica- tion to be monitored to high precision, but only 18 plots with stratification. Single-stratum project Area = 5,000 ha Plot size = 0.08 ha Mean stock = 101.6 t C/ha Standard deviation = 27.1 t C/ha N = 5,000/0.08 = 62,500 Desired precision = 10 % E = 101.6 x 0.1 = 10.16 @62,500 x 27.1#² nO 62,500² x 0.12 + 62,500 x 27.12 2² = 29 plots SourceBook for Land use, land-use change and forestry Projects 17 For three strata: Stratum 1 Stratum 2 Stratum 3 Total Area (ha) 3,400 900 700 5,000 Plot size 0.08 0.08 0.08 0.08 Mean carbon density 126.6 76.0 102.2 101.6 (t C/ha) Standard deviation 26.2 14.0 8.2 27.1 N 3,400/0.08 900/0.08 700/0.08 5,000/0.08 = 42,500 = 11,250 = 8,750 = 62,500 Desired precision (%) 10 E 101.6 x 0.1 = 10.16 G@42,500 x 26.2#+@11,250 x 14#+@8,750 x 8.2#F² nO 62,500² x 10.16² + @42,500 x 26.2² # + @11,250 x 14.0 ² # + @8,750 x 8.2² # 2² = 18 plots The more variable the carbon stocks, the more plots are needed to For example, using the data from the calculations above: attain targeted precision levels. However, if a stratified project area requires more measurement plots than an unstratified area, re- Stratum 1 move one or more of the strata. The purpose of the stratification is [ ] to allow more efficient sampling. @42,500 x 26.2# nh O x 18 If a project site is stratified, the following formula can be used to (42,500 x 26.2) + (11,250 x 14) + (8,750 x 8.2) allocate the calculated number of plots among the various strata: = 15 plots Number of plots for each stratum: Stratum 2 ] N h x sh [ nh O n x @11,250 x 14# L ∑ Nh x sh nh O x 18 h=1 (42,500 x 26.2) + (11,250 x 14) + (8,750 x 8.2) Where: = 2 plots n =  the total number of plots, nh = the number of plots in stratum h, Stratum 3 [ ] N =  the number of sampling units in the population, @8,750 x 8.2# Nh = the number of sampling units in stratum h, nh O x 18 s =  the standard deviation, (42,500 x 26.2) + (11,250 x 14) + (8,750 x 8.2) sh = the standard deviation in stratum h. = 1 plot 18 SourceBook for Land use, land-use change and forestry Projects The formulas above can equally be used with non-tree carbon pools 6.6. Determine Measurement Frequency or soil. Such plots will be temporary and new random locations should be chosen at each measurement period. It is recommended that for carbon accumulation, the frequency of measurements should be defined in accordance with the rate of However, since tree biomass will dominate total biomass (and change of the carbon stock. therefore will also dominate the summed variance for the project),  Forest processes are generally measured over periods of five- it is practical to estimate the number of plots needed for the other year intervals; carbon pools based loosely on the number of plots for the dominant  Carbon pools that respond more slowly, such as soil, are biomass component. For example, a single 100m line intersect (for measured every 10 or even 20 years. downed dead wood, see Section 7.4.2), four clip plots for herba- ceous vegetation and the forest floor, and four soil samples would be As verification and certification must occur every five years for sufficient per tree plot. CDM project activities, it is reasonable that at least the dominant biomass pool (trees) should be measured at the same frequency. 6.5.3. Location of Plots Indeed, it may not be possible to claim credit for pools not meas- ured with a five-year frequency. To maintain statistical rigour, plots must be located without bias. The entirety of the project site should be sampled. If plots follow a For pools accumulating carbon more slowly (for example, dead road or trail, then all locations in the project do not have an equal wood or soil) it would be logical to measure at time zero and again chance of selection and a systematic bias has been introduced. In- at the end of the project activity, and to claim credit at this time stead, the location of plots should either be random or located using for all sequestration that has occurred in these pools. a fixed grid that covers the entire area. Where multiple carbon pools are measured, it is reasonable to base the location of the secondary pool plots on the location of the orig- inal plot for the first census. However, these plots should be outside the original plot and all subsequent remeasurement censuses should occur in a new location. STEP 1 – P  repare a map of the project, with the project boundaries of strata within the project clearly delineated. STEP 2 – D ecide whether plots will be distributed systematically or randomly. STEP 3a – The random location of plots can be achieved using random number tables, the random function in Geographic Information Systems programmes or alternatively by using the millisecond counter in a stopwatch to take a random bearing and random distance for assigning plots on the map. STEP 3b – T he systematic location of plots within each stratum can be achieved by overlaying a grid on the project map and allocating plots in a regular pattern across the strata. SourceBook for Land use, land-use change and forestry Projects 19 7. f i e l d m e a s u r e m e n t s 7.1. Preparation for Fieldwork Efficient planning for fieldwork is essential to reduce unnecessary labour costs, avoid safety risks and ensure reliable carbon esti- mates. The equipment used for fieldwork should be accurate and durable to withstand the rigours of use under adverse conditions. The type of equipment required will depend on the type of measure- ments. The following list covers most of what is typically used: – Compass for measuring bearings – Fibreglass metre tapes (100m and 30m) for measuring distances – Global Positioning System (GPS) for locating plots – Plot centre marker (rebar/PVC tubing) for marking plots – Metal detector for locating belowground plot markers – Aluminium nail and number tags for marking trees – Tree diameter at breast height (dbh) tape for measuring trees – Clinometers (percent scale) for measuring tree height and slope – Coloured rope and pegs or a digital for marking plot boundaries measuring device (DME) – 100m line or two 50m lines for measuring dead wood – Calipers for measuring dead wood – Hand saw for collecting dead wood samples and cutting destructive samples – Spring scales (1kg and 300g) for weighing destructive samples – Large plastic sheets for mixing forest floor/understory sample – Soil sampling probes for sampling soil – Rubber mallet for inserting soil probes – Cloth (for example, Tyrek) or paper bags for collecting soil and understory samples If trees are to be tagged (see Section 7.2.1), aluminum nails  ones should be avoided as they can stretch and result in and tags should always be used to avoid rust. If fire is inaccurate measurements. Dbh tapes are relatively inex- prevalent at the site, use an aluminum nail and a steel tag. pensive and are readily available from suppliers such as: Plots can be marked either conspicuously (for example,  www.forestry-suppliers.com or www.benmeadows.com. with PVC) or inconspicuously (for example, by sinking For collecting soil samples, cloth bags are preferred as paper  iron rods below the ground and navigating to plot using a ones have a tendency to rip. Do not use plastic bags, as Global Positioning System and metal detector). they do not allow for the samples to dry, which can result For square or rectangular plots, mark the four corners of the  in increased respiration and inaccurate results. plots. During the measurement, run flagging tape between the corner markers to delineate the edges. A compass with a declination adjustment is preferred, so  that accurate and replicable bearings can be taken. Dbh tapes are critical when making tree measurements.  Steel or aluminum dbh tapes are normally used. Cloth 20 SourceBook for Land use, land-use change and forestry Projects 7.2. Trees, Palms and Lianas STEP 3c – T  o ensure accurate accounting of ingrowth (that is, trees that grow into the minimum size class 7.2.1. Trees of the nested plot), the position of new trees should be recorded at each census with regard The biomass and carbon stocks of trees are estimated using ap- to each of the nested plots. propriate equations applied to the tree measurements. For practi-  ccasionally trees will be close to the boundary STEP 3d – O cal purposes, tree biomass is often estimated from equations that of a plot. Plots are typically small and will be relate biomass to dbh. Although the combination of dbh and expanded to estimate biomass carbon on a per height is often superior to dbh alone, measuring tree height can be hectare basis. It is therefore important to time-consuming and will increase the expense of any monitoring carefully decide if a tree is in or out of a plot. If program. Furthermore, databases of trees from around the world more than 50 per cent of the trunk is within the show that highly significant biomass regression equations can be plot boundary, the tree is in. If more than 50 developed with very high accuracy using just dbh. In forestry, dbh per cent of the trunk is outside of the boundary, is defined as 1.3m above the ground. it is out and should not be measured. If the tree is exactly on the border of the plot, flip a coin to STEP 1 –  Accurately locate the plot centre (use of a GPS determine if it is in or out. ed approach). is the preferr STEP 2 –  I  f the plot is permanent, mark the centre (if plot is circular) or the boundaries (if plot is square) – experience has shown metal rods and/or Using a dbh tape PVC pipe work well. Assign a unique number to the plot. It is important that a dbh tape is used properly to ensure STEP 3a –  Starting at the north of the plot, begin consistency of measurement: measuring trees. Flag the first tree to mark Be sure to have a staff or pole measuring 1.3m in length so  the start/end point. Measure trees at dbh the dbh location on the tree can be accurately identified, using the guidance below. or use a sturdy stick (at least 2cm in diameter). Alterna- STEP 3b – After meauring a tree, move clockwise to the tively, each member of the team should measure the loca- next tree. If the plots are to be remeasured, tag tion of dbh (that is, 1.3m above ground) on their own the trees using an aluminum numbered tag and bodies and use that location to determine the placement nail. It is not necessary to record tree species of the tape. unless species with different forms exist in the Dbh tapes often measure diameter on one side and cir-  same area (for example, pines and broadleaf cumference on the other. It is important that all measur- species, or palms and early colonising species). ers know which measurements to record. If the tree is on a slope, always measure on the uphill  side. If the tree is leaning, the dbh tape must be wrapped ac-  Tagging trees cording to the tree’s natural angle (not straight across, par- allel to the ground). When trees are tagged, the numbered tag and nail should be If the tree is forked at or below the dbh, measure just be-  placed at 10cm below dbh to avoid errors arising from bumps low the fork point. If it is impossible to measure below the or other imperfections that can develop at the site where the fork, then measure as two trees. Traditional forestry dic- nail enters the tree. In future inventories, the dbh measure- tates that forked stems be measured as two separate trees ment will be taken by measuring 10cm up from the nail. The but when the focus is on biomass, it is more accurate to aluminum nail should be inserted deep enough to hold the measure as a single tree wherever possible. tag firmly but with enough nail exposed for the tree to grow. If the tree has fallen but is still alive, then place the measur-  If the trees in the project area will be subjected to harvest in ing stick towards the bottom and measure at dbh just as if the future, the nail and tag should be placed at the base of the the tree was standing upright. Trees are considered alive if tree to avoid any accidents with chainsaws or other equip- there are green leaves present. ment. Each plot should contain a description of the approach If a liana or vine is growing on a tree that is going to be  that was used, so that future measurements can be completed measured, do not cut the liana to clear a spot to measure efficiently and accurately. SourceBook for Land use, land-use change and forestry Projects 21 the tree’s dbh. If possible, pull the liana away from the trunk and run the dbh tape underneath. If the liana is too STEP 1 – D  etermine if palms are present in the intermedi- big to pull away from the trunk, then use the back of the ate-sized nested plot and if any exceed 1.3m in dbh tape and pull it across the front of the tree and esti- height. mate the diameter visually. Cutting a liana from a tree STEP 2 –  For any palms exceeding 1.3m, measure the should only be a last resort because, over time and with height using a clinometer (or directly if the palm repeated measurements, interfering with the natural dy- is only a few metres tall). Measure only the namics in the plot can make it different from the sur- height of the stem, that is, from the base up to rounding forest. The same standard should be followed the spot where the stem is no longer visible. for any other type of natural organisms (for example, STEP 3 –  If the plot is to be remeasured, insert an mushrooms, epiphytes, fungal growths, termite nests, etc.) aluminum numbered tag at 10cm below dbh. that are found on the tree. 7.2.3. Lianas Lianas are difficult to measure because they are often long and cross the plot in several places. Unless they form a significant component of the ecosystem, they should not be measured be- cause of these problems and also because it is hard to find biomass equations to use with them.  etermine if lianas are a significant biomass STEP 1 – D component. STEP 2 –  If necessary, measure at dbh. Take care that the same liana is not measured more than once. Lianas do not normally grow to more than 10cm in diameter, so only measure in the smallest nest. 7.3. Non-Tree Vegetation Non-tree vegetation is measured by simple harvesting techniques. For herbaceous plants, a square frame (30cm x 30cm) made from Dbh measurement locations for irregular and normally PVC pipe is sufficient for sampling. For shrubs and other large shaped trees non-tree vegetation, larger frames should be used (about 1–2m2, depending on the size of the vegetation). Alternative methods for measuring trees exist, including a basal area prism to estimate basal area/volume, which are commonly STEP 1 – P  lace the clip frame at the sampling site. If applied in commercial forestry. Methods are also provided for necessary, open the frame and place around estimating biomass carbon from volume in the IPCC Good Prac- the vegetation. tice Guidance on Land Use, Land-Use Change and Forestry (2003). STEP 2 – C  lip all vegetation within the frame to ground Unless local volume equations exist, or the project is part of a level. The frame should be viewed as extending commercial forestry operation, it is advisable to use the allometric vertically, and any vegetation falling outside the method of directly estimating biomass. boundaries of the plot (even it is begins inside the plot) should be excluded. 