Download to read offline
Uploaded byKROTOASA FOUNDATION
Atlasafricanag all
This document provides an overview and introduction to the Atlas of African Agriculture Research & Development. It summarizes that the atlas covers over 30 topics organized under 7 themes related to smallholder agriculture in Africa. The atlas uses maps and supporting text to illustrate each topic and the underlying data. It was created through contributions from many partner organizations to provide spatial information to support agricultural policy and investment decisions with the goal of improving livelihoods. An online version is planned to allow continued contributions from the research community.
Atlasafricanag all
- 1. ATLAS OF AFRICAN AGRICULTURE RESEARCH & DEVELOPMENT
- 3. Edited by KateSebastian A peer-reviewed publication International Food Policy Research Institute Washington, DC ATLAS OF AFRICAN AGRICULTURE RESEARCH & DEVELOPMENT
- 5. ATLAS OF AFRICAN AGRICULTURE RESEARCH & DEVELOPMENT
- 6. ABOUT THE MAPS Where not otherwise cited, the administrative boundaries and names shown and the designations used on the maps are from Global Administrative Unit Layers 2013 from the Food and Agriculture Organization of the United Nations (www.fao.org/geonetwork/). The use of these data does not imply official endorsement or acceptance by the International Food Policy Research Institute; the CGIAR Consortium; the CGIAR Research Program on Climate Change, Agriculture and Food Security; the Food and Agriculture Organization of the United Nations; the Bill & Melinda Gates Foundation; or any of the contributing authors or institutions. The 2013 boundaries and names used on the maps for Ruminant Livestock and Map 2 of Statistical Groupings were provided by the World Bank’s Map Design Unit. Where not otherwise cited, a total population figure for Africa of 1.03 billion was used, based on the United Nations’s estimate for 2010 from The World Population Prospects: The 2012 Revision (http://esa.un.org/wpp/Excel-Data/population.htm). For mapping and analysis, the Global Population of the World (GPW) population data projected for 2010 from the Center for International Earth Science Information Network and the Centro Internacional de Agricultura Tropical (http://sedac.ciesin. columbia.edu/data/set/gpw-v3-centroids, accessed Feb. 4, 2014) were used. Any opinions stated herein are those of the authors and are not necessarily representative of or endorsed by the International Food Policy Research Institute or any of the partner organizations. Copyright © 2014 International Food Policy Research Institute. All rights reserved. Contact [email protected] for permission to reprint. International Food Policy Research Institute 2033 K Street, NW Washington, DC 20006-1002, USA Telephone: +1-202-862-5600 www.ifpri.org DOI: http://dx.doi.org/10.2499/9780896298460 Library of Congress Cataloging-in-Publication Data Atlas of African agriculture research and development / edited by Kate Sebastian. pages cm Includes bibliographical references. ISBN 978-0-89629-846-0 (alk. paper) 1. Agriculture--Economic aspects--Africa. 2. Agriculture--Research--Africa. 3. Geospatial data-- Africa. I. Sebastian, Katherine L. II. International Food Policy Research Institute. HD2118.A87 2014 338.1096--dc23 2014008351 COVER PHOTO CREDITS Photos (l): Globe: © Hemera-Thinkstock; and starfield: Photodisc, C. Banker/J. Reed/Thinkstock. Photos (r, top-bottom): Maize in South Africa: Panos/J. Larkin; cassava field: Media Bakery/S. Paname/ Veer; village women and livestock in Niger: ILRI/S. Mann; road market in Tanzania: Media Bakery/A. Shalamov/Veer; women’s farming cooperative in the Democratic Republic of the Congo: Panos/G. Pirozzi; grain for food, straw for feed: ILRI/Gerard; carrying bags of coffee in Uganda: Panos/S. Torfinn. Cover design: Anne C. Kerns, Anne Likes Red, Inc. Book design and layout: David Popham, IFPRI Editor: Sandra Yin, IFPRI
- 7. Contents Foreword .....................................................................vii Acknowledgments ..........................................................ix Abbreviations and Acronyms ..............................................xi Introduction ................................................................xiii Political, Demographic, and Institutional Classifications ...............1 Administrative Boundaries ........................................................2 Statistical Groupings ...............................................................4 Public Agriculture R&D Investments ..............................................6 Africa’s Agricultural Research Pool ................................................8 CGIAR Research Program on Dryland Systems ..................................10 Works Cited .......................................................................12 Footprint of Agriculture ....................................................13 Farming Systems of Africa ........................................................14 Cropland and Pastureland ........................................................16 Irrigated Areas .....................................................................18 Cereal Crops ......................................................................20 Root Crops ........................................................................22 Livestock and Mixed Crop-Livestock Systems ...................................24 