E N V I R O N M E N T - - ^ ~~D E P A R T M E N T *t - PA P E RS PAPER NO. 56 - wJ TOWARD ENVIRONMENTALLY AND SOCIALLY SUSTAINABLE DEVELOPMENT ENVIRONMENTAL ECONOMICS SERIES The Use of Spatial Analysis at The World Bank Gerald C. Nelson David A. Gray September 1997 A Environmentally Sustainable Development The World Bank ESD Pollution and Environmental Economics Division The Use of Spatial Analysis at The World Bank Gerald C. Nelson David A. Gray September 1997 Papers in this series are notformal publications of the World Bank. They are circulated to encourage thought and discussion. The use and citation of this paper should take this into account. The views expressed are those of the authors and should not be attributed to the World Bank. Contents 1. Introduction 1 1.1 What are Spatial Data? 1 1-2 What is Spatial Analysis 2 1.3 The Value of Spatial Analysis 2 2. Spatial Analysis at the Bank 5 2.1 China: Land Use Planning in the Loess Plateau 5 2.2 Nepal: Effects of Alternate Access Road Location on Forest Degradation 6 2.3 Mali: Onsite Costs of Soil Erosion 9 2.4 Mexico: Predicting the Location of Air Pollution-Induced Illness in Coatzacoalcos and Minatitlan 11 2.5 Indonesia: Opportunity Cost of Protecting Biodiversity Rich Areas in Kalimantan 11 2.6 Sierra Leone: Soil Erosion Hot Spots 13 2.7 Burkina Faso: Targeting Public Works Programs 15 2.8 Chad: Effects of Woodlot Property Rights on Woodfiuel Prices and Fuel Biomass Availability 17 3. Decision Support Systems with Spatial Components 21 3.1 The Regional Air Pollution Information and Simulation Model for Asia (RAINS-ASIA) 21 3.2 Decision Support System for Industrial Pollution Control (DSS/lPC) 22 3.3 Multi-Objective Environmentally Sustainable Development Using Systems Analysis (MEDUSA) 24 4. Enhancing the Contribution of Spatial Analysis to Bank and Client Operations 27 4.1 New Approaches to Data Collection and Management 27 4.2 Development of Analytical Techniques 27 4.3 Support for Spatial Analysis in the Bank 28 5. Conclusion 29 References 31 Appendix 33 Environmental Economics Series The Use of Spatial Analysis at The World Bank Figures Figure 1: Critical Area For Strict Soil Conservation in Sijia Gou Basin 7 Figure 2: Priority Sites for Orchard Development in Sijia Gou Basin 7 Figure 3: Forest Areas of Potentially High Ecological Value Within 5 Km. of the Arun Dam Service Road 8 Figure 4: Foregone Farm Income From Erosion in the Current Year for a Given Location 9 Figure 5: Value of Yield Loss Due to Soil Erosion 10 Figure 6: Coatzacoalcos and Minatitlan Areas Under the "Plume" of SO2 and NOX Emissions from Point Sources 12 Figure 7: Preliminary Results for the Supply Curve for Biodiversity in West Kaliniantan 13 Figure 8: West Kalimantan: Preliminary Estimates of the Opportunity Cost of Biodiversity Conservation 14 Figure 9: Predicted Soil Loss Using 1986 Vegetation Cover 16 Figure 10: Biomass Around N'Djamena in 2016 With and Without the Project 18 Figure 11: Distribution of Sulfur From Sulfur Dioxide Emissions 23 Tables Table 1: Examples of Spatial Analysis 5 Table 2: Land Use Summary in 2 Km. and 5 Km. Buffer Zones on Route Alignment 9 ii Environmental Economics Series 1 Introduction Spatial analysis is necessary because location at the Bank and to describe how some recent matters. Most economic activities are location- advances in these techniques are being utilized in specific. Location provides special advantages - several on-going projects natural resource endowments or other desirable geophysical characteristics, proximity to markets After a brief introduction to spatial data and or other favorable socioeconomic attributes. spatial analysis, we provide short descriptions of Location also influences the severity of the eight examples of spatial analysis carried out at extemal effects of an activity, neighbors usually the Bank. Next, we discuss three larger modeling receiving offsite effects first. efforts designed to produce decision support systems that include a spatial component - Spatial analysis is the use of models that RAINS Asia, DSS/IPC, and MEDUSA. In the quantify the interactions of behavioral, final section, we discuss ways to improve the biological and physical relationships using contribution of spatial analysis to Bank spatial data.' The decline in price and growvth in operations focusing on trends in data collection, capacity of computing hardware and software for data management and analytical techniques. spatial analysis now makes possible analytical techniques that were impossible only a few years 1.1 What are Spatial Data? ago. Many of these techniques now have the potential to make important contributions to The key feature of spatial data is that every data World Bank operations, sector analysis and element includes location coordinates on the policy research. The purpose of this paper is to surface of the earth. These coordinates may be in highlight examples of spatial analysis carried out a variety of reference systems (e.g., latitude- longitude, UTM, state plane; see the glossary for This definition of spatial analysis is intentionally definition of these and other spatial analysis definition of spatial analysis is intentionally terms). Each data point represents a feature that broad. ESRI (a leading GIS software development company) uses a somewhat exists at the location. Data may be represented narrower definition. as points (location of a well), lImes (road), or polygons (municipal boundaries). Examples of "[Spatial analysis is] the process of modeling, data that can be spatially referenced include land examining, and interpreting model results. use (crop, forest, urban), elevation, rainfall, Spatial analysis is useful for evaluating p p p s social suitability and capability, for estimating and vraes, ' and pojects. predicting, and for interpreting and vanables, and Bank projects. understanding. There are four traditional types of spatial analysis: topological overlay and A growig number of spatially-referenced data contiguity analysis, surface analysis, linear sets are readily available through the intermet or analysis, and raster analysis." http:// by way of information initiatives involving www.esri.comnlbase/users/glossary/glossary.html. government, educational institutions, non- Environmental Economics Series The Use of Spatial Analysis at the World Bank governmental organizations and the private reform raises the price of maize. Other examples sector. This has not only dramatically reduced are mentioned below. the costs associated with spatial analysis but is also resulting in rapid improvements in data 1.3 The Value of Spatial Analysis quality - making it possible to use the approach in an ever-expanding range of applications. Spatial data and spatial analysis have much to offer the Bank and its clients throughout the 1.2 What is Spatial Analysis? project cycle. In particular, several features of this approach combine to produce benefits. These Spatial analysis generates quantitative estimates are illustrated in the examples given in Chapters of the spatial consequences of behavioral and 2 and 3: physical relationships. The results of spatial analysis are often depicted qualitatively (in * Simnulation of the effects of an activity maps). Both quantitative and qualitative facilitates low-cost evaluation of project representations can be used to reveal impacts in varying conditions. These can relationships that are not readily apparent. The range from investigation of the classic best-known set of tools for spatial analysis falls rang fro investigatio of the ssic under th rubn'c f geogrphic infrmationWhat ... if' scenario through to the use of under the rubrIc of geographic information complex models. Which forested areas are system (GIS). These tools permit manipulation of likely to be cleared following the spatial data, taking advantage of spatial construction of a road? When agricultural relationships. The standard tools of spatial subsidies are removed what locations will analysis, as implemented in a GIS, are see the largest effects on rice production and topological overlay and contiguity analysis, rural income? Which location for a water surface analysis, linear analysis, and raster treatment plant brings the greatest analysis. An example of classic GIS-based reductions in pollutant concentrations? analysis is to use overlays of several data sets to find specific locations that have key attributes. * Spatial disaggregation: Spatial analysis An example would be to identify all potential allows the investigation of policy and factory sites within 3 kilometers of a port, more project options in a spatially disaggregated than 300 meters from a riverbank, bounded by a framework utilizing both quantitative and major road, or which have compacted soil. qualitative approaches. This can enhance traditional approaches to cost/benefit More recently, it has become possible to combine anal argeti ostment ' . . . . ~~~~~~~~analys's, the targeting of investrnent, and the technical and/or behavioral relationships with investigation of tradeoffs and possible traditional spatial analysis tools and statistical mitigation measures software to simulate the effects of policy changes or project activities. Two examples illustrate this