7.2.2. Palms STEP 3 –  Weigh the sample and remove a well-mixed subsample for determination of dry-to-wet mass If palms are present, only the height should be recorded since bio- ratio. Weigh the subsample in the field, then mass in palms is more closely related to height than to diameter. oven-dry to a constant mass (usually at ~ 70°C). 22 SourceBook for Land use, land-use change and forestry Projects 7.4. Dead Wood  ssign each piece of dead wood to one of STEP 3 – A three density classes – sound, intermediate or 7.4.1. Standing dead wood rotten. To determine what density class a piece of dead wood fits into, each piece should be Within plots delineated for live trees, standing dead trees should struck with a saw or machete. If the blade does also be measured. The dbh and decomposition state of the dead not sink into the piece (that is, it bounces off ), it tree should be recorded. Decomposition classes for standing dead is classified as sound. If it sinks partly into the wood are defined practically as follows: piece and there has been some wood loss, it is classified as intermediate. If the blade sinks into Tree with branches and twigs and resembles a live tree 1.  the piece, there is more extensive wood loss and (except for leaves); the piece is crumbly, it is classified as rotten. Tree with no twig, but with persistent small and large 2.  STEP 4 –  Representative dead wood samples of the three branches; density classes, representing the range of species Tree with large branches only; 3.  present, should be collected for density (dry 4. Bole (trunk) only, no branches. weight per green volume) determination. Using a chainsaw or a hand saw, cut a complete disc from For classes 2, 3 and 4, the height of the tree and the diameter at the selected piece of dead wood. The average ground level should be measured and the diameter at the top diameter and thickness of the disc should be should be estimated. Height can be measured using a clinometer. measured to estimate volume. The fresh weight of the disc does not have to be recorded. The disc Top diameter can be estimated using a relascope or through the should be oven-dried to a constant weight. use of a transparent measuring ruler. Hold the ruler approximate- ly 10–20cm from your eye and record the apparent diameter of the top of the tree. The true diameter is then equal to: 7.5. Forest Floor (Litter Layer) Distance eye to tree (m) True diameter (m) O x Ruler measurement (m) Distance eye to ruler (m) The forest floor, or litter layer, is defined as all dead organic surface material on top of the mineral soil. Some of this material will still Distance can also be effectively measured with a laser range finder. be recognisable (for example, dead leaves, twigs, dead grasses and small branches) and some will be unidentifiable decomposed frag- 7.4.2. Downed dead wood ments of organic material. Note that dead wood with a diameter of less than 10cm is included in the litter layer. Lying dead wood is most efficiently measured using the line-inter- sect method [7, 8]. Only coarse dead wood (wood with a diameter Litter should be sampled at the identical time of year at each cen- > 10cm) is measured with this method – dead wood with a small- sus to eliminate seasonal effects. A square frame (30cm x 30 cm) er diameter is measured with litter. made from PVC pipe is suitable for sampling. STEP 1 – L  ay out two lines of 50m either in a single line or STEP 1 – P  lace the sampling frame at the sample site. at right angles. STEP 2 -  C  ollect all the litter inside the frame. A knife can  long the length of the lines, measure the STEP 2 – A be used to cut pieces that fall on the border of diameter of each intersecting piece of coarse the frame. Place all the litter on a tarpaulin dead wood (> 10cm diameter). Calipers work best beside the frame. for measuring the diameter. A piece of dead STEP 3a – W  eigh the sample on-site, then oven-dry to a wood should only be measured if: (a) more than constant weight. 50 per cent of the log is aboveground and (b) the  here sample bulk is excessive, the fresh STEP 3b – W sampling line crosses through at least 50 per cent weight of the total sample should be recorded of the diameter of the piece. If the log is hollow at in the field, and a subsample of manageable the intersection point, measure the diameter of size (approximately 80–100g) taken for the hollow; the hollow portion in the volume moisture content determination, from which estimates is excluded. the total dry mass can be calculated. SourceBook for Land use, land-use change and forestry Projects 23 7.6. Soil and then thoroughly mixed. The well-mixed sample should To obtain an accurate inventory of organic carbon stocks in min- not be oven-dried for the carbon analysis, but only air- eral or organic soil, three types of variables must be measured: (1) dried; however, the carbon concentration does need to be depth, (2) bulk density (calculated from the oven-dried weight of expressed on an oven dry basis at 105°C. The dry combus- soil from a known volume of sampled material), and (3) the con- tion method using a controlled-temperature furnace (for centrations of organic carbon within the sample. For convenience example, a LECO CHN-2000 or equivalent) is the recom- and cost-efficiency, it is advised to sample to a constant depth, mended method for determining total soil carbon [9] but maintaining a constant sample volume rather than mass. A 30cm the Walkley-Black method is also commonly used. probe is an effective measurement tool. STEP 1 – S  teadily insert the soil probe to a 30cm depth. If the soil is compacted, use a rubber mallet to fully insert. If the probe will not penetrate to the full depth, do not force it as it is likely a stone is blocking its route and, if forced, the probe will be damaged. Instead, withdraw the probe, clean out any collected soil and insert in a new location. STEP 2 – C arefully extract the probe and place the sample into a cloth bag. Because the carbon concentra- tion of organic materials is much higher than that of the mineral soil, including even a small amount of surface material can result in a serious overestimation of soil carbon stocks. STEP 3 – To reduce variability, aggregate four samples from each collection point for carbon concentra- tion analysis. STEP 4 – A t each sampling point, take two additional aggregated cores for determination of bulk density. When taking cores for measurements of bulk density, care should be taken to avoid any loss of soil from the cores. STEP 5 - S  oil samples can be sent to a professional laboratory for analysis. Commercial laboratories exist throughout the world and routinely analyse plant and soil samples using standard techniques. It is recommended the selected laboratory be checked to ensure they follow commonly accepted standard procedures with respect to sample preparation (for example, mixing and sieving), drying temperatures and carbon analysis methods. For bulk density determination, ensure the laboratory dries the samples in an oven at 105°C for a minimum of 48 hours. If the soil contains coarse, rocky fragments, the coarse fragments must be retained and weighed. For soil carbon determination, the material is sieved through a 2mm sieve 24 SourceBook for Land use, land-use change and forestry Projects 8. A n a lysis Most calculations determine values for the biomass of a particular 8.1. Live Tree Biomass carbon pool (except for soil, which usually measures carbon di- rectly). It is common practice to convert biomass to carbon by Biomass equations relate dbh to biomass. Equations may be for dividing by two: individual species or groups of species, but this literature is Biomass inconsistent and incomplete. Before applying a biomass equation, Carbon O 2 consider its original location, because trees in a similar functional group can differ greatly in their growth form between geographic However, if local values for the carbon content are available, use areas. these instead. The CDM Executive Board may, in the future, re- quire local measurements of mean carbon content. STEP 1 – S  earch for a suitable biomass equation. Either Extrapolating carbon stocks from a per plot basis to a per hectare use equations presented here (see Appendix C), basis requires the use of expansion factors, which indicate the area search the literature for equations, consult with each sample represents. This standardisation is required so that experts (perhaps in local universities or govern- results can be easily interpreted and also compared to other stud- ment forestry departments) or create new ies. The first step is to correct for slope so that all carbon values are equations (see Appendix B). reported on a horizontal projection. True horizontal radius is calculated using the formula: When making biomass calculations, the given maximum diameter L = Ls x cos S for the equation should be carefully observed. Using equations for trees that exceed the maximum diameters can lead to substantial Where: error (see [10] for ideas on how to address the problem of trees that L = the true horizontal plot radius, exceed the size limit of the database). Ls = the standard radius measured in the field along the slope, S = the slope in degrees, and The biomass equation should be verified for the project site. This cos = the cosine of the angle. can be done simplistically by estimating the volume of the tree stem (see Sections 7.4.1 and 8.4), using a standard factor of 1.2 to Correcting for slope after returning from the field results in a plot include the volume of branches, and multiplying by wood density of area: to attain biomass. Wood density values for most commercially important species are generally available (see [10]) or density can Circular Plot: Area = π x standard radius (Ls) x slope plot  be measured simply. The biomass equation can be verified through radius (L) comparison with estimations from a range of tree sizes. Area = Plot width x calculated true plot Rectangular Plot:  length (L) The importance of selecting an appropriate equation can be seen from the following example. In Appendix C, two biomass equa- For example, for a 20m radius plot on a slope of 25 degrees: tions are listed for pines in the USA – one for pines in the west and Ls = 20 x 0.91 or 18.1m (0.91 = cos 25). one for pines in the east. For a 50cm dbh tree, the western equa- Thus, the plot area = 3.142 x 20 x 18.1 = 0.11ha. tion produces a biomass estimate of 1.1 tonnes, while the eastern equation estimates 1.6 tonnes. A 1cm increment from 50cm to For a 25m square plot on a slope of 15 degrees: 51cm dbh results in a biomass increment of 54kg for the western Ls = 25 x 0.97 or 24.1 m (0.97 = cos 15). equation and 77kg for the eastern equation. Thus, the area of the plot = 25 x 24.1 = 0.06ha. STEP 2 – F  or each tree, calculate biomass using the chosen All expansion factors referred to from this point on are assumed to equation. use the slope-corrected area of the plot. The expansion factor is calculated as the area of a hectare in square metres divided by the area of the sample in square metres, that is: 10,000m2 Expansion factor O Area of plot, frame or soil core (m2) SourceBook for Land use, land-use change and forestry Projects 25 For example: A 55cm dbh tree was measured in moist tropical forest in Bolivia. A general equation for moist tropical forests was chosen (adapted from [10]): Biomass (kg) O exp (-2.289 + 2.649 x ln dbh - 0.021 x ln dbh2) A 55cm dbh is well within the maximum for this equation (148cm). 1. 2.649 x ln(55) = 10.615 2. 