Ruminant Livestock ..............................................................26 Cropping Intensity ................................................................28 Land Productivity for Staple Food Crops ........................................30 Works Cited .......................................................................32 Growing Conditions .........................................................33 Agroecological Zones .............................................................34 Climate Zones for Crop Management ...........................................36 Rainfall and Rainfall Variability ...................................................38 Soil Fertility ........................................................................40 Works Cited .......................................................................42
- 8. vi Role ofWater ................................................................43 Effects of Rainfall Variability on Maize Yields ....................................44 Blue and Green Virtual Water Flows .............................................46 Blue and Green Water Use by Irrigated Crops ...................................48 Rainfall Data Comparison ........................................................50 Works Cited .......................................................................52 Drivers of Change ...........................................................53 Influence of Aridity on Vegetation ...............................................54 Impacts of Climate Change on Length of Growing Period ......................56 Maize Yield Potential .............................................................58 Wheat Stem Rust Vulnerability ..................................................60 Benefits of Trypanosomosis Control in the Horn of Africa .....................62 Works Cited .......................................................................64 Access to Trade ..............................................................65 Market Access .....................................................................66 Accessing Local Markets: Marketsheds .........................................68 Accessing International Markets: Ports and Portsheds .........................70 Works Cited .......................................................................72 Human Welfare ..............................................................73 Severity of Hunger ................................................................74 Poverty ............................................................................76 Early Childhood Nutrition and Health ...........................................78 Works Cited .......................................................................80 About the Authors ..........................................................81 Glossary ......................................................................87
- 9. Foreword Africa isa paradox. This vast continent is home to almost half of the world’s uncultivated land fit for growing food crops—an estimated 202 million hectares—but much of it is off limits to farmers because it is difficult to farm or it is used for other purposes. Despite some recent economic successes, nearly a quarter of its population suffers from hunger, and Africa has the highest incidence of poverty in the world. It has long been recognized that Africa needs to significantly and sustainably intensify its smallholder agriculture. Low-input, low-productivity farming has failed to keep pace with food demands from a rising population. But achieving sustainable increases in smallholders’ productivity is not easy. In many areas erratic rainfall, poor soil fertility, and a lack of infrastructure and support services offer limited prospects and few incentives for poor farmers to invest in boosting productivity. Comparing and contrasting where the challenges to and opportunities for growth in productivity are located, and doing so at multiple scales and over time, can give us powerful insights that can enrich our understanding of the variables that affect agricultural productivity. The Atlas of African Agriculture Research & Development presents a broad range of geospatial data that relate to strategic agriculture policy, investment, and planning issues. The maps and analyses will give anyone who wants to learn about the role of agriculture in Africa, or find new ways to boost agricultural performance, a sense of the increasingly diverse geospatial data resources that can inform their work and guide decisionmaking on agricultural development. A better understanding of current and evolving growing conditions and how to increase productivity, despite obstacles, should aid in tailoring more pragmatic solutions for poor smallholder farmers. Shenggen Fan Director General International Food Policy Research Institute vii