approach. A process model of industrial * The value of visualization of data and pollution is combined with spatial data on relationships is often underestimated. The population distribution around a factory and cost use of spatial data and analysis can provide of treating pollution-based health effects to a readily understood graphic representation estimate the benefits of installing pollution of project activities and their relationship to control. A model of profit-maximizing other features - impossible in a tabular agricultural producers is combined with spatial form. Examples include the sirple display data on geophysical and socioeconomic of project activity in relation to other characteristics of individual plots (e.g., soil features in the landscape, the mapped quality, distance to nearest marketing center) to display of single or multivariate data (e.g. assess the extent of deforestation if a policy population density, or estimated pollution 2 Environmental Economnics Series Introduction loads) through to 3-dimensional models of spatial analysis are being combined with watershed run-off impacts. the power of information systems to produce systems which can support day- * Monitoring - combines the benefits of to-day policy-making. quantitative and qualitative analysis to examine change over time due to project Spatial data and analysis are the ideal tools for impacts and or policy changes. investigation of many Bank activities and the Increasingly, this approach utilizes effects of policy changes. Further, the results can environmental indicators or other indexes facilitate stakeholder understanding and based on readily available data to participation. Finally, when combined with provide the basis for decision-making. information systems, these techniques can provide the basis for cost-effective monitoring . . ~~~~~~~~~and decision support. * Decision support systems - increasingly, these capabilities of Environmental Economics Series 3 The Use of Spatial Analysis at the World Bank 4 Environmental Economics Series 2 Spatial Analysis at the Bank The following examples demonstrate how spatial of local land use plans using small watersheds as data and analysis have been (or could be) used in the basis for investment planning. The various Bank operations. These are summarized preparation of these plans seemed like a natural in Table 1. application of spatial analysis technology. As a part of project preparation work, the Bank 2.1 CHINA: LAND USE PLANNING IN THE developed a prototype of a multi-purpose GIS LOESS PLATEAU that could be used as a tool for land resource assessment, evaluation of investment alternatives, The Loess Plateau region of central China covers and as a basis for monitoring project over 620 million hectares. When dry and implementation. It was originally developed for undisturbed, loess soils can sustain large stresses the Sijia Gou Basin and the following examples and nearly vertical walls more than 20 meters are from there. However, the approach is now high are common. However, when there is an widely used throughout the Plateau. increase in soil moisture content, rapid breakdown of the soil structure occurs and Chinese planners had been using manual extremely high rates of erosion are common. An techniques to identify physical land suitability for integral part of the Chinese approach to land different land uses, but these practices are time management in this region is the development consuming and error prone. With the existing data digitized, it was possible to rapidly produce Table 1: Examples of Spatial Analysis Example Bank operations in which it was (or _ might be) used China: Land use planning in the Loess Plateau project preparation Nepal: Effects of altemate access road location on forest project preparation, environmental deSgadation assessment, prect evaluation Mali: Onsite costs of soil erosion country assistance strategy, project . preparation Mexico: Predicting the location of air pollution-induced project preparation and morntoring, illness in Coatzacoalcos and Minatitlan environmental assessment Indonesia: Opportunity cost of protecting biodiversity rich project preparation, environmental areas in Kalimantan assessment Sierra Leone: Soil erosion hot spots country assistance strategy Burkina Faso: Targeting public works programs country assistance strategy, project preparation Chad: Effects of woodlot property rights on woodfuel prices project preparation, environmental and fuel biomass availability assessment Environmental Economics Series 5 The Use of Spatial Analysis at the World Bank a variety of tables and maps of characteristics land use. If the road passes through an area important to land use planning. Given the nature where soil erosion is a potential problem, of loess soil, it is critical that areas with high biodiversity is high, or habitat is important, the erosion risk are identified and erosion control new land uses could have serious negative plans implemented. With the spatially referenced environmental consequences. Spatial analysis can data it was possible to estimate the number of be used to examine alternate road alignments to hectares with high risk or erosion and a three- minimize these negative effects. dimensional map showing their locations (see The service road for the proposed Aran Dami Figures 1 and 2). eastern Nepal could potentially pass through One method of erosion control is to plant tree imnportant forest areas. Using standard GIS crops. Fruit-bearing trees have the added techniques, "impact corridors" were generated by advantage of producing a valuable crop. Chinese identifying land use areas with buffers around the specialists defined favorable areas for apple and road of 1 to 10 kIn. Both a quantitative (areas pear orchard development as being south-facing affected, see Table 2), and qualitative (mapped) and having slopes between 8 and 25 degrees. The assessment of the kinds of land use that would be GIS software was used to estimate the number of included in the buffer areas was done (see Figure hectares with those characteristics and to produce 3). In the project review, the GIS database was a map with their locations. used interactively to assess the imnpact of different road alignments on sensitive forest This use of spatial analysis did not rely on formnal areas. For example, an alternate road alignment modeling. Instead it was used as an interactive to the east (not presented here) that was at a tool for visualizing and evaluating alternate land higher elevation was also evaluated. In addition use scenarios. The planners were able to suggest to facilitating the optimal placement of roads, the potential land uses and the software was used to maps and the statistical reports allow project generate data and maps that were then used to management teams to develop site-specific refine the plans or develop new alternatives. mitigation and management plans and facilitate long-term monitoring. Potential Use/Role in Bank Operations: An interesting extension would be to model the - land resource assessment behavioral effects explicitly. Geophysical factors - evaluation of land use investment alternatives such as soil type, temperature and rainfall, and socioeconomic factors such as the location- - moioigprjcmleettospecific prices of inputs and outputs and land For more information on the spatial analysis for tenure determine land use at a particular location. the Loess plateau project see: "Using The development of a road lowers transport GeographicInformation Systems to support costs, changing relative costs. It is possible to Watershed Management: Case Studies from estimate econometrically the effects of both Nepal and China," Glenn Morgan and Petter geophysical and socioeconomic characteristics on Nyborg, Asia Technical Department, The World existing land use. The estimated parameters Bank, ITLAB Technical Paper: GIS Series could then be used to predict more precisely land Number 1, 1993. use changes as a result of building a new road and suggest alternate approaches to mitigation of 2.2 NEPAL: EFFECTS OF ALTERNATE negative effects. ACCESS ROAD LOCATION ON FOREST DEGRADATION Potential Use/Role in Bank Operations: It is frequently necessary to build service roads - identify' locations potentially affected by for infrastructure development projects. These project development- roads often provide improved access to hitherto remote regions and encourage more intensive 6 Environmental Economics Series Figrures 1 & 2 China Land Use Planning in the Loess Plateau Figure 1 - Northeast perspective view of Sijia Gou Watershed Angle above surface 30 degrees Vertical exaggeration 1.5 Figure 2 - Priority sites for orchard development in Siiia Gou Basin Observation position: Azimuth: 180 degrees (Northfacing) I Southfacing slopes between 8 Angle above surface: 30 degrees and 25 degrees (8 Hectares ) Vertical exaggeration: 1.5 The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of the World Bank Group, an judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. Figure 3 Arun Valley- Nepal Forest areas of potentially high ecological value within 5 km of the proposed Arun Dam service road Nu *. o - . , - . poseid d. site Location map China T\ lumlin k '--- ,\ India