0.021 x ln(55)2 = 0.337 3. -2.289 + 10.615 – 0.337 = 7.989 4. exp (7.989) 2.95 tons of biomass = 2,948.3kg =  or 1.47 tons of carbon  or projects doing a one-time measurement, STEP 3a – F or for measurements with the purpose of establishing the required number of plots or the baseline carbon stock, sum the biomass of each tree in each nest then multiply by the expansion factor to get biomass per hectare for each nest. Finally, sum the nests to get the total estimated number of tons per hectare for that plot. STEP 3b – F  or projects that are tracking the accumulation of carbon in trees, subtract the biomass of a given tree at Time 1 from the biomass of the same tree at Time 2 to get the increment of accumulation. To be accurate in the calculations of change in carbon stocks, the biomass increment for ingrowth trees (that is, trees that were too small to be measured in the previous census) must be included correctly. To be conservative, the ingrowth tree is assigned the maximum dbh possible for that plot at the previous census. For example, if the minimum diameter for measurement is 10cm and a tree measured for the first time is 12.5cm, at the very least the tree has grown from just less than 10cm to 12.5cm dbh. Trees that die between censuses are given no increment of growth. They have left the live tree pool and entered the dead tree pool. Within nests, sum the increments and multiply the sum by the expansion factor. Finally, sum the nests to get the total estimated increment in tons per hectare for that plot. An example is provided overleaf. 26 SourceBook for Land use, land-use change and forestry Projects Calculating changes in aboveground tree carbon stocks from permanent, nested plots using allometric regression equations As a hypothetical example, a single plot will be examined. The plot consists of three nested, circular subplots: 4m radius for trees measuring 5cm to < 20cm dbh 14m radius for trees ≥ 20cm to < 50cm dbh 20m radius for trees ≥ 50cm dbh The figure below and table opposite show measurements over two time periods. Note at Time 2 the ingrowth of trees that were too small to be measured at Time 1 (trees 101 and 102 in the small nest and 103 in the intermediate nest) and outgrowth from one plot size and ingrowth into the next size when the maximum/ minimum thresholds are passed (trees 004 and 005 from small to intermediate, tree 009 from intermediate to large). The stars in the figure indicate the position of trees. At Time 2, the black stars indicate trees that remained in the same size class as at Time 1, the grey stars indicate trees that have grown into the next class, while white stars are trees that have exceeded the measure- ment minimum for the first time. Trees: 006, Tree: 010 007, 008, 009 Trees: 001, 002, 003, 004, 005 Time 1 Time 2 Trees: 001, 002, 003, 101, 102 Trees: 006, 007, 004, 005, 103 Trees: 010, 009 SourceBook for Land use, land-use change and forestry Projects 27 Time 1 Time 2 Tag Nest Dbh (cm) Biomass (kg) Tag Nest Dbh (cm) Biomass (kg) 001 Small 5.6 9.1 001 Small 6.1 11.4 002 Small 8.3 25.1 002 Small 8.9 30.0 003 Small 12.1 65.7 003 Small 13.2 82.0 004 Small 16.2 137.8 004 Intermediate 20.0 234.7 005 Small 18.1 182.4 005 Intermediate 22.1 301.9 006 Intermediate 20.2 240.6 006 Intermediate 20.9 262.2 007 Intermediate 22.3 308.8 007 Intermediate 23.3 344.8 008 Intermediate 38.6 1,221.9 008 DEAD DEAD 1,221.9 009 Intermediate 48.2 2,124.8 009 Large 51.0 2,444.9 010 Large 57.0 3,222.0 010 Large 58.0 3,364.0 101 Small 5.5 8.7 102 Small 5.9 10.5 103 Intermediate 20.3 243.7 Biomass increment in each subplot = For single (non-nested) plots the calculations are more simple. ( increments of trees remaining in subplot size class) + The minimum diameter for measurement must still be tracked but ( increments for outgrowth trees [= max biomass for size class there is no movement of trees between different plot sizes. – biomass at Time 1]) + ( increments for ingrowth trees [= bio- mass at Time 2 – min biomass for size class†]) 8.2. Belowground Tree Biomass Where = the sum of The measurement of aboveground biomass is relatively established † Minimum biomass for each size class is calculated by entering the and simple. Belowground biomass, however, can only be meas- minimum dbh for that size class into the regression equation (5cm for ured with time-consuming methods. Consequently, it is more the small plot, 20cm for the intermediate and 50cm for the large). In efficient and effective to apply a regression model to determine this example, 6.8 is the minimum biomass for the small plot, belowground biomass from knowledge of biomass aboveground. 234.7 for the intermediate and 2,327.5 for the large. The following regression models [11] are widely used: Small subplot [(11.4-9.1) + (30.0-25.1) + (82.0-65.7)] =  Boreal: + [(234.7-137.8) + (234.7-182.4)] + BBD (t/ha) = exp (-1.0587 + 0.8836 x ln ABD + 0.1874) [(8.7-6.8) + (10.5-6.8)] = 178.3kg [(262.2-240.6) + (344.8-308.8)] + Intermediate subplot =  Temperate: [(2,327.5-2,124.8)] + [(234.7-234.7) + BBD = exp (-1.0587 + 0.8836 x ln ABD + 0.2840) (301.9-234.7) + (243.7-234.7)] = 336.5kg Tropical: Large subplot (3,364.0-3,222.0)) + ((-)) + ((2,444.9- =  BBD = exp (-1.0587 + 0.8836 x ln ABD) 2327.5)) = 259.4kg Where: Biomass = the sum of biomass in each subplot x expansion factor BBD = belowground biomass density, and for that subplot: ABD = aboveground biomass density (t/ha) Small subplot 178.3 x 198.9 = 35,463.9 kg/ha Applying these equations allows an accurate assessment of below- Intermediate subplot 336.5 x 16.2 = 5,451.3 kg/ha ground biomass. This is the most practical and cost-effective Large subplot 259.4 x 8.0 = 2,075.2 kg/ha method of determining biomass of roots. For one-time measure- ments of root biomass, simply insert the aboveground biomass into Sum = 42,990.4 kg/ha = 43.0 t/ha for the time interval. the appropriate equation. 28 SourceBook for Land use, land-use change and forestry Projects For the calculation of increment in root biomass between two cen- 8.4. Standing Dead Wood suses, the exact usage of these equations is important. For tagged trees in permanent plots, it is not possible to simply calculate the  or decomposition class 1 (see Section 7.4.1), STEP 1 – F total aboveground biomass at Time 1 and Time 2, apply the equa- estimate the biomass of the tree using dbh and tions and then divide by the number of years, as this approach an appropriate equation as for live trees. cannot account for ingrowth or mortality trees. Instead below- STEP 2a – F  or class 1, subtract out the biomass of leaves ground biomass increments should be calculated using the follow- (about 2–3 per cent of aboveground biomass ing method: for hardwood/broadleaf species and 5–6 per cent for softwood/conifer species) (e.g., [12]).  alculate aboveground biomass at Time 1 using STEP 1 – C STEP 2b – F  or classes 2, 3 and 4, where it is not clear what allometric equations and the appropriate proportion of the original biomass has been expansion factors. lost, it is the conservative approach to estimate STEP 2 –  Calculate the increment of biomass accumulation the biomass of just the bole (trunk) of the tree. aboveground between Time 1 and Time 2 (see Section 8.1) and add to the Time 1 total biomass Volume is calculated using dbh and height measurements stock for an estimate of aboveground biomass and the estimate of the top diameter. It is then estimated density at Time 2. as the volume of a truncated cone. STEP 3 –  Apply the appropriate belowground equation Volume (m3) (Class 4) = ¹�₃ π h(r12 + r22 + r1 x r2) to estimate belowground biomass at each time interval. Where: STEP 4 –  ( Time 2 belowground – Time 1 belowground) / h = the height in metres, number of years = annual increment of biomass r1 = the radius at the base of the tree, belowground. r2 = the radius at the top of the tree. Volume is converted to dry biomass using an appropri- 8.3. Non-Tree Vegetation ate wood density. Biomass = Volume x Wood density (from samples)  alculate the dry mass of the sample. Where STEP 1 – C As the wood must be sound to support the still-standing a subsample was taken for determination of tree, the sound wood density from the downed dead wood moisture content: measurements (Section 8.5) can be used. Dry mass O [ subsample dry mass subsample fresh mass ] fresh mass of x whole sample 8.5. Downed Dead Wood STEP 2 –  The biomass density (the number of tons of biomass per hectare) is calculated by multiplying  alculate the wood density for each density class STEP 1 – C the dry mass by an expansion factor calculated (sound, intermediate and rotten, see Section from the sample-frame or plot size. 7.4.2) from the pieces of dead wood collected. Density is calculated by the following formula: Expansion factor O 10,000m2 mass (g) Area of plot (m2) Density (g/m3) O Volume (m3) Where: mass = the mass of the oven-dried sample, and volume = π x (average diameter/2)2 x average width of the fresh sample Average the densities to get a single density value for each class. SourceBook for Land use, land-use change and forestry Projects 29 8.6. Forest Floor (Litter Layer) STEP 2 -  For each density class, the volume is calculated separately as follows: STEP 1 – C  alculate the dry mass of the sample. Where Volume (m3/ha) O π2 x [ d12 + d22 ...dn2 8L ] a subsample was taken for determination of moisture content: where d1, d2 etc = diameters of intersecting pieces of dead wood in cm and L = length of the line in m. Dry mass O [ subsample dry mass subsample fresh mass ] x fresh mass of whole sample STEP 2 – T  he biomass density (the number of tons of STEP 3 – B  iomass of lying dead wood (t/ha) biomass per hectare) is calculated by multiplying = volume x density. the dry mass by an expansion factor calculated from the sample frame or plot size. In the following example, dead wood is sampled along 100m 10,000m2 Expansion factor O line (using the line-intersect method) to determine biomass Area of plot (m2) density. Diameters and density classes are recorded and a subsample collected to determine density in each of the three density classes (sound, intermediate, and rotten). The follow- 8.7. Soil ing numbers represent the hypothetical results: 13.8 cm sound STEP 1 – Calculate the bulk density of the mineral soil 10.7 cm sound core: 18.2 cm sound Bulk density (g/m3) = 10.2 cm intermediate Oven dry mass (g/m3) [ ] 11.9 cm intermediate Mass of coarse fragments (g) 56.0 cm rotten Core volume (m3) – Density of rock fragments (g/m3) Densities of subsamples: Sound: 0.43 t/m3 Intermediate: 0.34 t/m3 Where: Rotten: 0.19 t/m3 The bulk density is for the < 2mm fraction, coarse fragments are > 2 mm. The density of rock fragments is often given as Volume of sound wood: π2 x [d12 + d22…..dn2/8L] 2.65 g/cm3. π2 x [13.82 + 10.72 + 18.22/800] = 7.85m3/ha  sing the carbon concentration data obtained STEP 2 – U from the laboratory, the amount of carbon per Volume of intermediate wood: x π2 [10.22 + 11.92/800] unit area is given by: = 3.03m3/ha [(soil bulk density (gm3) x soil depth (cm) C (t/ha) =  x C)] x 100 Volume of rotten wood: π2 x [56.02/800] = 38.7m3/ha In this equation, C must be expressed as a decimal fraction – Biomass density  = (7.85 x 0.43) + (3.03 + 0.34) + (38.7 for example, 2.2 per cent carbon is expressed as 0.022 in the x 0.19) = 11.8t/ha equation. 30 SourceBook for Land use, land-use change and forestry Projects 8.8. Estimating Net Change Method 1 – Simple Error Propagation STEP 1 – I f results are initially calculated in tons of biomass  he plot-level results of increment of biomass STEP 1 – T per hectare, divide by two to give tons of carbon for living and standing dead trees, above- and per hectare. belowground, in permanent plots are averaged STEP 2 – The carbon stock for living and standing dead to give the mean and the 95 per cent confidence trees, above- and belowground, can be tracked intervals for the strata. through time for individual plots and the change STEP 2 – W  here temporary plots are used for trees, or the in carbon stocks calculated directly at the plot carbon pools of soils, downed dead wood, forest level. The change in carbon stocks for the floor or non-tree vegetation are included, the different components should be summed within uncertainty has to be calculated differently. The plots to give a per plot carbon stock change in confidence interval is then calculated as: t C/ha. The plot level results are then averaged Total 95% CI = √ 95% CITime 12 + 95% CITime 22 to give the mean for the stratum. STEP 3 – W here soils, downed dead wood, forest floor Where: and non-tree vegetation are included, they have 95% CITime 1 = 95% confidence interval for Time 1, and to be calculated differently. The change in 95% CITime 2 = 95% confidence interval for Time 2. carbon stock is calculated by subtracting the mean carbon stock at Time 2 from that at Time 1. STEP 3 -  The total confidence interval is calculated The annual increment is then calculated by as follows: dividing the change in stocks by the number of Total 95% CI = years between measurements. STEP 4 - T  he results of the various pools are combined √ 95% CIveg2 + 95% CIsoil2 + 95% CIDDW2 + 95% CIFF2 + 95% CINTV2 to produce an estimate of the total change. Where: STEP 5 – T he baseline is subtracted from the net change 95% CIveg = 95% confidence interval for vegetation, in carbon to calculate the net change in carbon 95% CIsoil = 95% confidence interval for soil, etc., and stock (or carbon benefit). DDW = downed dead wood, FF = forest floor and STEP 6 - I f the project were arranged into multiple strata, NTV = non-tree vegetation. then each would be calculated separately as detailed in Steps 1-4 and then combined. STEP 4 – I deally, the baseline will also have a 95 per cent STEP 7 - T  he mean change in carbon stocks per unit area confidence interval, in which case the confidence is then multiplied by the area of the project interval after the subtraction of means will equal: or entity to produce an estimate of the total Total 95% CI = √ 95% CICarbon Stocks2 + 95% CIbaseline2 change in carbon. STEP 8 - T  he total is then converted to tons of CO2 STEP 5 -  If the project was ordered into multiple strata, equivalent by multiplying by 3.67. then the new confidence interval for the combined strata would be estimated as follows: Total 95% CI = √ 95% CIs12 + 95% CIs22 ....... 