- 11. Acknowledgments This atlaswas supported by funding from the Bill & Melinda Gates Foundation (BMGF), HarvestChoice, the CGIAR Consortium for Spatial Information (CGIAR-CSI), and the CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS). An atlas covering such a broad set of topical issues on agricultural research and development in Africa required the expertise of many people who generously offered their time and insights. The editor, Kate Sebastian, and publisher, the International Food Policy Research Institute (IFPRI), wish to thank the following participating authors for their contributions and work as we created, refined, and finalized the content of the atlas: Christopher Auricht at Auricht Properties, Carlo Azzarri at HarvestChoice/IFPRI, Jason Beddow at the University of Minnesota (UMN), Nienke Beintema at Agricultural Science and Technology Indicators (ASTI)/IFPRI, Chandrashekhar Biradar at the International Center for Agricultural Research in Dry Areas (ICARDA), Jean-Marc Boffa at the World Agroforestry Centre (ICRAF), Giullano Cecchi at the Food and Agriculture Organization of the United Nations (FAO), Yuan Chai at UMN, Guiseppina Cinardi at FAO, Lieven Claessens at the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Guilia Conchedda at FAO, Cindy Cox at HarvestChoice/IFPRI, John Dixon at the Australian Centre for International Agricultural Research (ACIAR), Petra Döll at Goethe University, Kathleen Flaherty at ASTI/IFPRI, Karen Frenken at FAO, Heidi Fritschel at IFPRI, Dennis Garrity at ICRAF, Marius Gilbert at Université Libre de Bruxelles, Zhe Guo at HarvestChoice/IFPRI, Jawoo Koo at HarvestChoice/IFPRI, Raffaele Mattioli at FAO, Tolulope Olofinbiyi at IFPRI, Felix Portmann at Senckenberg Research Institute, Navin Ramankutty at McGill University, Timothy Robinson at the International Livestock Research Institute (ILRI), Alexandra Shaw (consultant), Stefan Siebert at the University of Bonn, Gert-Jan Stads at ASTI/IFPRI, Philip Thornton at CCAFS/ILRI, Antonio Trabucco at Euro-Mediterranean Center on Climate Change (CCMC), Justin Van Wart at the University of Nebraska, Klaus von Grebmer at IFPRI, Doris Wiesmann (consultant), William Wint at the Environmental Research Group Oxford, Stanley Wood at BMGF, Ulrike Wood-Sichra at HarvestChoice/IFPRI, Sandra Yin at IFPRI, Yisehac Yohannes at IFPRI, and Robert Zomer at ICRAF-China. Although each map theme is attributed to the contributing authors and their respective organizations, the authors would like to acknowledge the following for helping make this atlas possible: • Terrance Hurley and Philip Pardey at UMN, and Darren Kriticos at Commonwealth Scientific and Industrial Research Organisation (CSIRO), who contributed to Wheat Stem Rust Vunerability; • Jusper Kiplimo and An Notenbaert at ILRI, who assisted with mapping and analysis related to Livestock and Mixed Crop-Livestock Systems, Rainfall and Rainfall Variability, and Impacts of Climate Change on Length of Growing Period; • Peter Jones at Waen Associates, for his input on downscaling and climate data used in Rainfall and Rainfall Variability and Impacts of Climate Change on Length of Growing Period; • Michael Bell at the International Research Institute for Climate and Society (IRI), who helped extract the Weighted Anomaly of Standardized Precipitation data used in Rainfall and Rainfall Variability; ix
- 12. • Zhe Guo,Melanie Bacou, and Joseph Green at HarvestChoice/IFPRI, who assisted with data preparation and mapping related to Early Childhood Nutrition and Health, and Poverty; • Dany Plouffe at McGill University, who assisted with data preparation and analysis related to Cropland and Pastureland; • Mohamed Fawaz Tulaymat at ICARDA, who helped prepare the Dryland Systems map; • Verena Henrich at the Institute of Crop Science and Resource Conservation, University of Bonn, for her research related to Irrigated Areas; • FAO, World Bank, International Institute for Applied Systems Analysis (IIASA), HarvestChoice, and a large number of agriculturalists, for data and input related to the Farming Systems analysis; • Michael Morris and Raffaello Cervigni of the World Bank’s Agriculture and Rural Development, Africa Region group, who collaborated on the Ruminant Livestock analysis, which was partially funded by the World Bank as part of a study on vulnerability and resilience in African drylands; and • Jeffrey Lecksell and Bruno Bonansea of the World Bank’s Map Design Unit, who provided the country, lake, and continent boundaries used in the Ruminant Livestock and World Bank income group maps. The development of this atlas has been enriched by the support and advice of Deanna Olney and the main reviewer, Gershon Feder. It also benefited from the copyediting of IFPRI’s Patricia Fowlkes and John Whitehead and proofreading by Heidi Fritschel and Andrew Marble. The atlas would not be the product it is without the valuable input of IFPRI editor, Sandra Yin; designer, David Popham; and head of publications, Andrea Pedolsky. x
- 13. Abbreviations and Acronyms ACIAR Australian Centre for International Agricultural Research AEZ agroecological zone AgGDP agricultural gross domestic product ASTI Agricultural Science and Technology Indicators CAADP Comprehensive Africa Agriculture Development Program CCAFS Climate Change, Agriculture and Food Security CERES Crop Environment Resource Synthesis CGIAR-CSI CGIAR Consortium for Spatial Information CIESIN Center for International Earth Science Information Network CIMMYT International Maize and Wheat Improvement Center CMCC Euro-Mediterranean Center on Climate Change CRU-TS data from the University of East Anglia’s Climate Research Unit Time Series CV coefficient of variation DHS Demographic and Health Surveys DSSAT Decision Support System for Agrotechnology Transfer ERGO Environmental Research Group Oxford FAO Food and Agriculture Organization of the United Nations FAOSTAT Statistics Division of the FAO