I~ ~ ~ ~ ~~i )n:Y -''' --",''u>: /~~~~~~~~~~~~ (N ,-' -- . . -5 0 5 10 Kilometers li/ .eJ /V/Approx. road alignment Drainage Source. LEMP, Govrrnment of Nepal - Forest cover with crown density Kening Earth Savices 1980-1984 greater than 40% or mature stand The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of the World Bank Group, an judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. Spatial Analysis at the Bank Table 2: Land use summary in 2 km and 5 km buffer zones on route alignment 2 km buffer 5 km buffer Total area % of total Total area % of total (hectares) (hectares) Forest 16,954.9 52.5 40,251.5 49.7 Agriculture/Grazing 12,477.7 38.7 37,330.3 46.1 Misc. non-arable land cover 2,854.6 8.8 3,389.9 4.2 Total 32,287.2 100.0 80,971.7 100.0 - model nature and extent of project effects below). Yield losses were estimated using the - simulate effects of alternate location of following relationship: infrastructure - project preparation, environmental assessment, Y = C-* project evaluation where For more information on the spatial analysis for Y = yield in tons per hectare the Arun project see: "Using Geographic C = yield on uneroded (newly cleared) land Information Systems to support Watershed = estimated coefficient varying with crop Management: Case Studies from Nepal and and slope China," Glenn Morgan and Petter Nyborg, Asia and sope Technical Department, The World Bank, ITLAB Technical Paper: GIS Series Number 1, 1993. The study used values of f that implied annual 2.3 MALI: ONSITE COSTS OF SOIL yield losses ranging from 2 to 10 percent. EROSION Combiing this information with region-specific crop budgets, the income foregone by not Soil erosion has both off-site and on-site effects. implemetis terino tehne be The off-site effects include cloggimg of irngation estimated (Figures 4 aneroS). This loss can then be channels and river channels. An irnportant on- etmtd(iue n-) hsls a hnb chnnl adrercnel. Aniprtn.n compared with the costs of vanious technologies site effect is reduced crop yield. On-site effects of tomred sil erosio.oresaroundlBa iak soIl erosion can make investments m soil conservingltechnologies pay off for farmers. A where soil erosion control is potentially profitable conserving technologies pay off for farmers. A aesae mFgr .Ms fteeaeswr are shaded in Figure 5 . Most of these areas were country- or region-wide estimate of the yield located south of Bamako. losses associated with soil erosion can guide infrastructure planners to specific locations Figure 4. Foregone farmnincome from erosion in where investments in soil conservation the current year for a given location. technologies have positive rates of return. Farm An early use of spatial analysis at the World Income Bank that combined behavioral models with GIS techniques was a study that estimated the on-site Total income costs of soil erosion in Mali. The study looked at foregone a north-south swath containing one-third of the county's most productive agricultural areas and Incomew/o erosion identified areas where soil erosion might be causing yield declines. Mean rates of soil erosion Income with erosion caused by rainfall were estimated using the y Current Begin Begin Years ofcultvation Universal Soil Loss Equation (USLE, described Y er Begin Yoww/o erosion erosion Environmental Economics Series 9 Figure 5 Mali Bamako Area VALUE OF YIELD LOSS DUE TO SOIL EROSION (based on 1985 prices, assuming a 10 year horizon and a 10% discount rate) Value of yield loss due to soil erosion iAreai0X10 0 g W _ High Low City Isohyet (rainfall) Note: For all hatched areas the value of yield loss due to soil erosion is greater than 40,000 CFA/ha The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of the World Bank Group, an judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. Spatial Analysis at the Bank Potential Use/Role in Bank Operations: Only a small part of the residential area falls under the NOX plume (Figure 6). SO2 comes - identify areas where soil conserving primarily from a single source and has a wide techniques are profitable dispersion plume, covering most industrial and - evaluate the benefits of different soil residential areas. Because of the combined erosion control technologies patterns of dispersion there is a small part of the - country assistance strategy, project residential area that falls under the pollution preparation plumes of both NOX and SO2. For more information see "The On-Site Costs of Although this study did not do so, it would be Soil Erosion in Mail," Joshua Bishop and possible to estimate the health-related economic Jennifer Allen, World Bank Environment costs associated with pollution both to industrial Department, Environment Working Paper No. workers who receive frequent exposure to both 21, November 1989. NOX and SO2 and the residents of the area in both NOX and SO2 plumes. This could be 2.4 MEXICO: PREDICTING THE LOCATION compared with the cost of retrofitting NOX and OF AIR POLLUTION-INDUCED ILLNESS IN SO2 control equipment. COATZACOALCOS AND MINATITLAN Potential Use/Role in Bank Operations: Industrial air and water pollution are responsible for health-related economic costs around the - locate areas where inhabitants are exposed to world. Reducing industrial pollution can pollutants from different sources improve health and reduce both direct health - predict health costs of exposure to pollution costs such as medical expenses for treating and compare to retrofitting pollution control pollution-induced illnesses and indirect health devices costs such as lost workdays. However, - project preparation and monitoring, retrofitting firms with pollution control environmental assessment equipment can be very expensive. Spatial analysis makes it possible to estimate the For more information see Decision Support location-specific costs and benefits of pollution Svstem for Industrial Pollution Control. Volume amelioration. 1: Methodology and Results, Iona Sebastian and Henk de Koning, May 1996, World Bank and As part of the development of the Decision WHO-Pan American Health Organization. Support System for Industrial Pollution Control (DSS/IPC; described in more detail in Section 2.5 INDONESIA: OPPORTUNITY COST OF 3.2), the locations of neighborhoods at high risk PROTECTING BIODIVERSITY RICH AREAS IN of air pollution from industrial pollution in the KALIMANTAN Coatzacoalcos and Minatitlan, Mexico areas were identified. This was done using spatial data Areas rich in plant and animal species can also sets of the location of housing, industrial, mixed generate large private profits in other uses. For use and swamp areas and locations of firms instance, they might contain high-value tree emitting various kinds of industrial pollution. species, fertile soil for annual crop production, or Using a pollution dispersion model it was be located near roads and urban areas where possible to identify locations where plumes from conversion to industrial use or housing is different factories overlapped. There are many profitable. Using spatial analysis it is possible to sources of NOX but the dispersion is limited. construct a spatial "supply curve" of biodiversity Hence, high and medium-high NOX that measures the opportunity cost of setting concentrations were confined largely to industrial aside areas rich in biodiversity. The basic areas and mainly affected industrial workers. procedure has two steps. First, evaluate each Enviromnental Economics Series 11 Figure 6 MEXICO Coatzacoalcos & Minatitlan Areas Under the "Plume" of S02 & NOX Emissions From Point Sources E Housing (1) High S02 & NOX (2) High S02, medium high NOX Industrial Areas (3) High S02, medium NOX (4) High NOX, low S02 2K Mixed Use (5) Medium high NOX, low S02 Swamp areas (6) Low NOX & S02 ~~ Swamp areas ~~(7) Outside affected area r A Point source Distance from point source The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of the World Bank Group, an judgment on the legal status of any territory, or any endorsement or acceptance of such boundafies. Spatial Analysis at the Bank location for both its biodiversity and economic With an estimate of the biodiversity supply value. Second, construct a new variable - average curve, it is possible to perform policy cost per biodiversity adjusted hectare. By sorting experiments. For example, how does the supply locations from lowest to highest average curve shift if stumpage rates or trade policies for biodiversity cost, a supply curve is traced out. annual crops are changed. In addition, the Locations plotted close to the y-axis have low location of sensitive areas under pressure can be opportunity cost of preserving biodiversity while identified (Figure 8). areas to the right have higher opportunity cost. A map of areas with low opportunity cost also Potential Use/Role in Bank Operations: shows whether they are contiguous and hence easy to protect, or scattered widely. - identify locations where the cost of conserving biodiversity is low West Kalimantan, Indonesia is widely recognized as having a very high degree of biodiversity, but - modify infrastructure location to reduce impact other economic uses of the land also exist. on areas rich in biodiversity Smallholder cultivation of tree crops and rice cultivation are