95% CIsn 8.8.1 Uncertainty Where : There are two methods for calculating the total uncertainty for a 95% CIs1 = 95% confidence interval for stratum 1, project activity. The first method uses simple error propagation 95% CIs2 = 95% confidence interval for stratum 2, etc., through the root of the sum of the squares of the component for all strata (up to n) measured in the project. errors. The second method uses Monte Carlo simulations to propagate errors. The advantage of the first method is that it is  he total uncertainty in carbon stocks per unit STEP 6 - T simple to use and requires no additional computer software. area is multiplied by the area of the project or However, the second method should ideally be used where: entity to produce an estimate of the total change  Correlations exist between data sets – for example between in carbon. two carbon pools; STEP 7 -  The total is then converted to tons of CO2 Uncertainties are very large (greater than 100 per cent). equivalent by multiplying by 3.67. SourceBook for Land use, land-use change and forestry Projects 31 An example of the simple method is given below. In this case, the initial carbon stock in vegetation and soil on the land is assumed to remain constant throughout the estimation period. The base- line only has to be subtracted one time – at subsequent reporting intervals, the gross increment is the net increment. Calculating net change for the system The hypothetical example shown is a reforestation project on 500 hectares of degraded cropland. The baseline for carbon stocks in the absence of the project is continued coverage by annual crops with a carbon density of 0.9 t C/ha. The following table reports the carbon increment between years 1 and 10: Plot Number Increment in Carbon Pools (t C/ha) Sum (t C/ha) Living Biomass Dead Organic Matter Aboveground Trees Belowground Standing Dead Wood Plot 1 12.1 2.4 0.0 14.5 Plot 2 11.5 2.3 0.0 13.8 .... ... ... ... ... .... ... ... ... ... Plot 31 12.6 2.5 0.0 15.1 Plot 32 10.9 2.2 0.0 13.1 Mean of summed biomass increment in above- and belowground tree and standing dead wood = 13.8 t C/ha 95% CI = 2.4 + Increment in non-tree vegetation = 1.8 t C/ha 95% CI = 0.1 + Increment in downed dead wood = 0.1 t C/ha 95% CI = 0.1 + Increment in forest floor = 0.2 t C/ha 95% CI = 0.1 + Increment in soil organic carbon = 0.5 t C/ha 95% CI = 0.1 – Baseline biomass carbon stock = 0.9 t C/ha 95% CI = 0.1 = NET change in carbon stock = 15.5 t C/ha 95% CI = 2.4 Net change in stocks over project area: 15.5 t C/ha x 3.67 t CO2e/ha / t C/ha x 500ha ± the 95% CI: 2.4 t C/ha x 3.67 t CO2e/ha / t C/ha x 500ha Therefore the net change is: 28,443 ± 4,419 t CO2e over the measurement interval Method 2 – Monte Carlo Simulations Comparison of two methods for a single dataset The principle of Monte Carlo analyses is to perform the summing In theory, almost all LULUCF calculations should be performed of uncertainties many times using the uncertain stocks or incre- using Monte Carlo simulations because independence between ments chosen randomly by the computer software from within the various uncertainty values does not exist. For example, Time the distribution of uncertainties that the user initially inputs. 1 is always going to be correlated with Time 2 and dead wood stocks are going to be correlated with live tree biomass. These analyses can be carried out using Monte Carlo software such as Simetar, @Risk or Crystal Ball (www.simetar.com, www.pali- In the following example, calculations are carried out using the sade.com/html/risk.asp, www.crystalball.com). two methods outlined here on a single dataset. 32 SourceBook for Land use, land-use change and forestry Projects Data were collected from 111 plots in closed tropical forest in Belize. The pools sampled included live aboveground trees, standing dead wood, downed dead wood, herbaceous vegetation and litter. Live aboveground trees: 1  23.3 t C/ha ± 9.9 (mean ± 95% confidence interval) Standing dead wood: 3.5 t C/ha ± 1.0 Downed dead wood: 3.9 t C/ha ± 1.1 Herbaceous vegetation: 0.5 ± 0.1 Litter: 2.8 ± 0.3 Propagation of errors Total stock = 123.3 + 3.5 + 3.9+ 0.5 + 2.8 = 134.0 t C/ha Uncertainty = √ 9.92 + + 0.1 + 1.0 + 1.1 2 2 2 0.3 = 10.0 2 (95 % confidence interval) Monte Carlo analysis The data were fit to distribution curves: Log normal: Live aboveground trees; Normal: Litter; Exponential: Standing dead wood, lying dead wood and her- baceous vegetation. The products of the distributions were modeled through 100 iterations with the following result: Total stock = 134.6 t C/ha Uncertainty = 10.1 (95 % confidence interval) The propagation of errors therefore produced a confidence interval equal to 7.45 per cent of the mean. The equivalent for the Monte Carlo analysis was 7.50 per cent. The confi- dence intervals differed by 1.1 per cent. Clearly in the example above, there was little difference between the two methods. However, the measurements were relatively precise for all pools and there was little correlation between pools. Care should be taken when there is a high degree of cor- relation and/or the measured pools are highly variable. SourceBook for Land use, land-use change and forestry Projects 33 9. N o n - CO 2 G a s e s Other gases influence climate change as directly as CO2. Two 9.3. Fire gases related to land-use change activities are methane (CH4) and nitrous oxide (N2O). Although these gases are produced in small- Biomass burning is the greatest natural (or semi-natural) source of er quantities than CO2, their effect for a given mass on global non-CO2 gas production [13]. The quantity released can be esti- warming is greater. This is illustrated by the calculated global mated using emission factors based on the quantity of C released warming potential. Over a 100-year period, CH4 is expected to [13]. Fire emissions would have to be considered if site prepara- have a global warming potential equal to 21 times that of CO2 tion for planting involved prescribed burns. and N2O has a potential equal to 310 times that of CO2 [1]. Con- sequently, these gases need only be produced in quantities equal to CH4 emissions = Carbon released x 0.016 x CO2EFM 4 per cent and 0.3 per cent respectively of the mass of CO2 emit- Where CO2EFM = CO2 equivalent factor of 21 ted to have an equal effect with respect to climate change over 100 years. N2O emissions = Carbon released x 0.00011 x CO2EFN Where CO2EFN = CO2 equivalent factor of 310 CH4 and N2O are produced mainly as the result of anthropogenic activities, such as the use of machinery, fires, the draining of wet- land regions and the fertilisation of land [1]. Methods for estimating these non-CO2 greenhouse gas emissions are provided in the IPCC Good Practice Guidance for Land Use, Land-Use Change and Forestry [13] and the IPCC Revised 1996 Guidelines for National Greenhouse Gas Inventories [14]. Tier 1 methods (the most simple ones) are presented here – if any sourc- es are found to be significant (that is, more than 1 per cent of the total), then the users should return consider a Tier 2 or Tier 3 methodology. 9.1 Transport and Machinery Methods exist for calculating emissions even under Tier 1, but require complex, varied inputs. If gasoline or diesel are consumed heavily as part of project activities, then users should consult the methodology in the IPCC Revised 1996 Guidelines [14]. 9.2. Fertilisation If fertilisers are used to enhance tree growth, then N2O emissions should be considered. (FSN x EF1) x Direct N2O emissions from fertilisation =  CO2EFN Where: FSN =Annual amount of synthetic fertiliser nitrogen applied to soils EF1 =Emission factor for N2O emissions from fertilisation in unit of N (default value = 1.25 per cent) CO2EFN = CO2 equivalent factor of 310 34 SourceBook for Land use, land-use change and forestry Projects 10. Q ua li t y Ass u r a n c e a n d Q ua li t y Co n t r o l For verifiable and certifiable measurements of changes in carbon For all the verified plots: stocks, provisions are required for quality assurance (QA) and Measurement error (%) O quality control (QC) to be implemented. A QA/QC plan pro- (Biomass before corrections – Biomass after corrections) vides confidence to all stakeholders that the reported carbon cred- x 100 its are reliable and meet minimum measurement standards. The Biomass after corrections plan should become part of project documentation and cover pro- cedures for: (1) collecting reliable field measurements; (2) verify- QA/QC for Sample Preparation 10.2.  ing laboratory procedures; (3) verifying data entry and analysis techniques; and (4) data maintenance and archiving. To ensure and Laboratory Measurements these procedures are carried out in a repeatable manner, a set of Standard Operating Procedures should be prepared for each step. Standard operating procedures should also be prepared and rigor- ously followed for sample preparation and analyses. In many in- 10.1. QA/QC for Field Measurements stances, it is likely that commercial laboratories will be used. If so, it is important that their procedures follow accepted standards. For example, soil bulk density samples should be dried at 105°C Collecting reliable field measurements is an important step in the (221°F) in a drying oven to constant weight. By definition, soil QA plan. Those responsible for the carbon measurement work organic carbon is that which passes through a 2mm sieve, thus it should be fully trained in all aspects of the field data collection and is important that the laboratory follow this step. The well-mixed data analyses and Standard Operating Procedures should be fol- sample should not be oven-dried for the carbon analysis, but only lowed rigidly to ensure accurate measurement and remeasure- air-dried; however, the carbon concentration does need to be ex- ment. The Standard Operating Procedures should be detailed pressed on an oven-dry basis at 105°C (221°F). enough that any new person sent to the field would be able to ac- curately repeat the previous measurements. For example, the For QC, all combustion instruments for measuring carbon should Standard Operating Procedures should cover all aspects of the be calibrated using commercially available certified carbon stand- field measurements, including steps such as where to measure the ards. For example, blanks and samples of known carbon concen- dbh of a tree, how to classify dead wood and how to clearly delin- trations should be analysed in each batch/run. Similarly, all bal- eate the litter from the mineral soil. The detailed methods pre- ances for measuring dry weights should be periodically calibrated sented in this sourcebook are appropriate for creating Standard against known weights. Where possible, 10–20 per cent of the Operating Procedures for the field phase of a QA/QC plan. soil samples could be reanalysed/reweighed to produce an error estimate. Similar procedures should be applied to plant material Field crews should receive extensive training so they are fully cog- such as litter or understory. nisant of all procedures and understand the importance of collect- ing data as accurately as possible. An evaluation of the field crews Measurement error (%) O should be conducted to identify errors in field techniques, verify (Number of errors among checked sample) measurement processes and correct any identified problems before x 100 they carry out measurements. Total number checked If the calculated measurement error is greater than 10 per cent, A second type of field evaluation should be used to quantify meas- then rerun all the analyses. urement errors. To implement this type of evaluation, a complete remeasurement of a number of plots by people other than the original field crews is performed at the end of the fieldwork. The 10.3. QA/QC for Data Entry verifying crew should be experienced in forest measurement and highly attentive to detail. The auditing crew enters the field and Field data are either collected directly on electronic media or on remeasures every tree in about 10–20 per cent of the plots. After field sheets. If entered electronically in the field, then the field measurement, a comparison is made with the original data and data entry step is not needed – however, errors in field data entry discrepancies are reverified. Field data collected at this stage can can occur and efforts should be made to check this step. If col- be compared with the original data. Any errors found should be lected on field sheets, the accurate entry of data into the data corrected and recorded, and could be expressed as a percentage of analysis software is important. all plots that have been rechecked to provide an estimate of the measurement error. To check for data entry errors, it is suggested that another inde- pendent person should enter data from about 10–15 per cent of SourceBook for Land use, land-use change and forestry Projects 35 the field sheets into the data analysis software. These two data sets can then be compared to check for errors. Any errors detected should be corrected in the master file. Measurement error (%) O (Number of errors among checked sample) x 100 Total number checked If the calculated measurement error is greater than 10 per cent, re-enter the data. Data analysis software could be developed so that it has checks built into it to highlight potential errors in data entry. For exam- ple, such checks could include tests to check that the diameter limits for a given nested plot (if used) is within the limits set by the field work. Common sense should be used when reviewing the results of the data analysis, to make sure the results fit within the realm of real- ity. Errors can be reduced if the entered data are reviewed using expert judgment and, if necessary, through comparison with inde- pendent data. All personnel involved in measuring and analysing data should communicate closely to resolve any apparent anoma- lies before final analysis of the monitoring data is completed. 