and FAO’s primary portal for its statistical database FCC Soil Functional Capacity Classification System FTE full-time equivalent (refers to researchers) GADM Global Administrative Boundaries GAUL Global Administrative Unit Layers GCWM Global Crop Water Model GDD growing degree days GHI Global Hunger Index GIS Geographic Information Systems GLC2000 Global Land Cover for the year 2000 GMTS data from the University of Delaware’s Gridded Monthly Time Series GNI average national income GRUMP Global Rural-Urban Mapping Project GYGA-ED Global Yield Gap Atlas Extrapolation Domain ha hectares ICARDA International Center for Agricultural Research in Dry Areas ICRAF World Agroforestry Centre ICRISAT International Crops Research Institute for the Semi-Arid Tropics IDO intermediate development outcome IFPRI International Food Policy Research Institute IIASA International Institute for Applied Systems Analysis ILRI International Livestock Research Institute ISRIC World Soil Information kg kilograms km kilometers LGP length of growing period MIRCA monthly irrigated and rainfed crop areas mm millimeters MODIS Moderate Resolution Imaging Spectroradiometer xi
- 14. MSc master ofscience degree NEPAD New Partnership for Africa’s Development PET potential evapotranspiration pH measure of acidity PhD doctoral degree PPP purchasing power parity R&D research and development RS remote sensing RUE rain-use efficiency SPAM Spatial Production Allocation Model SSA Africa south of the Sahara UMN University of Minnesota UN United Nations UNSALB United Nations Second Administrative Level Boundaries US$ United States dollars VoP value of production WASP Weighted Anomaly of Standardized Precipitation WDI World Development Indicators WHO World Health Organization WorldClim global climate data layers xii
- 15. Introduction The Atlasof African Agriculture Research & Development is a multifaceted resource that highlights the ubiquitous nature of smallholder agriculture in Africa; the many factors shaping the location, nature, and performance of agricultural enterprises; and the strong interdependencies among farming, natural resource stocks and flows, rural infrastructure, and the well-being of the poor. Organized around 7 themes, the atlas covers more than 30 topics. Maps illustrate each topic, complemented by supporting text that discusses the content and relevance of the maps, the underlying source data, and where to learn more. The atlas is part of an eAtlas initiative that includes plans for an online, open-access resource of spatial data and tools generated and maintained by a community of research scientists, development analysts, and practitioners working in and for Africa. The atlas got its start in 2009, when Joachim von Braun, a former director general of IFPRI, was invited to head up the development of the first CGIAR Strategy and Results Framework (SRF). He asked Stanley Wood, then coordinator of the CGIAR Consortium on Spatial Information (CSI), to assemble relevant spatial data and analysis to support the analytical work of the SRF team. Wood first turned to the geographic information system (GIS) specialists at the CGIAR centers to contribute to that effort. Over time researchers at other organizations were invited to contribute. The many partners and contributors to the atlas share a belief that a better understanding of the spatial patterns and processes of agriculture research and development in Africa can contribute to better-targeted policy and investment decisions and, ultimately, to better livelihoods for the rural poor. To learn more about the eAtlas initiative, visit http://agatlas.org. xiii
- 17. ATLAS OF AFRICANAGRICULTURE RESEARCH & DEVELOPMENT POLITICAL, DEMOGRAPHIC, AND INSTITUTIONAL CLASSIFICATIONS Administrative Boundaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Statistical Groupings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Public Agriculture R&D Investments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Africa’s Agricultural Research Pool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 CGIAR Research Program on Dryland Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Works Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 1
- 18. POLITICAL, DEMOGRAPHIC, ANDINSTITUTIONAL CLASSIFICATIONS Administrative Boundaries Kate Sebastian WHAT IS THIS MAP TELLING US? The most common ways to present data for research, demo-graphic, political, and other reporting purposes is by admin-istrative unit or the unit of measure that recognizes the political boundaries and area of a country. The map shows Africa divided into nation equivalent (zero-level) units. The majority of these zero-level units represent countries that are further divided into smaller subnational (first-level) units, such as departments or states, which vary in size and num-ber per country. Drawing boundary lines is often easier said than done. Discrepancies occasionally occur due to faulty input data or, on occasion, disputed land areas. An example of this is the Hala’ib Triangle, a small area of land over which both Egypt and Sudan claim sovereignty. Most of the reporting in this atlas is done at a regional level and both Egypt and Sudan fall in the northern region so the regional reporting is not affected. Additionally, South Sudan gained its independence as a country in 2011 so it is shown separately on the maps unless the data reported are