important agricultural activities. - develop simulations of the consequences of A biodiversity index was constructed with data policy changes and projects for conversion of drawn primarily from interpretation of satellite areas rich in biodiversity imagery. Forty-five different land use types were identified. From these land use types it is - project preparation, environmental assessment possible to estimate the extent of species-richness For further information see, "The Economic and diversity and construct a biodiversity Supply of Biodiversity in West Kalimantan: richness variable. These values were weighted Preliminary Results," William B. Magrath, by a conservation priority index to avoid bias Charl es ults, Nallm B. agrat agans smllbut rarer areas. The opportunity Charles M. Peters, Nalin Kishor and Puneet against small bu ae ra.TeopruiyKishor, in Environmnental and Economic Issues in cost of leaving the land uncultivated was Fore m Casentudi es in determined by identifying the altemate land use Forestry: Selected Case Studies in Asia, Susan with the highest private profitability. Shen and Arnoldo Contreras-Hermosillo, editors. Prelimninary results for the biodiversity supply 2.6 SIERRA LEONE: SOIL EROSION HOT curve (Figure 7) show that for West Kalimantan, SPOTS a large fraction of the land could be set aside with little or no opportunity cost. As a country plans infrastructure development such as roads, irrigation works, or dams, it needs Figure 7: Preliminary results for the supply curve to know about possible unintended negative for biodiversity in West Kalimantan consequences of these developments. An example is soil erosion cause by road development that opens new areas to cultivation that are - E 6,000 - susceptible to erosion. Irrigation works and .> @ 4,000 ____ - IZIZIIiIIIIIIII dams can have shortened lives if eroded soil fills G 3,000 canals or lakes. .2,O000 0 0 = 15000 The Universal Soil Loss Equation (USLE) can be o 5,000 10,000 15,000 used to assess the potential for soil loss from Area (million hectares) sheet and rill erosion under a variety of geophysical and land use conditions. Environmental Economics Series 13 Figure 8 INDONESIA WEST KALIMANTAN Preliminary Estimates of the Opportunity Cost of Biodiversity Conservation < ~~~~~~~~~~~~~~~~~~~~ 1-1000 Areas having lower opportunity costs are 1001 - 2000 predominantly forested with relatively high levels of _ 2001 - 4000 biodiversity; they also possess limited capacity to Greater than 4001 support economically viable alternative land uses Outside study area The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of the World Bank Group, an judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. Spatial Analysis at the Bank The equation is: environmental action plan, project development A = RKLSCP where: For more infornation, see "Assessment of Soil Erosion in Sierra Leone," Petter A. Nyborg, A - computed soil loss per unit area Consultant, The World Bank, November 1995. R - rainfall factor K - soil erodibility factor 2.7 BLRKINA FAso: TARGETING PUBLIC L - a slope length factor WORKS PROGRAMS S - a slope degree factor C - a crop practice factor When designing projects to provide social P - a conservation practice factor. infrastructure, it helps to know where the very The USLE was oniginally developed im the U.S. poor live. Infrastructure in these communities for field-specific conditions. However, with can be subsidized while facilities in better off certain assumptions it can be used over larger communities can be developed on a cost-recovery areas. For the Sierra Leone National basis. Traditional survey methods to identify the Environmental Action Plan, the USLE was poorest communities are expensive and spatial modified to estimate predicted soil loss under analysis can provide a low-cost altemative. different cropping practices. Spatial data used include soil type, rainfall, slope and crop The principal goal of a research project practices. These data are available for most parts underway in the ISP Division of the Africa of the world. region is to develop criteria for narrow geographical targeting of public projects and The analysis generated a map of the soil loss per programs of individual villages or of clusters of unit area highlighting areas of low and high soil two or three neighboring villages. These criteria erosion potential under existing cropping will allow a comprehensive analysis of the cost- practices in 1985 (Figure 9). The highest risk of effectiveness of individual projects and the soil erosion is in the coastal delta areas and in selection of the most desirable policy parts of the northwest region where soil type and interventions. By identifying the poorest villages slope combine to increase the threat of soil the planner can achieve the maximum benefits for erosion. A second map not shown here identifies the poor. The criteria will also identify the areas of high erosion risk if the most appropriate better-off villages in which user-fees can be agricultural practices were adopted introduced. Sensitivity analysis also can be performed. For The focus on the community as the target of example, if an investment project in a particular poverty-reduction activities introduces another area will bring about a change in cropping dimension both to the analysis of geographical practices, say conversion of forest to annual targeting and the implementation of these projects crops, the effect on soil loss in that area can be and programs. Whereas geographical targeting estimated. often focuses on agroclimatic conditions, Potential Use/Role in Bank Operations: community level targeting takes into account community-specific environmental factors (such as the proximity to a source of water for drinking and/or irrigation), location factors (micluding the - simulating the impact of policy changes or proximity to the main market center and the projects on soil erosion quality of the access road), and demographic, economic and institutional factors. - identifyig areas for soil conservation project development The analysis will combine spatial, agro-climatic, - countrv assistance strategy, national and census data available for all communities Environrental Economics Series 15 Figure 9 N C!~~~~ierra e I one Predicted Soil Loss using 1 986 Vegetation Cover i n.0v 5S.cXi9^0i_004. dS ................. S.E .00|000 .g. .0~~~~~~~~~~s s AtlanticS5-;0SA.40--0-ltt00200-:;-i5 --50:0000000-ES0SS^SS-000lSU--05llSp00200A-4S-;S00W0 Ocean 0 <^0- , :000g:g : v;0->> Soil L g.g > g.g ;0;g Agg-i.g- aoss L;2g 50:;y g-gpPg t.-2. L o00...0w<.5i5 9n0 S 0 000 W: 0:t 000 0o Low toWW moderateEi WWW W WWW0k-WWWWWW0WWWWW 3W WW W- :EWWLWW.tE8M0fiz .. : 2- Mod 000S0era ite ; tj^j- -0j.tS{ C Si0S000s iW0 Mod00erat'-'i';'-fi''''l.l'^e..' to hi zghi0 High0.0 0i0.08-.0, j0 08'88@ s '0 8 t -8 t ; 4000 08j888208:j880800 8800i80008 Very high0j80 8,0s58'888888880'g8 8gi^j8S ' 80;g S 8 (, 3,08:;, E5UXXLXgti:ggt:0:00008088108StS0: Ex '0i0l00''0i00tremely' 0l0 h'0igh000 2g00 0X ' 08 2i0008 l 2 40 Y i Sii'80 tometers000lg$l8000000000800g000;0t The boundar0''X ies, co2000lorgs, eomntin andzl0 ' any othe information,' show on0 this: map8 do not imly onteprto h W rdBn .Group anl80X00 t 0'X :X: jugmn onll the legal8 status of0 an tertoy or any0 enoremn or acetac of:lg g05 : 0 suc bonai es. 0l0 88XX0008880880 l i88 Spatial Analysis at the Bank with data from the Household Income and effectively open to anyone to collect wood Expenditure Survey, the Agricultural Survey and without payment. When this situation occurs the Demographic and Health Survey which are there is no incentive to manage the wood resource only available for a sample of communities. The for higher productivity. A central feature of the use of limited dependent variable econometric Chad Household Energy Project is to grant techniques makes it possible to estimate the villages near the city of N'Djamena exclusive probability that in a given community the rights to manage and harvest woodlots in their majority of the households are relatively poor. It vicinity. The hope is that villagers will increase also makes it possible to assess quantitatively the yields above regeneration rates. The paper contribution of individual spatial and develops a model where owners are assumed to demographic characteristics to that probability. manage (spatially referenced) woodlots to maximize intertemporal profits. The model is The second part of the study will use the used to develop a woodfuel supply curve with characteristics of communities to develop and without the assignment of property rights. operational rules for determining the location of public facilities in health, education, water Spatial analysis makes it possible to derive the supply, road construction, etc. that are cost following data sets for every 1 km square around effective and reach as many of the poor as N'Djamena: possible given resource constraints. These rules will also be used to identify better-off * transport cost per ton to N'Djamena for communities and suggest guidelines for wood and charcoal establishing user-fees there. * spatially-explicit prices of wood and charcoal, with and without the project This research has already started in Burkina Faso * future land use and biomass, with and and will start soon in Cote d'Ivoire. without the project - potential yield under sustainable and P'otential Use/Role in Bank Operations:ustanbexpoain unsustainable exploitation - use small, intensive village surveys to develop * spatial distribution of income changes parameters to predict whether villages outside brought about by the-project. the sample are poor The output of the analysis is a spatial data set that predicts rights (and other project activities) - simulate effectsoaene roam ope on land use and biomass (Figure 10). By foormulaton oneffecivenes in eachig themapping the predicted biomass levels it is possible to identify areas where there are likely to - country assistance strategy, project preparation be environmental pressures (soil erosion and other nontimber woodland benefits), with and For more information contact David Bigman, ISP without the project. The spatial analysis can also Division, Afiica Region, World Bank. show how income is distributed spatially as a consequence of the project. 