10.4. QA/QC for Data Archiving Because of the relatively long-term nature of forestry activities, data archiving (maintenance and storage) will be an important component of a project. Copies of all data analyses and models, the final estimate of the amount of carbon sequestered, any GIS products and copies of all measuring and monitoring reports should all be stored in a dedicated and safe place. Given the time frame over which a project may take place, and the pace of production of updated versions of software and new hard- ware for storing data, electronic copies of data and reports should be periodically updated or converted to a format that can be ac- cessed by any future software applications. 36 SourceBook for Land use, land-use change and forestry Projects 11. G u i d a n c e o n L e a k ag e Leakage is very difficult to calculate. BioCarbon Fund projects, with their focus on sustainable development, should not be great- ly susceptible to leakage as community alternative livelihood pro- grams will automatically be built into projects, diminishing the risk of the local community leaking carbon benefits outside the project boundaries. Leakage should, however, be considered and here we present a decision tree to determine the importance of leakage on a project- by-project basis. At a simple level, leakage can be split into three categories: activity shifting, market effects and super-acceptance. Activity shifting occurs when activities that cause emissions are not permanently avoided, but are simply displaced to another area. For example, if one area is set aside for reforestation, cattle farmers who were farming the area might deforest an alternative area outside the project boundaries to replace their lost grazing land. Market effects occur when emission reductions are countered by emissions created by shifts in supply and demand of the products and services affected by the project. This is of minimal impor- tance for farming activities, but can be important for large-scale commercial timber harvesting. For example, a stop-logging project might decrease the supply of timber, leading other practi- tioners to increase their rate of harvest. Market effects leakage is not likely to be a problem, however, for afforestation/reforestation project activities. Super-acceptance may result from the alternative livelihoods ac- tivities created for the project. If the activities are very successful, they can draw in people from the surrounding regions. The result may be positive1 or negative leakage. It will be positive if the im- migrants were previously deforesting or practising a similarly high greenhouse gas-emitting lifestyle, but negative if the immigrants previously had lower greenhouse gas-emitting lifestyles and now have access to new land, for example, to deforest. Adapted from [15] The science of evaluating leakage is not well developed. However if it is suspected that leakage may occur, for example, with dis- placed farmers cutting forest to replace land that is reforested as part of the project, a significant alternative livelihoods programme could diminish the impact. The decision tree opposite helps identify whether leakage is likely to occur and what form the leakage might take. 1 Positive leakage is currently not permitted under the CDM. SourceBook for Land use, land-use change and forestry Projects 37 Does the project include an alternative livelihoods programme? no yes Activity shifting leakage Has the local community likely to occur no engaged in alternative livelihoods options? yes Was the local community previously engaged in commercial activities? Or was a commercial operator active in the area prior to the project? no yes Is there evidence of super-acceptance Market effects leakage of the alternative livelihoods possible programme by either the local community or external actors? no yes No further analysis Leakage (positive or needed: no leakage negative) possible due expected to super-acceptance Adapted from [15] 38 SourceBook for Land use, land-use change and forestry Projects 12. R e f e r e n c e s  oughton, J.T., Y. Ding, D.J. Griggs, M. Noguer, P.J. van der [1] H Cairns, M. A., S. Brown, E. H. Helmer, and G. A. Baumgardner. [11]  Linden, X. Dai, K. Maskell and C.A. Johnson (eds). 2001. 1997. Root biomass allocation in the world’s upland forests. Climate Change 2001: The Scientific Basis. Contribution Oecologia 111: 1-11 to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, [12]  Jenkins, J.C., D.C. Chojnacky, L.S. Heath, and R.A. Birdsey. Cambridge, UK, and New York, USA. 881 pp. 2003. National-scale biomass estimation for United States tree species. Forest Science 49: 12-35.  ixon, R. K., S. Brown, R. A. Houghton, A. M. Solomon, M. C. [2] D Trexler, and J. Wisniewski. 1994. Carbon pools and flux of [13]  Namburs, G-J, N.H. Ravindranath, K. Paustian, A. Freibauer, global forest ecosystems. Science 263: 185-190. W. Hohenstein and W. Makundi (eds). 2004. Chapter 3: LUCF Sector Good Practice Guidance. In: Penman, J, M. [3] Brown, S., J. Sathaye, M. Cannell, and P. Kauppi. 1996. Gytarsky, T. Hiraishi, T. Krug, D. Kruger, R. Pipatti, L. Buendia, Management of forests for mitigation of greenhouse gas K. Miwa, T. Ngara, K. Tanabe and F. Wagner. Good Practice emissions. Chapter 24 in R. T. Watson, M.C. Zinyowera, and Guidance for Land Use, Land-Use Change and Forestry, R.H. Moss (eds.), Climate Change 1995: Impacts, Adapta- IPCC National Greenhouse Gas Inventories Programme, tions and Mitigation of Climate Change: Scientific- Intergovernmental Panel on Climate Change. Technical Analyses. Contribution of Working Group II to the Second Assessment Report of the Intergovernmental Intergovernmental Panel on Climate Change. 1996. Revised [14]  Panel on Climate Change, Cambridge University Press, 1996 IPCC Guidelines for National Greenhouse Gas Cambridge, UK, and New York, USA. Inventories, Volumes 1, 2 and 3. Intergovernmental Panel on Climate Change Secretariat, Geneva, Switzerland. [4] Dutschke, M, B. Schlamadiner, J. Wong and M. Rumsberg. 2004. Value and risks of expiring carbon credits from CDM Aukland, L., P. Moura Costa and S. Brown. 2003. A concep- [15]  afforestation and reforestation. HWWA Discussion Paper tual framework and its application for addressing leakage on 290. 34 pp. avoided deforestation projects. Climate Policy 3: 123-136. [5] Avery T.E. and H.E. Burkhart (eds.). 1994. Forest Measure- ments, 4th edition. McGraw-Hill, New York.  rown, S. 2002. Measuring, monitoring, and verification of [6] B carbon benefits for forest-based projects. Phil. Trans R. Soc. Lond. A 360: 1669-1683.  rown, J. K. 1974. Handbook for inventorying downed [7] B woody material. General Technical Report INT-16. Ogden, Utah: USDA Forest Service Intermountain Forest and Range Experiment Station.  armon, M. E. and J. Sexton. 1996. Guidelines for Measure- [8] H ments of Woody Detritus in Forest Ecosystems. US LTER Publication No. 20. US LTER Network Office, University of Washington, Seattle, USA.  elson, D.W., and L.E. Sommers. 1996. Total carbon, [9] N organic carbon, and organic matter. p. 961-1010. In: D.L. Sparks et al. (eds.) Methods of soil analysis. Part 3. Chemical methods. SSSA, Madison, USA. Brown, S. 1997. Estimating Biomass and Biomass Change [10]  of Tropical Forests: A Primer. UN FAO Forestry Paper 134, Rome. 55 pp. SourceBook for Land use, land-use change and forestry Projects 39 A P P END I X A : G LO S S ARY Accuracy: how close a measurement is to its true value. atmosphere are water vapour, carbon dioxide, nitrous oxide, methane Activity shifting: when activities that cause greenhouse gas emis- and ozone. sions are not permanently avoided through project implementation, Hardwoods: this botanical group of trees has broad leaves and pro- but are instead displaced to another area causing carbon leakage (see duces a fruit or nut. Section 11 for more information). Leakage: the loss of carbon outside the boundaries of the project as Baseline: the emission or removal of greenhouse gases that would a result of project activities. There are three categories of leakage: ac- occur without the project. tivity shifting, market effects and super-acceptance. Biomass: organic material (above- or belowground, live or dead). Market effects: when emission reductions under a project are coun- Boreal: mean annual temperature of less than 0oC. tered by emissions created by shifts in supply and demand of the Carbon pool: organic material containing carbon. products and services affected by the project (see Section 11 for more Carbon stock: the quantity of carbon in a given pool or pools per information). unit area. Mean: is the sum of observations divided by the number of observa- Confidence interval: a measure of the spread of the data. It gives a tions. Mean is calculated in Microsoft Excel using: =AVERAGE range of values in which there is a percentage probability (usually 95 (...list of observations...). per cent) of the true mean occurring. Calculated by multiplying the Precision: the repeatability of a measure or the range of value be- standard error by the appropriate t value. T values for calculating the tween which the true value may lie. 95 per cent confidence interval are given below. Sequestration: the process of increasing the carbon stock in an eco- system. Number of t value Number of t value Softwoods: softwoods and conifers (from the Latin word meaning Observations Observations cone-bearing) have needles. Standard deviation: a measure of the spread of the data. It is calcu- 5 2.776 60 2.001 lated in Microsoft Excel using: =STDEV (...list of observations...). 10 2.262 65 1.998 Standard error: a measure of the spread of the data. It is calculated 15 2.145 70 1.995 by dividing the standard deviation by the square root of the number 20 2.093 75 1.993 of observations. 25 2.064 80 1.990 Super-acceptance: occurs when alternative livelihoods activities 30 2.045 90 1.987 created for a project are very successful and draw in people from the 35 2.032 100 1.984 surrounding regions. The result may be a positive or negative carbon 40 2.023 110 1.982 45 2.015 120 1.980 leakage (see Section 11 for more information). 50 2.010 150 1.976 Temperate: mean annual temperature between 0oC and 20oC. 55 2.005 200 1.972 Tropical: mean annual temperature greater than 20oC. Variance: a measure of the spread of the data. It is calculated in Mi- crosoft Excel using: =VAR (...list of observations...). Cropland: defines any land on which non-timber crops are grown. Without-project scenario: see baseline. This includes both herbaceous crops and higher carbon-content sys- tems including vineyards and orchards. diameter at breast height (dbh): tree diameter parallel to the ground at 1.3m above the ground. Usually measured using a dbh tape, which is calibrated to diameter when the user measures the cir- cumference of the tree. Forests: includes all land with a canopy cover greater than 30 per cent. This can include natural forest, plantations, forested wetlands and mangroves. Grazing land: a very broad category that includes managed pastures, prairies, steppe and savannas. Grazing lands will often include trees, but only when the canopy cover is less than 30 per cent. Aquatic systems, such as flooded grasslands and salt marshes, are also included in this category. Greenhouse gases: gases in the atmosphere (both natural and anthropogenic) that absorb and emit radiation. This property of the gases causes the greenhouse effect. The primary gases in the earth’s 40 SourceBook for Land use, land-use change and forestry Projects A P P END I X B : C r e at i n g Bi o m a ss R e g r e ssi o n Eq uat i o n s Method 1: Developing Biomass Equations Method 2: Mean Tree Biomass Estimate Developing local biomass equations can be a resource-expensive To avoid felling a large number of trees (>30) and the cost operation. When dealing with native forests, it is highly likely of estimating their mass, the mean tree biomass method is an that general equations exist (such as those in Appendix C). option, although this method is not as accurate as the species- However, for many multi-purpose species, this may not be the specific biomass equation derived using Method 1. case and it is necessary to develop local biomass equations. Procedures for developing location- and species-specific biomass equations involves the following steps: Step 1 –  Using dbh data from field measurements, prepare frequency tables using appropriate class intervals (for example, 5cm for each tree Step 1 – Select the dominant tree species. species). The smaller the class interval, the Step 2 – Select about 30 trees to represent the full range lower the error. of diameter classes existing or expected, but Step 2 –  Locate a tree with a dbh close to the mean dbh with a bias towards large trees (which will value in the forest or plantation for each class. dominate biomass). Step 3 –  Harvest the selected tree and estimate the mass Step 3 – Measure dbh and height of each tree. using the dry mass estimation described in Step 4 – Harvest the selected trees to the ground. Method 1. Step 5 –  Cut the trees into appropriate sizes to directly Step 4 –  Estimate the total mass of all trees in each dbh estimate their fresh mass. class using the mass of the tree with mean dbh  f cutting a large tree trunk for weighing is not Step 6a – I and the number of trees in the dbh class. feasible, estimate the volume using data on diameter at both ends of the trunk and the length of the trunk ([Volume = π r12 + π r22 ]/2 Below is an illustrative example of the mean tree dbh method for x L), where r1 = radius at one end of the trunk, estimating aboveground biomass in moist tropical forest. r2 = radius at the other end of the trunk and L = length of the trunk Dbh Mean Mean mass No. of Total biomass STEP6b –  Collect a complete cross-sectional sample of class Dbh (cm) of tree trees/ha (dry mass fresh wood from each log, estimate the (cm) (kg/tree) kg/ha) volume, oven-dry it and measure the dry mass. Estimate the density (g/cm3) by dividing the dry mass by its volume. 