country-level statistical data that predate 2011. In such cases South Sudan is not separately designated (for example, p. 75). WHY IS THIS IMPORTANT? As more aid is dispensed and research decisions are made based on the visualization and mapping of data, it is increas-ingly important that the boundaries be both accurate and precise. In creating this atlas, for consistency’s sake, it was imperative that each map use the same administra-tive boundaries. There are a number of publicly available worldwide boundary datasets but the Food and Agriculture Organization of the United Nations’ (FAO) Global Administrative Unit Layers (GAUL) is the standard used in the atlas, because it constantly revises and updates adminis-trative boundaries to present the most up-to-date data avail-able, and it has the highest boundary accuracy rate for the developing countries of Africa (Figure 1) when compared to the Global Administrative Boundaries (GADM) and the UN’s Second Administrative Level Boundaries. Using con-sistent boundaries allows users to easily compare data by region and even identify patterns. For example, a quick look at cropland area by region (p. 16) and the average value of staple food crop production by region (p. 30) shows that southern Africa not only has the smallest share of total area devoted to cropland but also the lowest productivity. Knowing that the same boundaries were used across the maps gives the reader confidence that these values are based on the same area totals and thus can be analyzed together. WHAT ABOUT THE UNDERLYING DATA? GAUL country boundaries and disputed areas are from the UN Cartographic Section (FAO 2013). The secondary boundaries are based on information gathered from both international and national sources. Data are continuously being updated and corrected and are released yearly. The data are licensed strictly for noncommercial use by FAO, which cannot be held accountable for the accuracy, reliabil-ity, or content of the information provided. This is important to note due to the political nature of the data. Thus, by pre-senting these boundaries, FAO, and subsequently the orga-nizations involved in this atlas, are not expressing an opinion concerning the legal status of any area or its authorities or concerning the delimitation of its boundaries. WHERE CAN I LEARN MORE? GAUL 2013 boundaries: www.fao.org/geonetwork/ GADM boundaries: www.gadm.org UNSALB boundaries: http://bit.ly/RJ12kD FIGURE 1 Accuracy of different administrative boundary datasets 0 5 10 15 20 25 30 35 Number of countries with correct subnational boundaries GADM UNSALB ADM1 & GADM ADM2 GAUL ADM1 & GAUL ADM2 GADM ADM2 GAUL ADM1 Source: Adapted from Brigham, Gilbert, and Xu 2013. Note: GADM = Global Administrative Boundaries; GAUL = Global Administra-tive Unit Layers (FAO); UNSALB = UN Second Administrative Level Boundaries; ADM1 = First-level administrative boundaries; ADM2 = Second-level administrative boundaries. 2
- 19. Data source: FAO2013. ATLAS OF AFRICAN AGRICULTURE RESEARCH & DEVELOPMENT Algeria Libya Egypt Mali Mauritania Morocco Tunisia Niger Nigeria Chad Togo Benin São Tomé and Principe Seychelles Ghana Senegal Guinea e Gambia Sudan South Sudan Liberia Sierra Leone Guinea- Bissau Western Sahara Central African Republic Democratic Republic of the Congo Ethiopia Eritrea Djibouti Somalia Kenya Uganda Tanzania Rwanda Republic Burundi of Congo Equatorial Guinea Gabon Côte d’Ivoire Cameroon Angola Zambia Malawi Zimbabwe Botswana Namibia South Africa Mozambique Lesotho Swaziland Madagascar Burkina Faso Comoros Country boundaries First-level boundaries Contested areas MAP 1 Country and first-level administrative boundaries 3
- 20. POLITICAL, DEMOGRAPHIC, ANDINSTITUTIONAL CLASSIFICATIONS Statistical Groupings Stanley Wood and Kate Sebastian WHAT ARE THESE MAPS TELLING US? The agriculture research and development community makes extensive use of two primary sources of national statistics: those compiled by the Food and Agriculture Organization of the United Nations (FAO), and those compiled by the World Bank. When presenting summary statistics across countries in Africa, however, the two organi-zations use different regional aggregation approaches. FAO data, accessible through its FAOSTAT portal, are summa-rized by five geographically contiguous, subregional coun-try groupings: northern Africa, western Africa, middle Africa, eastern Africa, and southern Africa (Map 1). The World Bank on the other hand uses an income-based grouping schema for data accessible through its World Development Indicators (WDI) portal. Map 2 reflects the World Bank’s four categories of average national income per person (GNI per capita in US dollars): low ($1,025), lower middle ($1,026–$4,035), upper middle ($4,036–$12,475), and high ($12,475) income (World Bank 2013a). WHY IS THIS IMPORTANT? FAO regional aggregates better reflect similarities in agroecol-ogy, language and culture, and market integration opportu-nities across contiguous constituent countries. World Bank aggregates reflect similarity in the narrowly defined status of economic development across geographically dispersed countries (although the sources of economic growth—such as minerals or agriculture or the exploitation of other natural resources such as timber—can vary widely among countries in the same economic development category). As shown in the graphical comparison of different aggregates in Figure 