2.8. CHAD: EFFECTS OF WOODLOT PROPERTY RIGHTS ON WOODFUEL PRICES With the results of the analysis, several AND FUEL BIOMASS AVAILABILITY interesting policy experiments are possible. For example, it is possible to compare the effects of In many developing countries, wood and charcoal imposing a property tax or a wood tax at the are important energy sources for cooking and gates of N'Djamena on land use and wood prices. heating. Fuelwood supplies often come from open-access resources; that is, wood lots are It is also possible to estimate a rate of return on project expenditures. Based on the spatial Environmental Economics Series 17 Figure 10 Biomass around N'Djamena in 2016 with and without the project Biomass 1995 (tons/km2) Projected Biomass 2016 with the project [] 0- 100 tons/krn2 E5 100-200 tonslkrn2 O3 200-1000 tonslkrm2 * 1000+ tonslkm2 w~~~~~~ Projected Biomass 2016 without the project ~ 2~ The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of the World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. Spatial Analysis at the Bank analysis, the present discounted value of fuel - simulate effects of altemate tax and subsidy supply cost savings as a result of the project is policies for fuels about $5.9 million. The present discounted value - project preparation, environmental of project costs is about $3.9 million. assessment Potential Use/Role in Bank Operations: For further information, see "An Economic Analysis of Woodfuel Management in the Sahel: - quantify overall costs and benefits of The Case of Chad," Kenneth M. Chomitz and assigning property rights Charles Griffiths, Development Research Group, - simulate effects of different pricing regimes Policy Working Paper 1788, World Bank, June for fuels 1997. Environmental Economics Series 19 The Use of Spatial Analysis at The World Bank 20 Environmental Economics Series 3 Decision Support Systems with Spatial Components In addition to specific uses of spatial analysis m air pollutants, and their potential environmental various units around the Bank, we describe three impacts. The model was developed as an larger modeling efforts incorporating spatial international cooperative venture, involving analysis in various stages of development - scientists from Asia, Europe and North America. Regional Air Pollution Information System It is designed to run on standard IBM-compatible (RAINS-ASIA), Multi-objective environ- computers and be user-friendly. Individual mentally-sustainable development Using Systems modules can guide users through the sequence of Analysis (MEDUSA), and the Decision Support steps necessary for creating and evaluating System for Industrial Pollution (DSS/IPC). emission control plans. The RAINS-ASIA model Tlhese activities provide a generalized approach consists of three modules, each addressing a with applicability beyond a single project or different part of the acidification process. The program. Substantial effort is devoted to Regional Energy and Scenario Generator developing underlying data sets (e.g., engineering (RESGEN) module estimates energy pathways coefficients relating processes to pollution based on socioeconomic and technological outputs) and software to perform complex assumptions. The Energy and Emission (ENEM) calculations and display the results. module uses the energy scenarios to calculate sulfur emissions and costs of control strategies. 3.1. The Regional Air Pollution The Deposition and Critical Loads (DEP) module Information and Simulation Model for calculates the levels and patterns of sulfur Asia (RAINS-ASIA) deposition resulting from a given scenario, and then assesses the resulting environmental Growing concem about acid rain prompted a impacts. series of expert meetings in Asia during the late 1980s. A consensus emerged that an assessment In its current version, the model is designed to tool was needed to understand acid rain in Asia analyze emissions and impacts of sulfur dioxide and to help develop strategies to mitigate or avert at a regional level (the spatial resolution is 1 the problem. A project to develop an integrated degree). It assesses only the indirect impacts of assessment model called RAINS-ASIA (The sulfur deposition on soil. The model does not Regional Air Pollution Information and include the effect of sulfur dioxide on terrestrial Simulation Model for Asia) emerged from this ecosystems through direct exposure, or the effect consensus. on human health, aquatic ecosystems, and materials damage. RAINS-ASIA is a computerized scientific tool to help policy makers assess and project future A number of scenarios have already been trends in ermissions, transport, and deposition of analyzed using the RAINS-ASIA model. These Environmental Economics Series 21 The Use of Spatial Analysis at The World Bank are based on forecasts of future socioeconomic the "Best Available Technology" (BAT) strategy. conditions and energy demand, and predicted In this scenario, sulfur dioxide emissions are levels of energy use, emissions, and halved in 30 years, from 33.6 million tons in environmental pollution. The starting point of 1990 to 16.3 million tons by 2020. As a result, these analyses is the "base case" or the status nearly all areas of Asia attain sustainable levels quo scenario which forecasts future conditions of sulfur deposition that avoid ecosystem assuming that no changes are made in present damage, although problems still exist in areas of rates of economic and population growth or in China where there is heavy industrial activity. present economic, energy and environmental The cost of implementing the BAT strategy is policies. In the base case, total energy demand estimated at US$90 billion per year, or about 0.6 increases at an average rate of 4 percent per year percent of the region's gross domestic product over the period 1990-2020, and the relative (GDP). importance of coal in primary energy production remains relatively stable at, or near, 1990 levels The RAINS-ASIA model can be used for a of 41 percent of total fuel use. Because of the variety of purposes. Examples include: high rate of economic growth forecast for the region, sulfur emissions are projected to increase * energy and environmental planning, from 33.6 million tons in 1990 to over 110 * identifying critical ecosystems and their million tons by 2020, an increase of 230 percent, sulfur carrying capacities, if no actions are taken to restrict emissions * following emissions from an area or point (Figure 11). source to estimate deposition, 0 identifying the contributing sources to This huge increase in energy consumption and deposition in an ecosystem, sulfur dioxide emissions brings about similar * exploring different mitigation strategies and increases in sulfur deposition. Many estimating associated costs, industrialized areas of Indonesia, Malaysia, the * selecting pre-defined energy pathways; Philippines, and Thailand experience sulfur modifying pathways to explore impacts of deposition levels of 5-10 grams per square meter alternative energy development strategies, per year, while local "hot spots" in some and industrialized areas of China receive over 18 . defining control strategies for individual fuel grams of sulfur per square meter every year. In types, economic sectors, emission control comparison, the maximum levels reached in the technologies, and sub-regions or countries. most heavily polluted parts of Central and Eastern Europe -- the black triangle -- were For more information on RAINS- ASIA, contact approximately 15 grams per square meter per Jitendra Shah, (ASTEN) (20)458-1598. year. These levels resulted in the premature death of many tree species in an area covering 3.2. Decision Support System for southwest Poland, northwest Czech Republic and Industrial Pollution Control (DSS/IPC) southeast Germany. The model projects that large sections of southern and eastern China, The Decision Support System for Industrial northem and eastern India, the Korean peninsula, Pollution Control (DSS/IPC) was developed to and northern and central Thailand would receive serve a variety of operational and policy research levels of acid deposition beyond the carrying needs of the environmental and industrial capacity of