5-10 8 23 5 115 STEP6c –  Estimate mass of trunk using volume and wood 10-15 12.5 73 25 1,834 density (Mass = Volume x Density) and add to 15-20 18 190 20 3,797 the other components (for example, branches, leaves, etc. ) to obtain total mass of the tree. 20-25 24 402 15 6,028 Step 7 – Develop biomass equations linking tree 25-30 28 601 8 4,805 biomass data to dbh alone, or dbh and height. >30 33 922 5 4,609 Simple equations can be created by fitting a regression line to the data in the graphing feature of Microsoft Excel. Methods for References developing the linear or non-linear biomass equations using data on dbh, height and mass of trees are given in most text books on Brown, S. 1997. Estimating biomass and biomass change of statistics or forest mensuration. Further discussion regarding tropical forests: a primer. FAO Forestry Paper 134, Rome, Italy. development of biomass equations and their use can be found in Brown (1997) and Parresol (1999). Parresol, B.R. 1999. Assessing tree and stand biomass: a review with examples and critical comparisons. Forest Science 45, 573- One of the limitations of this method is that harvesting of about 593. 30 trees of a given species may not be feasible or permitted, except for plantation species. SourceBook for Land use, land-use change and forestry Projects 41 A P P END I X c : P UB L I S HED B I OMA S S REGRE S S I ON E Q UAT I ON S Some examples of biomass equations are presented below. For more sources of equations, review: IPCC Good Practice Guidance for Land Use, Land-Use Change  and Forestry (www.ipcc-nggip.iges.or.jp/public/gpglulucf/ gpglulucf.htm) Winrock International Ecosystem Services website  (http://www.winrock.org/Ecosystems/publications.aspm) Temperate equations General Species Group Equation Source Data   Max dbh Classification originating from Hardwood General 0.5 + ((25000 x dbh2.5)/ Biomass =  Schroeder et Eastern 85.1cm (dbh2.5 + 246872)) al. (1997) USA Softwood Pine Biomass = 0.887 + ((10486 x dbh2.84) Brown and Eastern 56.1cm /(dbh2.84 + 376907)) Schroeder (1999) USA Softwood Fir/spruce Biomass = 0.357 + ((34185 x dbh2.47)/ Brown and Eastern 71.6cm (dbh2.47 + 425676)) Schroeder (1999) USA Hardwood General Biomass = Exp(-2.9132 + 0.9232 x Winrock Eastern 85.1cm ln(dbh2 x height) USA Hardwood Aspen/alder/ Biomass = Exp(-2.2094 + 2.3867 Jenkins et al. USA 70cm cottonwood/ x lndbh) (2003) willow Hardwood Soft maple/ Biomass = Exp(-1.9123 + 2.3651 Jenkins et al. USA 66cm birch x lndbh) (2003) Hardwood Mixed hardwood Biomass = Exp(-2.4800 + 2.4835 ) Jenkins et al. USA 56cm x lndbh (2003) Hardwood Hard maple/oak/ Biomass = Exp(-2.0127 + 2.4342 Jenkins et al. USA 73cm hickory / beech x lndbh) (2003) Softwood Cedar/larch Biomass = Exp(-2.0336 + 2.2592 Jenkins et al. USA 250cm x lndbh) (2003) 42 SourceBook for Land use, land-use change and forestry Projects General Species Group Equation Source Data   Max dbh Classification originating from Softwood Douglas-fir Biomass = Exp(-2.2304 + 2.4435 Jenkins et al. USA 210cm x lndbh) (2003) Softwood True fir/hemlock Biomass = Exp(-2.5384 + 2.4814 Jenkins et al. USA 230cm x lndbh) (2003) Softwood Pine Biomass = Exp(-2.5356 + 2.4349 Jenkins et al. Western 180cm x lndbh) (2003) USA Softwood Spruce Biomass = Exp(-2.0773 + 2.3323 Jenkins et al. Western USA 250cm x lndbh) (2003) Woodland Juniper/oak/ Biomass = Exp(-0.7152 + 1.7029 Jenkins et al. USA 78cm mesquite x lndbh) (2003) Hardwood Beech Biomass = Exp(-3.0366 + 2.5395) Joosten et al. Germany ~ 70cm x lndbh) (2004) Softwood Scots Pine Biomass = 0.152 x dbh2.234 Xiao and The 9.87cm Ceulemans Netherlands (2004) SourceBook for Land use, land-use change and forestry Projects 43 Tropical equations General Species Group Equation Source Data Max dbh Classification originating from Dry (900–1500mm General Biomass = 0.2035 x dbh2.3196 Brown 63cm rainfall) (unpublished) Dry (< 900mm General Biomass = 10(-0.535+log10basal area) Brown (1997) Mexico 30cm rainfall) Moist (1500–4000mm General Biomass = exp(-2.289+2.649 Brown (1997, 148cm rainfall) x lndbh-0.021 x lndbh2) updated) Wet (> 4000mm General Biomass = 21.297 – 6.953 x dbh Brown (1997) 112cm rainfall) + 0.740 x dbh2 Cecropia Cecropia species Biomass = 12.764 + 0.2588 x dbh2.0515 Winrock Bolivia 40cm Palms Palms Biomass = 6.666 + 12.826 x height0.5 Winrock Bolivia 33m (asai and pataju) x ln(height) height Palms Palms (motacu) Biomass = 23.487 + 41.851 x Winrock Bolivia 11m (ln(height))2 height Lianas Lianas Biomass = exp(0.12+0.91xlog Putz (1983) Venezuela 12cm (BA at dbh)) 44 SourceBook for Land use, land-use change and forestry Projects Agroforestry equations General Species Group Equation Source Data Max dbh Classification originating from Agroforestry All Log10Biomass = -0.834 + 2.223 (log10dbh) Segura et al. Nicaragua 44cm Shade Trees (2006) Agroforestry Inga spp. Log10Biomass = -0.889 + 2.317 (log10dbh) Segura et al. Nicaragua 44cm Shade Trees (2006) Agroforestry Inga punctata Log10Biomass = -0.559 + 2.067 (log10dbh) Segura et al. Nicaragua 44cm Shade Trees (2006) Agroforestry Inga tonduzzi Log10Biomass = -0.936 + 2.348 (log10dbh) Segura et al. Nicaragua 44cm Shade Trees (2006) Agroforestry Juglans Log10Biomass = -1.417 + 2.755 (log10dbh) Segura et al. Nicaragua 44cm Shade Trees olanchama (2006) Agroforestry Cordia alliadora Log10Biomass = -0.755 + 2.072 (log10dbh) Segura et al. Nicaragua 44cm Shade Trees (2006) Shade grown Coffea arabica Biomass = exp(-2.719 + 1.991 (ln(dbh))) Segura et al. Nicaragua 8cm coffee (log10dbh) (2006) Pruned coffee Coffea arabica Biomass = 0.281 x dbh2.06 Van Noordwijk Java, 10cm et al. (2002) Indonesia Banana Musa X paradisiaca Biomass = 0.030 x dbh2.13 Van Noordwijk Java, 28cm et al. (2002) Indonesia Peach palm Bactris gasipaes Biomass = 0.97 + 0.078 x BA – 0.00094 x BA2 Schroth Amazonia 2–12cm + 0.0000065 x BA3 et al. (2002) Rubber trees Hevea brasiliensis Biomass = -3.84 + 0.528 x BA + 0.001 x BA2 Schroth Amazonia 6–20cm et al. (2002) Orange trees Citrus sinensis Biomass = -6.64 + 0.279 x BA + 0.000514 x BA2 Schroth Amazonia 8–17cm et al. (2002) Brazil nut trees Bertholletia excelsa Biomass = -18.1 + 0.663 x BA – 0.000384 x BA2 Schroth Amazonia 8–26cm et al. (2002) SourceBook for Land use, land-use change and forestry Projects 45 References Brown, S. 1997. Estimating biomass and biomass change of tropical forests: a primer. FAO Forestry Paper 134, Rome, Italy. Brown, S.L. and P.E. Schroeder. 1999. Spatial patterns of above ground production and mortality of woody biomass for eastern US forests. Ecological Applications 9: 968-980. (errata: Brown, S.L., Schroeder, P.E. 2000. Ecological Applications 10: 937). Jenkins, J.C., D.C. Chojnacky, L.S. Heath, and R.A. Birdsey. 2003. National-scale biomass estimation for United States tree species. Forest Science 49: 12-35. Joosten, R., J. Schumacher, C. Wirth and A. Schulte. 2004. Evaluating tree carbon predictions for beech (Fagus sylvatica L.) in western Germany. Forest Ecology and Management 189: 87-96. van Noordwijk, M., S. Rahayu, K. Hairiah, Y.C. Wulan, A. Farida and B. Verbist. 2002. Carbon stock assessment for a forest-to-coffee conversion landscape in Sumber-Jaya (Lampung, Indonesia): from allometric equations to land use analysis. Science in China C 45 suppl: 75-86. Available at: http://www.globalcarbonproject.org/ PRODUCTS/Table_of_contents_land_use%20(Canadell_Zhou_ Noble2003)/Noordwijk_yc0075.pdf. Putz, F.E. 1983. Liana biomass and leaf area of a ‘Tierra Firme’ forest in the Rio Negro Basin, Venezuela. Biotropica 15: 185-189 Schroeder, P., S. Brown, J. Mo, R, Birdsey and C. Cieszewski. 1997. Biomass estimation for temperate broadleaf forests of the United States using inventory data. Forest Science 43: 424-434. Schroth, G., S.A. D’Angelo, W.G. Teixeira, D. Haag and R. Lieberei. 2002. Conversion of secondary forest to agroforestry and monoculture plantations in Amazonia: consequences for biomass, litter and soil carbon stock after 7 years. Forest Ecology and Manage- ment 163: 131-150. Segura, M., M. Kanninen and D. Suárez. 2006. Allometric models for estimating aboveground biomass of shade trees and coffe plants in agroforestry systems in Matagalpa, Nicaragua. Submitted to Agroforestry Systems. Xiao, C-W and R. Ceulemans. 2004. Allometric relationships for below- and aboveground biomass of young Scots pine. Forest Ecology and Management 203: 177-186. 46 SourceBook for Land use, land-use change and forestry Projects App e n d ix D : C h e c k lis t f o r C d m a f f o r e s tat i o n / r e f o r e s tat i o n P r o j e c t s Author: Igino M. Emmer with support from Wolfram Kägi Glossary of terms (BSS) AE Applicant Entity AR or A/R Afforestation or reforestation CDM Clean Development Mechanism CDM AR WG CDM Working Group for A/R CDM-AR-NMB CDM A/R New Baseline Methodology form CDM-AR-NMM CDM A/R New Monitoring This checklist can be used during both Project Idea Note and Methodology form Project Design Document writing stages for either small-scale or normal-sized afforestation/reforestation CDM project activities. CDM-AR-PDD CDM A/R PDD form Issues and activities for the Project Idea Note are indicated with CDM-SSC-AR-PDD CDM Small-Scale A/R PDD form an asterix; those for small-scale or normal project activities are CER Certified Emission Reduction indicated with an “S� or “N� respectively. COP Conference of the Parties to the UNFCCC Information sources, formats to be used and issues to be addressed or demonstrated are also identified in the comments DNA Designated National Authority column. In certain cases, topics are elaborated in more detail in DOE Designated Operational Entity a dedicated text box. EB Executive Board EB21 21st meeting of the Executive Board While this checklist gives general guidance to developing GHG Greenhouse gas afforestation/reforestation CDM project activities, in specific GPG Good Practice Guidance areas more detailed information is provided, based on the growing experience with the approval procedure for baseline and IPCC  Intergovernmental Panel on Climate monitoring methodologies. By no means does this checklist Change intend to cover all aspects of CDM afforestation/reforestation lCER Long-term CER project development. MA Marrakech Accords MOP  Meeting of the Parties (to the Kyoto A basic knowledge of the UNFCCC and the CDM is assumed, although references to essential documentation are also provided. Protocol) NM New methodology Main themes NMB New baseline methodology NMM New monitoring methodology 1. Capacity – knowledge of the process ODA Official Development Assistance 2. Participation requirements PDD Project Design Document 3. Baseline methodology 4. Monitoring methodology and monitoring plan PIN Project Idea Note 5. Project Design Document SSC Small scale 6. Legal issues tCER Temporary CER UNFCCC United Nations Framework Convention on Climate Change VER Verified Emission Reduction SourceBook for Land use, land-use change and forestry Projects 47 COP decisions from the checklist 11/CP.7: http://unfccc.int/resource/docs/cop7/13a01. pdf#page=54 (Land use, land-use change, and forestry) 17/CP.7: http://unfccc.int/resource/docs/cop7/13a02. pdf#page=20 (Modalities and procedures for a Clean Development Mecha- nism as defined in Article 12 of the Kyoto Protocol) 18/CP.9: http://unfccc.int/resource/docs/cop9/06a02. pdf#page=5 (Guidance to the Executive Board of the Clean Development Mechanism) 19/CP.9: http://unfccc.int/resource/docs/cop9/06a02. pdf#page=13 (Modalities and procedures for afforestation and reforestation project activities under the Clean Development Mechanism in the first commitment period of the Kyoto Protocol) 14/CP.10: http://unfccc.int/resource/docs/cop10/10a02. pdf#page=26 (Simplified modalities and procedures for small-scale afforesta- tion and reforestation project activities under the clean develop- ment mechanism in the first commitment period of the Kyoto Protocol and measures to facilitate their implementation) Checklist for Afforestation/Reforestation CDM project activities 48 2 Requirement / activity PIN 3 Scale 4 Reference 5 Sourcebook Section Comment CAPACITY – KNOWLEDGE OF THE PROCESS A/R CDM project activity cycle (general) * N/S 11/CP7; §2 http://cdm.unfccc.int/Projects/pac/index.html and (A/R) CDM modalities 17/CP.7; http://cdm.unfccc.int/Projects/pac/pac_ar.html 19/CP.9 http://unfccc.int/documentation/decisions/items/2964.php Steps towards CDM registration N/S 19/CP.9 G Text box 1 PARTICIPATION REQUIREMENTS Host country must be Party to Kyoto Protocol * N/S 17/CP.7 http://unfccc.int/files/essential_background/kyoto_ protocol/ F.30 application/pdf/kpstats.pdf DNA must have been established * N/S 17/CP.7 Text box 1 F.29 DNA must have selected a definition of ‘forest’ * N/S 19/CP.9 F §5.3 Text box 2 DNA must have formulated sustainable * N/S 17/CP.7 Text box 3 development criteria G.40 Written approval from DNA for the A/R CDM N/S 17/CP.7 Text boxes 1 and 3 project activity G.40 Voluntary participation * N/S 19/CP.9 Text box 1 G.15a ODA eligibility * N/S 17/CP.7 See ‘Baseline methodology – Additionality’ Eligibility of projects that already started N/S See ‘Project Design Document – Crediting period’ implementation Eligible A/R CDM project activities * N/S See ‘Baseline methodology’ SourceBook for Land use, land-use change and forestry Projects Eligibility of land – ‘31 December 1989 rule’ * N/S See ‘Baseline methodology’ 2 Check when completed. 