1, including, for further contrast, total Africa and a split between landlocked and nonlandlocked country groupings (Map 3), different regional aggregation schema provide significantly different insights into the variation of key agricultural per-formance indicators. Not shown in the maps for reasons of scale are small African island nations, such as Cape Verde in western Africa and Reunion in southern Africa. While geo-graphically dispersed, they often face common development challenges and opportunities (for instance, limited food pro-duction potential, sea level rise, and large tourist populations). The different logical groupings of nations often translate into formal country associations that represent and promote their specific common interests. The Convention on Transit Trade of Land-locked States and the Small Island Developing States is one example. WHAT ABOUT THE UNDERLYING DATA? FAO compiles and disseminates agricultural production, consumption, price, input, land use, and related food and nutrition indicators from country-reported data, while the World Bank primarily compiles and harmonizes a broader range of cross-sectoral and macroeconomic data from FAO, the International Monetary Fund, the International Labour Organization, the World Health Organization, and other pri-mary sources. Of particular note, however, are the global responsibilities of FAO and the World Bank to derive, track, and report on the Millennium Development Goal indicators of hunger and poverty respectively. WHERE CAN I LEARN MORE? FAO’s primary data portal, FAOSTAT, including extensive metadata descriptions: http://faostat3.fao.org Other FAO reports and data sets: http://www.fao.org/publications/ The World Bank’s WDI data portal: http://bit.ly./1aS5CmL Extensive WDI poverty-specific datasets: http://bit.ly/1d7kk9U FIGURE 1 Comparing different aggregates, cereal yields and fertilizer use, 2010 0 1 2 3 4 5 6 0 2 4 6 8 10 12 Africa High income* Upper-middle income Lower-middle income Low income Western Africa Southern Africa Eastern Africa Middle Africa Nothern Africa Landlocked With coastline Cereal yield (tons per hectare) Fertilizer consumption (10 kilograms per hectare) Cereal yield Fertilizer consumption TOTAL INCOMEBASED REGIONAL LAND LOCKED Data source: FAO 2012a; FAO 2012b; World Bank 2013b. Note: Because the figure is based on values from 2010, statistics do not include South Sudan (independent since 2011), in the landlocked countries total. Cereal crops include barley, buckwheat, canary seed, fonio, maize, millet, oats, quinoa, rice, rye, sorghum, triticale, and wheat. Fertilizer consumption is based on fertilizer application for all crops. *Data unavailable. 4
- 21. Data source: Map1—FAO 2012a; Map 2—World Bank 2013b and Lecksell/World Bank 2013; Map 3—Authors. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Western Africa Northern Africa Eastern Africa Middle Africa Southern Africa Algeria Libya Egypt Mali Mauritania Morocco Tunisia Niger Nigeria Chad Benin Togo Ghana Senegal e Guinea Gambia Sudan South Sudan Liberia Sierra Leone Guinea Bissau Central African Republic Ethiopia Eritrea Somalia Djibouti Kenya Uganda Tanzania Rwanda Republic Burundi Democratic Republic of the Congo of Congo Equatorial Guinea Gabon Cameroon Angola Zambia Zimbabwe Namibia Botswana South Africa Mozambique Lesotho Swaziland Madagascar Cote d'Ivoire Burkina Faso Malawi Mali Niger Chad Ethiopia Uganda Rwanda Burundi Zambia Zimbabwe Botswana Lesotho Swaziland South Central African Sudan Republic Burkina Faso Malawi Low income Lower-middle income Upper-middle income High income Landlocked country R180 G46 B52 R154 G58 B32 R255 G204 B104 R52 G51 B22 R227 G137 B27 R198 G113 B41 MAP 1 FAO regional groups MAP 2 World Bank income groups MAP 3 Landlocked countries 5
- 22. POLITICAL, DEMOGRAPHIC, ANDINSTITUTIONAL CLASSIFICATIONS Public Agriculture RD Investments Gert-Jan Stads, Nienke Beintema, and Kathleen Flaherty WHAT ARE THESE MAPS TELLING US? Growth in public agriculture research and development (RD) spending levels in Africa south of the Sahara (SSA) varied widely from 2008 to 2011 (Map 1). Continent-wide growth was driven by a handful of larger countries. However, 13 of the 39 countries for which Agricultural Science and Technology Indicators (ASTI) data are avail-able experienced negative annual growth in public agricul-tural RD spending during 2008/09–2011.1 Another way of comparing commitment to public agricultural RD invest-ment across countries is to measure intensity (Map 2)—that is, total public agricultural RD spending as a percentage of agricultural output (AgGDP). Overall investment lev-els in most countries are still well below the levels required to sustain agricultural RD needs. In 2011, SSA as a whole invested 0.51 percent of AgGDP on average. Just 10 of the 39 countries met the investment target of one percent of AgGDP set by the African Union’s New Partnership for Africa’s Development (NEPAD). Some of the smallest coun-tries in Africa, such as Lesotho, Swaziland, Burundi, Eritrea, and Sierra Leone, have such low and declining levels of investment that the effectiveness of their national agricul-tural RD is questionable. In addition, compared with other developing regions, agricultural RD is highly dependent on funding from donor organizations and development banks such as the World Bank (Figure 1). This type of funding has been highly volatile