the ecosystem. development community in the World Bank, in developing countries and elsewhere. While the base case scenario may be used as the worst-case scenario, since it assumes that no new The system is built from four groups of software measures are undertaken to control emissions, modules: (1) an inventory of the pollution loads one can also investigate the best-case scenario, of geographically distributed by media of discharge; 22 Enviromnental Economics Series Figure 11 Distribution of sulfur depositions from sulfur dioxide emissions in Asia Sulfur deposition in Asia, 1990 (g/m2/yr) Ambient levels of sulfur dioxide concentration in 2020 under the reference scenario =~~S W _ The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of the World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. The Use of Spatial Analysis at The World Bank (2) evaluation of the resulting ambient pollution What are the best control strategies given the concentrations; (3) the identification of priority current state of the art in pollution control pollutants, their main industrial contributors, and technologies, the degree of market concentration the available control measures and technologies; in the sectors generating the most hazardous and (4) the costs of alternative control measures pollutants, the cost structure of the industry, the and technologies, by type of pollutant and by availability of budgetary resources to implement industrial sectors and processes generating the pollution control measures, the level of risk faced pollution. The DSS/IPC software package by the most exposed and their risk preferences? includes the following databases: See Section 5.4 for an example of how DSS/IPP * emission factors by media of discharge of the can be used. most often encountered processes used in energy production, industry, agriculture and Questions or inquiries should be directed to public services, as well as of mobile sources Kseniya Lvovsky at in transportation, KLvovsky@WORLDBANK.ORG. * a comprehensive review of the most 3.3. Multi-objective environmentally- important air and water pollutants for which sustainable development Using Systems the estimated concentrations of no concern Analysis (MEDUSA) have been evaluated, and y * a list of the most commonly encountered The goal of the South Asia Development triangle pollution control measures and technologies initiative is to identify benefits to India, available in each industrial sector, their Bangladesh, Nepal and Bhutan of cooperation in potential emiission reduction effect, and the utilizing the region's natural and manmade standardized costs associated with adopting uiiigtergo' aua n amd stheacndardizdoss m asuresoc.atedth adoptingresources. A central challenge is to coordinate the control measures. information (and later projects and programs) in Whereas the RAINS Asia model is designed various sectors within and among the countries mainly for regional modeling, the DSS/IPC can and extenal lending institutions. model effects in the region of a single factory or a T D heavily industnialized area. The Decision Support Systems unit of ASTEN hea-il industrialized area. has proposed the development of MEDUSA, an Questions the DSS/IPC was designed to answer information system and a set of models, to include: examine specific resource issues. The information system will consist of a network of Which regions are most susceptible to socioeconomic, resource and development environmental pollution damage, or where are the information on the triangle, including user- most probable concentrations of pollution, given friendly, interactive spreadsheets, maps, web some preliminary comparative analysis of pages, and an inventory of useful web sites. The pollution loads; in other words, where are the first model, MEDUSA-1, will analyze the water pollution "hot-spots"? resources of the region. It is an interactive multiple-objective optimization model that In any one area, what are the most hazardous focuses on potentials and tradeoffs in water use, pollutants for the habitat? agricultural output, hydropower, in-stream water quality, ground water recharge, and policy, legal What is the most cost effective, economically and other constraints under a variety of feasible option to control the most hazardous scenarios. The objective is to make the best use pollutants? of available data to illustrate tradeoffs, identify areas of cooperation, estimate the value of additional information, and examine physical and 24 Environmental Economics Series Decision Support Systems with Spatial Components social options for the supply-side and demand- * What are the capacity constraints that hurt side for water resources planning and most in various forms of transport? management in a river basin framework. * What are the shadow prices of energy- related commodities under scenarios with The MEDUSA 2 model will look at transport different objectives, constraints, and infrastructure for various commodities including assumptions? coal, oil, hydropower and agricultural and industrial goods. The model uses an innovative Both models are currently at a nascent stage - methodology to support intersectoral further development would include the infrastructure and policy design and evaluation. participation of technical experts in the countries, It focuses on multi-commodity production, updating data, and an initial prioritization of consumption, trade, and transport issues. development constraints. Questions that MEDUSA 2 could answer are: For further information, contact Nagaraja Rao Harshadeep, (ASTEN), (202)473-9173. * What are the benefits of better connecting Nepal to the northem India power grid? Environmental Economics Series 25 The Use of Spatial Analysis at the World Bank 26 Environmental Economics Series 4 Enhancing the Contribution of Spatial Analysis to Bank and Client Operations Spatial analysis has demonstrated its value to The recent expansion of the Internet and the Bank operations in selected cases but clearly has World Wide Web offers the most exciting the potential to make a far greater contribution. potential for the mainstreaning of spatial data There are a number of developments likely to and spatial analysis. Despite difficulties accelerate the use of spatial data and analysis in associated with the large data volume of many Bank activities in the next few years. These can GIS data sets, it is now possible to make them be traced to dramatic improvements in data available anywhere in the world via an Internet availability and quality due largely to the advent connection. As this technology is implemented, it of the Internet. This is providing the stimulus for will no longer be necessary to centralize data or the production and ready exchange of spatially rely on lengthy and time-consuming delivery referenced data and increasing interest in the schedules. Data compatibility issues will also development and application of spatial analysis become less of a problem. techniques. These innovations, and the renewed interest in 4.1 New Approaches to Data Collection data exchange accompanying them, have and Management encouraged previously independent and disbursed groups of organizations to work together to By its nature, spatial data lends itself well to facilitate data harmonization and dissemination. qualitative representation in maps. However, These groups (e.g., Consortium for International maps are often difficult to incorporate in a Earth Science Information Network (CIESIN) or meaningfully way into traditional reports and are the Biodiversity Conservation Information often a static and inefficient way of conveying System (BCIS)) are improving data availability quantitative information - depending largely on and quality and are beginning to provide data of the cartographic skills of those producing them immediate utility for spatial analysis for (see Tufte, 1983 and 1990). In addition to maps, practically any area in the world. the Bank is experimenting with innovative and dynamic ways of combining qualitative and 4.2 Development of Analytical quantitative representations of data. Recent Techniques advances in information technology allow the integration of spatial data with simple interactive The history of spatial analysis has resulted in the spatial analysis tools on CD-ROMS. These can development of two different sets of techniques to be produced and disseminated at low cost - analyze spatial data - vector and raster analysis. allowing the user to investigate and manipulate Geographers were instrumental in developing the data and produce output tailored to a GIS software that manipulated data for particular demand. irregularly shaped areas such as counties, cities, and bodies of water. Traditionally GIS software Environmental Economics Series 27 The Use of Spatial Analysis at the World Bank focused on the manipulation of "vector" data; needed to test the validity of these approaches in that is, a point, line or polygon represents a a variety of settings. location. An attribute (e.g., population, and water quality) has a constant value for that area. It should be possible to develop a set of "current best practices" for spatial analysis with wide use Remote sensing experts with digital data from in Bank operations. These would include many of aerial photographs and satellite images developed the approaches used in the examples described techniques for analyzing raster data. A raster above. Other, less-well developed spatial analysis data element is a pixel, which represents an techniques could be tried on an experimental attribute value for a small, regularly shaped area basis. An important issue is how to promote (rectangle or square). Typically the data value is these best practices. For the foreseeable future it an index that represents the intensity of light of a seems likely that the level of expertise needed to given frequency received at a sensor on the manipulate the software and hardware implies satellite or airplane.1 Modem GIS software now some centralization, either for the whole Bank or incorporates both vector and raster analysis in the regions. techniques. 