3 Asterix (*) indicates that this step is required for the Project Idea Note (PIN). 4 N = normal/standard CDM project, S = small-scale project, N/S = applies to both normal and small-scale projects. 5 Reference to relevant COP decision. Requirement / activity PIN Scale Reference Sourcebook Section Comment Maximum 8000 t CO2-e/y on average over * S 19/CP.9 A.1i Text box 4 5 years Not de-bundled: multiple project sites at least * S 14/CP.10 C Text box 4 1 km apart BASELINE METHODOLOGY Check for relevant guidelines and instructions N/S http://cdm.unfccc.int/EB/Meetings; http://cdm. of EB unfccc.int/EB/Meetings/021/eb21repan18.pdf Use existing approved or nearly approved N Text box 5; http://cdm.unfccc.int/methodologies/ baseline methodologies, if appropriate ARmethodologies Use existing approved or nearly approved S Text boxes 4 and 5; http://cdm.unfccc.int/ methodologies simplified baseline methodologies for small-scale activities, if appropriate Submit for approval if new (simplified) N/S EB21; EB22 §5.6 Text box 5; CDM-AR-NM form at: methodology http://cdm.unfccc.int/EB/Meetings/022/eb22_repan14.pdf; See also: http://cdm.unfccc.int/EB/Meetings/021/eb21repan18.pdf; http://cdm.unfccc.int/methodologies/Reference/Documents Eligible A/R CDM project activities * N/S 19/CP.9 F §5.1 Text box 2 Eligibility of land – ’31 December 1989 rule’ * N 11/CP.7 §5.2 Text box 2; http://cdm.unfccc.int/EB/Meetings/022/eb22_ A.1c; EB22 repan16.pdf; Eligibility of land – ’31 December 1989 rule’ * S Text boxes 2 and 4; Attachment A in: http://cdm.unfccc.int/ Panels/ar/ARWG06_repan2_AR_SSC_Meth.pdf, Appropriate use of definitions N/S 19/CP.9 A; Text box 6; http://cdm.unfccc.int/EB/Meetings/021/ EB21 eb21repan19.pdf Accuracy, assessment of uncertainties, §6.1 Text box 6 substantiation of conservative approach Selection of baseline approach * N/S 19/CP.9 G.22 §4.2/5.4 Text box 7 Accounting for non-CO2 gases N 11/CP.7 E21; §9 Text box 8 SourceBook for Land use, land-use change and forestry Projects 19/CP.9 A1 49 50 Requirement / activity PIN Scale Reference Sourcebook Section Comment Accounting for CO2 only S Text box 4 Ecosystem compartments in soil and N 19/CP.9 G.21 §6.4/9 Text box 8 biomass included Only above- and belowground biomass included S Text box 4 Project boundary definition * N/S 19/CP.9 §6.2 See ‘Leakage’ Compliance with national policies N/S Additionality – qualitative and quantitative * N §4.1/5.3 Text box 9; Additionality tool for CDM A/R; http://cdm.unfccc. int/EB/Meetings/021/eb21repan16.pdf Additionality – qualitative and quantitative * S Text box 9; Additionality assessment method; http://cdm.unfccc. int/Panels/ar/ARWG06_repan2_AR_SSC_Meth.pdf Leakage: all sources addressed or accounted for N 19/CP.9 §4.3/11 Text box 10 Ex-post leakage as identified in simplified S Text box 4 baseline methodology Ex-ante calculation of actual net GHG removal N/S Text box 6 MONITORING METHODOLOGY AND 19/CP.9 H §6 / 7/8 MONITORING PLAN Check for relevant guidelines and instructions N/S http://cdm.unfccc.int/EB/Meetings of EB Use existing approved or nearly approved N Text box 5; http://cdm.unfccc.int/methodologies/ methodologies, if appropriate ARmethodologies Use existing approved or nearly approved S Text box 4; http://cdm.unfccc.int/methodologies/ simplified methodologies, if appropriate ARmethodologies; http://cdm.unfccc.int/Panels/ar/ARWG06_ SourceBook for Land use, land-use change and forestry Projects repan2_AR_SSC_Meth.pdf Submit for approval if new (simplified) N/S §5.6 Text box 5; CDM-AR-NM form at: methodology http://cdm.unfccc.int/EB/Meetings/022/eb22_repan14.pdf; See also: http://cdm.unfccc.int/EB/Meetings/021/eb21repan18.pdf; http://cdm.unfccc.int/methodologies/Reference/Documents Requirement / activity PIN Scale Reference Sourcebook Section Comment All greenhouse gases included N §9 Text box 8 Only carbon stock changes required S Text box 4 All ecosystem compartments in soil and N §6.4 / 9 Text box 8 biomass included Only above and belowground biomass included S Text box 4 Monitoring methodology in agreement with N/S Text box 6 baseline methodology Leakage N/S §11 Text boxes 4 and 10 Quality assurance N/S §10 Text box 6 PROJECT DESIGN DOCUMENT Check for relevant guidelines and instructions N/S http://cdm.unfccc.int/EB/Meetings of EB Use the CDM-AR-PDD form N/S http://cdm.unfccc.int/methodologies/Reference/Documents Crediting period and operational lifetime * N/S §4.4 / 5.5 Text box 11 Address participation requirements N/S See ‘Participation requirements’ Address project design N/S Address additionality N/S See ‘Baseline methodology – Additionality’ Address baseline study N/S See ‘Baseline methodology’ Address leakage N/S See ‘Baseline methodology’ Address monitoring methodology N/S See ‘Monitoring methodology’ and monitoring plan Environmental Impact Assessment N/S 17/CP.7 G.37 Must be included in the PDD SourceBook for Land use, land-use change and forestry Projects Stakeholders’ comments N/S 17/CP.7 A1; Must be included in the PDD G37; I62 51 52 Requirement / activity PIN Scale Reference Sourcebook Section Comment LEGAL ISSUES - OWNERSHIP Legal issues guidebook N/S www.uneptie.org ; www.cd4cdm.org Land titles or customary rights to the land * N/S Text box 12 Entitlement to GHG reductions / CERs (‘seller’) * N/S Text box 12 CERs versus VERs N/S Text box 12 Payment of transaction costs N/S Text box 12 Types of risks to be addressed N/S Text box 12 Liabilities and indemnities N/S Text box 12 SourceBook for Land use, land-use change and forestry Projects SourceBook for Land use, land-use change and forestry Projects 53 1. Steps towards CDM registration There are two categories of eligible A/R CDM project activities, viz. ‘afforestation’ and ‘reforestation’. Forest management or An A/R CDM project activity must be registered to be issued avoidance of deforestation are not eligible A/R CDM project CERs. For registration it has to go through the following steps: activities for the first commitment period. (17/CP.7 Art. 7a; 11/ CP.7 Annex D.12) The A/R CDM project activity has to be described using the  CDM-AR-PDD form. Afforestation is the direct human-induced conversion of land, that has not been forested for a period of at least 50 years, to  If the A/R CDM project activity does not employ approved forested land through planting, seeding and/or the human- baseline and monitoring methodologies, the new methodolo- induced promotion of natural seed sources. (11/CP.7 Annex gies must be submitted first for approval (see Text box 5). A.1b) The PDD has to be submitted to a DOE. Reforestation is the direct human-induced conversion of non- forested land to forested land through planting, seeding and/or  The DOE checks the application and the PDD against the human-induced promotion of natural seed sources, on land that CDM requirements. was forested but that has been converted to non-forest land. For the first commitment period, reforestation activities will be  The A/R CDM project activity proponent must have approval limited to reforestation occurring on those lands that did not from the host Party’s DNA. The DNA will state that the host contain forest on 31 December 1989. (11/CP.7 Annex A.1c) Party has ratified the Kyoto Protocol, assess whether project participation is voluntary and whether the A/R CDM project In practice, no distinction is made under the CDM between activity meets the sustainable development criteria (see Text afforestation and reforestation. Therefore, the criterion that all A/ box 3). The approval is required prior to registration, not R CDM project activities must meet, is no forest to be present necessarily prior to the DOE’s validation procedure. within the project boundaries between 31 December 1989 and the start of the A/R CDM project activity. The CDM EB  If the DOE determines the proposed A/R CDM project provides a tool to define the eligibility of land. (http://cdm.unfccc. activity to be valid, it submits a request to the EB for int/EB/Meetings/022/eb22_repan16.pdf ) registration of the A/R CDM project activity. This request takes the form of a validation report. In addition, the PDD In the Marrakech Accords it is stated that A/R CDM project and the host Party approval are handed in. The EB charges a activities must contribute to the conservation of biodiversity and registration fee. sustainable use of natural resources. (11/CP.7)  The COP and the EB have set deadlines for various steps in For the first commitment period, the total of additions to a the review and registration procedures. Procedures and Party’s assigned amount resulting from A/R CDM project deadlines may change. Therefore check the EB web site activities may not exceed 1% of the base year emissions (1989) of regularly. (http://cdm.unfccc.int/EB/Meetings) that Party, times 5. (17/CP.7 Art. 7b ;11/CP.7 Annex D.14) (17/CP.7 Annex G; 18/CP.9 Annex II; 19/CP.9 Annex G) 3. Sustainable development criteria 2. Definition of ‘forest’, eligible A/R CDM project One requirement of a CDM project activity is that it must activities, ‘31 December 1989 rule’ contribute to the sustainable development of the host party. The DNAs have been given the role to define criteria for sustainable The decision of what constitutes a forest has implications for development. These criteria are likely to include the following: what lands are available for afforestation and reforestation Environmental impact activities. DNAs have been given the role of deciding for their Social impact country where to lay the thresholds from the available range: Economic impact Technology transfer Minimum tree crown cover value between 10 and 30 percent Minimum land area value between 0.05 and 1 hectare Meeting these criteria will be part of the approval procedure by Minimum tree height value between 2 and 5 metres the DNA. (11/CP.7 Annex A.1a; 19/CP.9 Annex F) (17/CP.7 Annex G.40) 54 SourceBook for Land use, land-use change and forestry Projects 4. Small-scale A/R CDM project activities Land eligibility, Baseline scenario, Small-scale A/R CDM project activities may not generate more Project scenario, than a maximum of 8,000 t CO2-e/y on average over five years. Additionality, Example: assuming an average net carbon sequestration of Leakage and 10 tC/ha, this implies a maximum area of 218 ha of forest. Estimation of greenhouse gas benefits generated by the A/R  (19/CP.9 Annex A.1i) CDM project activity. Small-scale A/R CDM project activities may not be the result of A monitoring methodology describes how the GHG effects of a de-bundled larger scale activity. The three following criteria the A/R CDM project activity are to be measured / monitored. must all apply for projects to be deemed de-bundled: the same project participants, registered within the previous two years and For a new methodology to be approved, the following steps need boundaries within 1km. For example, a set of small-scale A/R to be taken: CDM project activities from the same proponent and registered at the same time should fulfil the criterion to be at least 1km The project proponent shall propose a new A/R methodology,  apart. (14CP10 Annex B.4c and App C) through a DOE or an AE. The following completed docu- ments are needed: a CDM-AR-NM (for both baseline and Indicative simplified methodologies are provided (14/CP.10 monitoring methodologies – previously there were two Appendix B) and, so far, one detailed baseline and one related separate documents, NMB and NMM; http://cdm.unfccc.int/ monitoring methodology for small-scale A/R CDM project EB/Meetings/022/eb22_repan14.pdf) and a draft CDM-AR- activities have been proposed by the AR WG, for grassland and PDD (with completed sections A-D). A methodology can be cropland to forested land. In this methodology, only carbon submitted only in combination with a concrete A/R CDM stock changes in above- and belowground biomass need to be project activity that applies the methodology. quantified and leakage can be estimated ex-post. (http://cdm. unfccc.int/Panels/ar/ARWG06_repan2_AR_SSC_Meth.pdf ) The DOE/AE and the CDM AR WG go through an  interactive reviewing process with short response times Modalities for A/R CDM project activities partly apply to small- for the project proponent. scale A/R CDM project activities (19/CP.9 1-11). For the latter, simplified modalities have been defined. (14/CP.10) The EB attributes a final rating to the methodology (A:  approval, B: resubmit – to be resubmitted with required improvements within 5 months or C: non-approval). 