over time, leaving research programs vul-nerable and making long-term planning difficult. WHY IS THIS IMPORTANT? A closer look at growth in public agricultural RD invest-ment levels over time reveals important cross-country differ-ences and challenges. While the intensity ratio of investment (measured as a share of AgGDP) provides a relative measure of a country’s commitment to agricultural RD, monitoring investments is also key to understanding agriculture RD’s contribution to agricultural growth. Research managers and policymakers can use agricultural RD spending information to formulate policies and make decisions about strategic plan-ning, priority setting, monitoring, and evaluation. The data are also needed to assess the progress of the Comprehensive Africa Agriculture Development Program (CAADP), which is designed to boost investments in agricultural growth through research, extension, education, and training. WHAT ABOUT THE UNDERLYING DATA? The data are from primary surveys of 39 countries in SSA conducted during 2012–2013 by ASTI and national partners. ASTI provides comprehensive datasets on agricultural RD investment and capacity trends and institutional changes in low- and middle-income countries. The datasets are updated at regular intervals and accessible online. WHERE CAN I LEARN MORE? ASTI datasets, publications, and other outputs by country: www.asti.cgiar.org/countries ASTI methodology and data collection procedures: www.asti.cgiar.org/methodology FIGURE 1 Donor funding as a share of total agriculture RD funding, 2011 0 10 20 30 40 50 60 70 80 Madagascar Mali Burkina Faso Mozambique Rwanda Eritrea Liberia Malawi Senegal e Gambia Benin Burundi Central Afr Rep Mauritania Togo Ethiopia Tanzania Uganda Mauritius Congo, Dem Rep Guinea-Bissau Guinea Kenya Côte d'Ivoire Sudan Percent Source: ASTI 2013. Note: Donor funding includes loans from development banks and funding from subregional organizations. Figure excludes countries with donor shares of less than 5 percent. 1 Due to scale, not all ASTI countries are visible on the maps. 6
- 23. Data source: ASTI2013. Notes: AgGDP=agricultural output. Intensity of agricultural RD spending=public agricultural RD spending per $100 of agricultural output. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Mauritania Mali Chad Ghana Togo Benin Senegal Gambia Guinea Sudan Liberia Ethiopia Eritrea Kenya Uganda Tanzania Rwanda Burundi Gabon Zambia Malawi Zimbabwe Botswana Namibia South Africa Mozambique Lesotho Swaziland Madagascar Côte d'Ivoire Sierra Leone Central African Republic Republic of Congo Burkina e Faso Nigeria Mauritania Mali Chad Ghana Togo Benin Senegal Gambia Guinea Sudan Liberia Ethiopia Eritrea Kenya Uganda Tanzania Rwanda Burundi Gabon Zambia Malawi Zimbabwe Botswana Namibia South Africa Mozambique Lesotho Swaziland Madagascar Côte d'Ivoire Sierra Leone Guinea- Bissau Central African Republic Republic of Congo Democratic Republic of the Congo Burkina e Faso Nigeria −3 −23 0 3 No data or non-ASTI country Annual spending growth rate (%) 0 0.5 2.00 1.00 No data or non-ASTI country $ of RD spending per $100 AgGDP INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI MAP 1 Change in public agriculture RD spending levels, 2008–2011 MAP 2 Intensity of agriculture RD spending, 2011 7
- 24. POLITICAL, DEMOGRAPHIC, ANDINSTITUTIONAL CLASSIFICATIONS Africa’s Agricultural Research Pool Nienke Beintema, Gert-Jan Stads, and Kathleen Flaherty WHAT ARE THESE MAPS TELLING US? Absolute levels of staffing in public agriculture research and development (RD) vary considerably across the 39 countries in Africa south of the Sahara participat-ing in the Agricultural Science and Technology Indicator (ASTI) survey (Map 1). In 2011, Ethiopia, Ghana, Kenya, Nigeria, South Africa, Sudan, and Tanzania each employed more than 500 full-time equivalent (FTE) researchers. In contrast, 11 countries employed fewer than 100 FTE researchers each.1 Despite recent challenges, many west-ern African countries have maintained relatively large pools of well-qualified researchers (those holding PhD and MSc degrees) (Map 2). In contrast, less than half of researchers in Botswana, the Democratic Republic of the Congo, Eritrea, Ethiopia, Guinea, Guinea-Bissau, Lesotho, Liberia, Mozambique, and Zimbabwe hold graduate degrees. Map 3 shows the number of FTE researchers per 100,000 people who are economically active in agricul-ture. While the overall average for ASTI countries is 7 FTE researchers per 100,000, only Botswana, Cape Verde, Gabon, Mauritius, Namibia, Nigeria, and South Africa each employ more than 20 FTEs per 100,000 agriculture sector workers. WHY IS THIS IMPORTANT? There is growing concern about the ability of African agricul-ture research and development (RD) systems to respond to current and emerging development challenges. Some of Africa’s smallest countries have such low, and in a few instances, declining levels of researcher numbers that the effectiveness of their national agricultural RD systems is questionable. Structural problems also persist in the age and sex composition of RD personnel (Figure 1 provides a national example), where the limitations of an aging research workforce and knowledge base are exacerbated by the low participation of female researchers (especially when com-pared to their much broader participation in the sector as farmers, farm workers, and traders). Furthermore, despite