4.3 Support for Spatial Analysis in the The analytical techniques and computer Bank algorithms for classical GIS and remote sensing analysis are well developed. However, the The Bank has a fully operational unit capable of development of empirical behavioral models with undertaking the tasks associated with spatial spatial attributes is relatively recent. analysis, from data compilation and automation Incorporation of statistical techniques into through development and application of commercial GIS software is just beginning. techniques to the production of output (web There are especially challenging econometric pages, maps, CD-ROMS). The ENGIS unit issues that arise with the use of spatial data (see supports the activities of the GIS Working Anselin, 1988). Economic models of land use Group, a Bank-wide group of professionals that take advantage of spatial characteristics and dedicated to the development and application of can be tested econometrically have recently been spatial analysis and related technologies in Bank developed (see Chomitz and Gray, 1995, Nelson projects (see Prevost and Gilruth, 1997). Many and Hellerstein, 1997), but additional research is of the examples listed in this publication have arisen due to collaborative exercises between A commonly used and widely available source of ENGIS and staff in other units. satellite images is the Landsat Multispectral Sensor MSS). This camera was first carried aboard a Landsat satellite in 1973 and images from successive versions are available for most of the world since that time. Each sensor in the camera captures the light reflected from an area about 80 meters by 80 meters. Four frequencies are recorded - green, red and two near infrared. These frequencies were chosen because they were absorbed or reflected in different ways by chlorophyll, soil, and water. Combinations of these frequencies make it possible to identify areas of annual and perennial vegetation, urban development, and other land uses. MSS images are raster data; each pixel represents the intensity of light in a given frequency as reflected from an 80-meter square area. 28 Environmental Economics Series 5 Conclusion Spatial analysis has already made important referenced digital data sets is improving rapidly, contributions to Bank operations as the examples driven by the development of the Internet and the above illustrate. The continued decline in cost of World Wide Web. The Bank can play a key role software and hardware opens up new in hastening the contribution of spatial analysis opportunities to use these emerging techniques to to its clients by supporting developments in the improve Bank support for its clients. In addition, activities outlined above. the availabilitv of high quality, spatially Environmental Economics Series 29 The Use of Spatial Analysis at The World Bank 30 Environmental Economics Series References Anselin, Luc. Spatial Econometrics: Methods and Models. Studies in Operational Regional Science, Dordrecht: Kluwer Academic Publishers, 1988. Bishop, Joshua and Jennifer Allen, "The On-Site Costs of Soil Erosion in Mail," World Bank Environment Department, Environment Working Paper No. 21, November 1989. Chomitz, Kenneth M. and David A. Gray. "Roads, Land, Markets and Deforestation: A Spatial Model of Land Use in Belize." World Bank Economic Review, Vol. 10, No. 3, pp 487 - 512, 1995. Chomitz , Kenneth M. and Charles Griffiths, "An Economic Analysis of Woodfuel Management in the Sahel: The Case of Chad," Development Research Group, Policy Working Paper 1788, World Bank, June 1997. Magrath, William B., Charles M. Peters, Nalin Kishor and Puneet Kishor, "The Economic Supply of Biodiversity in West Kalimantan: Preliniinary Results," in Environmental and Econormic Issues in Forestrv: Selected Case Studies in Asia, Susan Shen and Arnoldo Contreras-Hermosillo, editors. Morgan, Glenn and Petter Nyborg, "Using Geographic Information Systems to support Watershed Management: Case Studies from Nepal and China," Asia Technical Department, The World Bank, ITLAB Technical Paper: GIS Series Number 1, 1993. Nelson, Gerald C. and Daniel Hellerstein. "Do Roads Cause Deforestation? Using Satellite Images in Econometric Analysis of Land Use," American Journal of Agricultural Economics, February, 1997. Nyborg, Petter A., "Assessment of Soil Erosion in Sierra Leone," Consultant, The World Bank, November 1995. Prevost, Yves A. and Peter Gilruth, "Environmental Information Systems in Sub-Saharan Africa." Building Blocks for Africa 2026, Paper No. 12, June 1997. Sebastian, Iona and Henk de Koning, Decision Support System for Industrial Pollution Control. Volume 1: Methodology and Results, May 1996, World Bank and WHO-Pan American Health Organization. Tufte, Edward R.. The Visual Display of Quantitative Information. Graphics Press, Cheshire, Connecticut. 1983. . Envisioning Information. Graphics Press, Cheshire, Connecticut. 1990. Environmental Economics Series 31 The Use of Spatial Analysis at The World Bank 32 Environmental Economics Series Appendix This glossary is derived from a GIS glossary http://www.env.gov.bc.ca/gis/glosstxt.htmn. made available on the World Wide Web by the Some terms have deleted or modified; others British Columbia Ministry of Environment, have been added. Lands, and Parks. The address is Appendix 1: A GIS Glossary Accuracy 1) The closeness of an estimated (measured or computed) value to a standard or accepted value of a particular quantity, i.e., relates to the quality of the result. 2) With regards to numbers in a mathematical table or those produced by a computer Accuracy may mean (a) the number of significant digits in the numbers, (b) the order of magnitude of the least significant digit. ( See also Precision) Algorithm A set of rules for solving a problem. An algorithm must be specified before the rules can be written in a computer language. Analog A continuously varying electronic signal (contrast with Digital). Also refers to traditional paper mapping products and aerial photographs. Arc A line connecting a set of points that make up one side of a polygon. Area A fundamental unit of geographical information. See polygon. Attribute Non-graphic information associated with a point, line, or area element in a GIS. Bit map A pattem of bits (i.e. ON/OFF) on the grid stored in memory and used to generate an image on a raster scan display. CAD(D) Computer-Aided Drafting (Design) Cadastral Related to records of land-related tenure, whether surface or sub-surface and whether linear, parcel-based or defined as a single point. Cell The basic element of spatial information in the raster (grid) description of spatial entities. Chain A sequence of coordinates defining a complex line or boundary. See Arc. Choropleth map A map consisting of areas of equal value separated bv abrupt boundaries. Contour A line connecting points of equal elevation. Environmental Economics Series 33 The Use of Spatial Analysis at the World Bank Coordinate system A system of linear or angular quantities which designate the position that a point occupies in a given reference frame or system. Also used as a general term to designate the particular kind of reference frame or system. There are several coordinate systems that can be roughly grouped into the following types: Cartesian (or plane) coordinate system, geographical (or spherical) coordinate system and rectangular coordinate systems. All of these systems establish a grid of lines and upon a two- or three-dimensional surface. See also, state plane and universal trans mercator. Database A collection of interrelated infonnation usually stored on some form of mass- storage system such as magnetic tape or disk. A GIS database includes data about the position and the attributes of geographical features that have been coded as points, lines, areas, pixels or grid cells. Digital elevation A quantitative model of lanfform in digital forn. Also digital terrain model model (DEM) (DTM). Digital The ability to represent data in discrete, quantized units or digits. Digitizer A device for entering the spatial coordinates of mapped features from a map or document to the computer. Edge matching The process of ensuring that detail along the edge of two adjacent map sheets matches correctly. Element A fundamental geographical unit of information, such as a point, line, area, or pixel. May also be known