5. Steps towards new baseline and monitoring methodologies (19/CP.9 Annex H; http://cdm.unfccc.int/EB/Meetings/021/ eb21repan18.pdf ) Existing approved methodologies or parts of these methodologies should be used as much as possible, if applicable, to the proposed Modalities for monitoring of CDM project activities are new A/R CDM project activity, to avoid or reduce the bureauc- provided in the Marrakech Accords and COP 9 decisions. racy of getting a new methodology approved by the CDM EB. (17/CP.7 Annex H; 19/CP.9 Annex H) Submissions of different methodologies for similar A/R CDM The COP and the CDM EB have set deadlines for various steps project activities in the same country or region should be in the review and registration procedures. Procedures and avoided. deadlines may change. Therefore check the EB web site regularly. (http://cdm.unfccc.int/EB/Meetings) The PDD asks project developers to use an approved A/R methodology. Where no approved methodology exists which 6. Technical standards for documentation could be applied to the A/R CDM project activity in question, a new methodology has to be formulated and submitted through At its 21st meeting in September 2005, the CDM EB published a DOE. Once they are approved, other project developers can a second version of guidelines on formulating the A/R PDD, use them as well. A baseline methodology includes a number of NMB and NMM (Clean Development Mechanism Guidelines issues, not just the baseline (the name is thus somewhat for Completing the Project Design Document for A/R [CDM- misleading) including: AR-PDD]) (http://cdm.unfccc.int/EB/Meetings/021/eb21repan19. SourceBook for Land use, land-use change and forestry Projects 55 pdf), the proposed new methodology for A/R: Baseline (CDM- 7. Selection of baseline approach AR-NMB) and the proposed new methodology for A/R: Monitoring (CDM-AR-NMM)). The guidelines are very specific Three approaches to creating a baseline are available for selection. and give relatively clear instructions. It is strongly recommended Project developers have to select the most appropriate approach to go through this document when writing an A/R PDD and and justify their selection: NM. The CDM A/R WG expects high standards for CDM A/R documentation. This pertains to completeness, the proper use of Existing or historical, as applicable, changes in carbon stocks a)  definitions and accuracy. The above-mentioned guidelines in the carbon pools within the project boundary; include a glossary that provide guidance in using the right language for the documentation. Furthermore, it is recommend- Changes in carbon stocks in the carbon pools within the b)  ed to check and take into account information and clarifications project boundary from a land use that represents an economi- published by the CDM EB. cally attractive course of action, taking into account barriers to investment; Some specific recommendations include: Changes in carbon stocks in the pools within the project c)  Use proper definitions for additionality, leakage and project  boundary from the most likely land use at the time the A/R boundary. CDM project activity starts. Ex-ante calculations of net GHG removals must be included  (19/CP.9 Annex G.22) in the baseline methodology. It is not sufficient to define the methodology for quantifying these ex-post. The baseline scenario can either be estimated and validated upfront and then “frozen� for the first phase of the crediting The selection of the most plausible baseline scenario must be  period (30 years, or the first 20 years of up to 60 years) (19/CP.9 separated from the additionality assessment. Annex G.23), or it is also possible to monitor the baseline during the A/R CDM project activity. Make sure that the estimation of actual net GHG removals is  performed in a complete, transparent, conservative and It is advisable to define more than one alternative baseline verifiable manner. For definitions of these terms see the above- scenarios. The project scenario should at this stage be regarded as mentioned glossary. one of these scenarios. The baseline scenario is the most plausible of alternatives identified and its choice must be substantiated. Accuracy must be adequate. Quantifications (ex-ante as well as  ex-post) must be accompanied by error assessments and A baseline must be established in a transparent and conservative outcomes must be conservative. Formulae etc. must be well manner. (19/CP.9 Annex G.20) defined, contain no errors and be adequately referenced. Take note of the relevant specific guidelines from the CDM EB. 8. GHG gases and ecosystem compartments to  Quality assurance must be taken seriously. For verifiable and be considered certifiable measurements of changes in carbon stocks Two other gases besides carbon dioxide (CO2) that are related to provisions for quality assurance and quality control to be land-use change activities are methane and nitrous oxide. implemented are required, providing confidence to all Although these gases are produced in smaller quantities than stakeholders that the reported carbon credits are reliable and CO2, their effect for a given mass on global warming is greater meet minimum measurement standards. (21 and 296 times that of CO2, respectively). Methodologies must be described in a logical, step-wise ‘cook Methane and nitrous oxide are produced mainly as the result of book’ approach with unambiguous use of terminology. anthropogenic activities, for example the use of machinery, fires, the draining of wetland regions, and the fertilisation of land. Baseline and monitoring methodologies must be mutually consistent, as they must also be proposed and approved together. Methods for estimating these non-CO2 GHG emissions can be found in the IPCC Good Practice Guidance on Land Use, Land- Use Change and Forestry (2003). 56 SourceBook for Land use, land-use change and forestry Projects There are six carbon pools applicable to A/R CDM project activi- For example, leakage can be due to displaced agricultural ties: aboveground tree biomass, aboveground non-tree biomass, activities and cattle raising (CO2 and non-CO2), or due to belowground biomass, litter, dead wood and soil organic matter. displaced farmers cutting forest to replace land that is reforested (19/CP.9 Annex A.1) However, not all six pools will be signifi- as part of the project. cantly impacted in a given project. (11/CP.7 Annex E.21) Project participants may choose not to account for one or more carbon It is recommended to address leakage in the project design pools, subject to the provision of transparent and verifiable (19/CP.9 Annex G.24) or otherwise account for it by subtracting information that the choice will not increase the expected net it from the project performance. Only negative leakage (in- anthropogenic greenhouse gas removals by sinks. Therefore pools creased GHG emissions) must be included. Positive leakage can be excluded as long as it can reasonably be shown that the (reduced GHG emissions) – although a beneficial result of the pool will not decrease as part of the project activity or will not activity – may not be accounted for. increase as part of the baseline. Definitions of pools can be found in the IPCC Good Practice Guidance on Land Use, Land-Use 11. Crediting period and operational lifetime Change and Forestry (2003) (http://www.ipcc-nggip.iges.or.jp/ public/gpglulucf/gpglulucf.htm). A/R CDM project activities generate expiring CER units in two forms: tCER (temporary CERs) and lCER (long-term CERs). 9. Determination of additionality These types of CER have been instituted to address the issue of non-permanence. tCERs expire at the end of the commitment Additionality is not the mere difference between baseline and period following the one during which they were issued, that is, project scenarios. The additionality assessment is to show that they last for five years if subsequent commitment periods are five the project activity would not have occurred in the absence of years. (19/CP.9 Annex A.1g) lCERs last for the entire length of the A/R CDM project activity. (17/CP.7 Annex F.34; 19/CP.9 the crediting period. (19/CP.9 Annex A.1h) For both types of Annex G.10d) Nevertheless, there must be consistency between CERs, there is a choice between a single crediting period of a the determination of the baseline scenario (Text box 7) and the maximum of 30 years or a period of 20 years with the possibility determination of additionality. of renewal twice (totalling 60 years). These two choices must be made in the PDD. (19/CP.9 Annex A.1gh/G.23/K) The EB developed a step-wise tool to test the additionality of prospective project activities (Tool for the demonstration and Normally, the crediting period can only start after the date of assessment of additionality in A/R CM project activities – http:// registration. However, A/R CDM project activities that have cdm.unfccc.int/EB/Meetings/021/eb21repan16.pdf). This tool already started (with a start date after 1 January 2000) can covers a wide range of activities but can be adapted if need arises. register with the EB after 31 December 2005 and begin the For small-scale A/R CDM project activities, the AR WG has crediting period as early as 1 January 2000. Decisions 17/CP.7 developed a specific method for the assessment of additionality. 12 and 13 do not apply to A/R CDM project activities, as stated (http://cdm.unfccc.int/Panels/ar/ARWG06_repan2_AR_SSC_Meth. by the EB at its 21st meeting. (http://cdm.unfccc.int/EB/ pdf; Attachment B) Meetings/021/eb21rep.pdf, paras 63 and 64) Therefore, A/R CDM project activities can accumulate CERs from 1 January 2000 on Further considerations include: which can be used for compliance purposes in the commitment ODA eligibility: potential public funding for the A/R CDM  period 2008-2012. project activity from Parties in Annex I shall not be a diversion of official development assistance. (17/CP.7) The operational lifetime must be at least the same as the crediting period. The date on which the project start implementing, In case of the existence of a background reforestation or tree  resulting in the actual net GHG removal is the same as the start planting programme, the project must substantiate that there of the crediting period. will be no interference with this programme to demonstrate additionality. 12. Legal issues 10. Leakage A project developer must deal with a variety of legal issues during the project development cycle. The issues have been dealt with in Leakage is the increase in GHG emissions occurring outside the some detail in the UNEP Legal Issues Guidebook to the Clean project boundary of an A/R CDM project activity which is Development Mechanism. For the purpose of drafting a PIN, it measurable and attributable to the activity. (19/CP. Annex A.1e) is sufficient to assess land titles or customary rights to land, as SourceBook for Land use, land-use change and forestry Projects 57 this has a bearing on who will have ownership of the products of the CDM A/R project activity, depending on local legislation. In particular the following issues must not be overlooked in the PDD writing stage: Entitlement to GHG reductions/CERs: Check local legislation  to assess if the host country government has pre-existing rights on CERs or if land owners also own the CERs generated on their land. Establish who exactly is the seller of the CERs. CERs versus VERs: Establish the nature of the rights being  sold. CERs are not generated if the project fails, but in that case VERs may still be a second option. Payment of transaction costs: It must be clear who will pay for  the cost of creating CERs, including hiring a DOE, registra- tion and monitoring and verification. If these costs are not part of the CER’s price, they must be allocated to either the buyer or the seller. Types of risks to be addressed: Policy risk (political and  regulatory uncertainties in developing countries) and A/R CDM project activity risk (occurring in any kind of project) can be dealt with in contracts and are usually reflected in the purchasing price of the CERs. For example, European companies buy emission reductions from the European Emission Trading system (low risk) at a higher price than CERs from CDM projects (higher risk). Kyoto Protocol risks are specific to this legal framework and include, amongst others, unexpected changes in international agreements, opposition of NGOs, CER market risks, failing compliance with Kyoto Protocol and related rules, etc. These risks must be contractu- ally assigned. Liabilities and indemnities: Ensure that no liabilities exist that  are beyond the control of the project developer. (www.uneptie.org/energy/publications/pdfs/CDMLegalIssuesguide- book.pdf or www.cd4cdm.org/Publications/CDM%20Legal%20Issues%20Guid ebook.pdf )