as an entity. Feature Code A set of characters (alpha, alpha/numeric or numeric) within the GIS, which uniquely identifies a feature class or homogeneous group of features. The following examples are from the TRIM 120 000 Mapping Specifications 33750000 - Wooded Area 25100190 - Paved Road 25000120 - Loose Surface Road (Gravel) Filter In raster graphics, particularly image processing, a mathematically defined operation for removing long-range (high-pass) or short range (low-pass) variation. Used for removing unwanted components from a signal or spatial pattern. Geodesy The scientific study of the size and shape of the earth and detennination of positions on it. Geodetic A spatial framework of points whose position has been precisely determined framework/ network on the surface of the earth. Geographic A system of capturing, storing, checking, integrating, analyzing and Information System displaying data about the earth that is spatially referenced. It is normally (GIS) taken to include a spatially referenced database and appropriate applications software. Geo-referencing The process of delimiting a given object, either physical (e.g. a lake) or conceptual (e.g. an administrative region), in terms of its spatial relationship to the land. The geographic reference thus established consists of points, lines, areas or volumes defined in terms of some coordinate system (usually latitude and longitude, or UTM northings and eastings, and elevation). Geocoding The activity of defining the position of geographical objects relative to a standard reference grid. 34 Environmental Economics Series Appendix Geographics or Representation of the earth's surface as a projection onto rectangular lines of geographic latitude and longitude. projection Global Positioning A system of earth satellites, each providing precise time and position System information that enables a GPS receiver to compute the distance to each satellite. The distance measurements of at least three satellites are required to fix the receivers position in latitude and longitude. Measurements from a fourth satellite are required to provide vertical (altitude) positioning. Grey scales Levels of brightness (or darkness) for displaying information on monochrome display devices. Grid 1. A network of uniformly spaced points or lines on the CRT for locating positions. 2. A set of regularly spaced sample points. 3. In cartography, an exact set of reference lines over the earth's surface. 4. In utility mapping, the distribution network of the utility resources, e.g. electricity or telephone lines. Image analysis The processing and interpretation of graphic images held in digital form. Isopleth map A map displaying the distribution of an attribute in terms of lines connecting (Isoline) points of equal value; See contour, contrast with Choropleth map. LANDSAT The generic name for a series of earth resource scanning satellites launched by the United States of America. Latitude Angular distance expressed in degrees and minutes along a meridian north or south of the equator. Legend The part of the drawn map explaining the meaning of the symbols used to code the depicted geographical elements. Longitude The angular distance east or west from a standard meridian to another meridian on the earth's surface, expressed in degrees and minutes. Line One of the basic geographical elements, defined by at least two pairs of xy coordinates. Look-up table An array of data values that can be quickly accessed by a computer program to convert data from one form to another, e.g. from attribute values to colors. Map projection The basic system of coordinates used to describe the spatial distribution of elements in a GIS Meta-Data Meta-data is data about data. It typically includes information such as currency, accuracy, extent, custodianship, and collection methodology. Meta- data is typically stored in data models, dictionaries, schemas and other representations. Multi-spectral A device often carried in airplanes or satellites for recording received radiation scanner system in several wavebands at the sarne time. (MSS) Network 1. Two or more interconnected computer systems for implementation of specific functions. 2. A set of interconnected lines (arcs, chains, strings) defining the boundaries of polygons. Node The point at which areas (lines, chains, strings) in a polygon network are joined. Nodes carry information about the topology of the polygons. Environmental Economics Series 35 The Use of Spatial Analysis at the World Bank Overlay The process of stacking digital representations of various spatial data on top of each other so that each position in the areas covered can be analyzed in terms of these data. Pixel Contraction of picture element. The smallest unit of information in a grid cell map or scanner image. Polygon A multi-sided figure representing an area on a map. Precision A measure of the tendency of a set of random numbers to cluster about a number determined by the set. The usual measure is either the standard deviation with respect to the average i.e., relates to the quality of the method by which the measurements were made and is distinguished from accuracy which relates to the quality of the result. Projection The representation on a plane surface of any part of the surface of the earth. Raster-to-vector The process of converting an image made up of cells into one described by lines and polygons. Raster A regular grid of cells covering an area. Reference Ellipsoid An ellipsoid associated with a geodetic reference system or geodetic datum, whose surface is equipotential and approximates the geoid in size and position. Reference ellipsoids are most commonly ellipsoids of revolution and are sometimes called reference spheroids. Resolution The smallest spacing between two display elements; the smallest size of feature that can be mapped or sampled. Rubber sheeting The transformation of spatial data to stretch or compress them to fit with other data Scale The relation between the size of an object on a map and its size in the real world. SPOT A series of earth resource satellites with high-resolution sensors. France launched the first in January 1986. State Plane The State Plane coordinate system (SPCS) is a rectangular system of x/y Coordinate System coordinates defined by the U.S. Geological Survey and unique to each state. Locations within a state are given in terms of their positions relative to some defined origin with that state. Coordinates are measured in feet. An example coordinate is 380,930N 1,996,821E Wisconsin South. Tesselation The process of splitting an area into tiles. Thematic map A map displaying selected kinds of inforrmation relating to specific themes, such as soil, land-use, population density, suitability for arable crops, and so on. Thematic information may be represented as labeled polygons. lines or points, chloropleth maps, isolines, etc. Tile A part of the database in a GIS representing a discrete part of the earth's surface. By splitting a study area into tiles, considerable savings in access times and improvements in system performance can be achieved. Topographic map A map showing natural and man-made features as well as relief, often in the form of contours. Topography The configuration of a planetary surface including its relief and the position of its natural and man made features. 36 Environmental Economics Series Appendix Topology The way in which geographical elements are related to each other. The topology of the data must be defined before GIS analysis can be performed. Toponymy The study of the place names of a region. A toponym is a place name. Transform The process of changing the scale, projection, or orientation of a mapped image. Universal Transverse The Universal Transverse Mercator (UTM) coordinate system is a Transverse Mercator (UTM) Mercator projection of the ellipsoid to which specific parameters have been Coordinate System applied. The UTM projection grid was adopted by the US Army in 1947 and NATO in the 1950's for designating rectangular coordinates on large-scale mnilitary maps of the world. The UTM projection maps the Earth between latitudes 84°N. and 80°S. in 60 zones, each 6° wide in longitude. The bounding meridians of each zone are divisible by 60 and each zone has a unique central meridian and zone number. Zones are numbered consecutively from 1 to 60 beginning with the zone bounded by meridians 180°W. and 174°W. and continuing eastwards. Each UTM zone has a true origin at the intersection of the central meridian and the equator and a false origin established so that grid coordinates within a zone are positive. In the northern hemisphere, the origin of grid coordinates (the false origin) is on the equator and 500,000 m west of the true origin. In the southern hemisphere, the origin of grid coordinates is 500,000 meters west and 10,000,000 south of the true origin. Units of measurement are meters. Vector graphics A means of coding line and area information in the form of units of data structure expressing magnitude, direction, and connectivity. Environmental Economics Series 37 Environment Department The World Bank 1818 H Street, N.W. Washington, D.C. 20433 202 473 3641 202 477 0565 FAX @ Printed on 100% post-consumer recycled paper