stable growth in the number of agricultural researchers, many research agencies experienced high staff turnover as a consequence, in part, of researchers retiring from the work-force (Beintema and Stads 2011). Aging scientist populations and the deterioration of average degree levels in many coun-tries imply a chronic erosion of domestic innovation capac-ity. Ongoing monitoring of national agriculture research capacity can contribute to the formulation of appropri-ate responses. WHAT ABOUT THE UNDERLYING DATA? Underlying primary data are from 39 national surveys con-ducted during 2012–2013 by the ASTI initiative and national partners. ASTI generates and curates comprehensive and comparable agriculture RD institutional, investment, and capacity data for low- and middle-income countries. The datasets are periodically updated and are accessible online. WHERE CAN I LEARN MORE? ASTI datasets, publications, and other outputs by country: www.asti.cgiar.org/countries ASTI methodology and data collection procedures: www.asti.cgiar.org/methodology Other ASTI resources: www.asti.cgiar.org/about FIGURE 1 Age and sex structure of agricultural RD staff: Senegalese Agricultural Research Institute, 2008 28 21 14 7 0 7 14 21 28 30–34 35–39 40–44 45–49 50–54 55–59 Female Male Age Number of researchers Source: Sène et al. 2012. 1 Due to scale, not all ASTI countries are visible on the maps. 8
- 25. Data source: Maps1 and 2—ASTI 2013; Map 3— ASTI 2013 and FAO 2013. Notes: Maps 1 and 3—FTE = full-time equivalent. FTE values take into account only the proportion of time spent on research and development; Map 2—Researchers with postgraduate degrees earned PhDs or MScs; Map 3—Farmers include all agricultural sector workers. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Mauritania Mali Niger Chad Ghana Togo Benin Gambia Guinea Sudan Liberia Ethiopia Eritrea Kenya Uganda Tanzania Rwanda Burundi Gabon Zambia Malawi Zimbabwe Botswana Namibia South Africa Mozambique Lesotho Swaziland Madagascar Democratic Republic of the Congo Côte d'Ivoire Sierra Leone Guinea- Bissau Central African Republic Republic of Congo Burkina e Faso Nigeria Senegal Mauritania Mali Chad Ghana Togo Benin Gambia Guinea Sudan Liberia Ethiopia Eritrea Kenya Uganda Tanzania Rwanda Burundi Gabon Malawi Zimbabwe Botswana Namibia Mozambique Lesotho Swaziland Madagascar Democratic Republic of the Congo Sierra Leone Guinea- Bissau Central African Republic Republic of Congo Burkina e Faso Nigeria Senegal Mauritania Mali Niger Chad Ghana Togo Benin Gambia Guinea Sudan Liberia Ethiopia Eritrea Kenya Uganda Tanzania Rwanda Burundi Gabon Zambia Malawi Zimbabwe Botswana Namibia South Africa Mozambique Lesotho Swaziland Madagascar Democratic Republic of the Congo Côte d'Ivoire Sierra Leone Guinea- Bissau Central African Republic Republic of Congo Burkina e Faso Nigeria Senegal 0 100 200 500 No data or non-ASTI country Total FTE researchers 0 50 75 90 No data or non-ASTI country Percent of total 2 10 20 40 No data or non-ASTI country FTEs per 100,000 farmers INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI MAP 1 Number of agricultural researchers, 2011 MAP 2 Share of agricultural researchers with postgraduate degrees, 2011 MAP 3 Concentration of agricultural researchers, 2011 9
- 26. POLITICAL, DEMOGRAPHIC, ANDINSTITUTIONAL CLASSIFICATIONS CGIAR Research Program on Dryland Systems Chandrashekhar Biradar WHAT IS THIS MAP TELLING US? The map shows the distribution of dryland agricultural production systems (also known as the CGIAR Research Program on Dryland Systems) in Africa. Dryland systems are characterized by low and erratic precipitation, persistent water scarcity, extreme climatic variability, high susceptibil-ity to land degradation, including desertification, and higher than average loss rates for natural resources, such as biodiver-sity. The lack of water is the main factor that limits profitable agricultural production. Dryland systems consist of combina-tions of plant and animal species and management practices farmers use to pursue livelihood goals based on several fac-tors including climate, soils, markets, capital, and tradition. Dryland Systems is a multidisciplinary research program that aligns the research of CGIAR research centers and partners. It aims to tackle complex development issues in two key stra-tegic research themes known as intermediate development outcomes (IDOs). The first IDO focuses on low-potential and marginal drylands and developing strategies and tools to minimize risk and reduce vulnerability. The second IDO focuses on higher-potential dryland regions and supporting sustainable intensification of agricultural production systems. Within each large target area, a number of representative action sites and complementary satellite sites serve as test sites where most of the research will be conducted. These sites—which include the Kano-Katsina-Maradi Transect in Nigeria and Niger; Wa-Bobo-Sikasso Transect in Ghana, Burkina Faso, and Mali; Tolon-K and Cinzana along West African Sahel and dryland savannas in Ghana and Mali; the Nile Delta in Egypt; Béni Khedache-Sidi Bouzid inTunisia; the Ethiopian Highlands; and Chinyanja Triangle in Malawi, Zambia, and Mozambique—were identified based on crite-ria relating to aridity index, length of growing period, market access, hunger and malnutrition, poverty, environmental risk, land degradation, and demography. WHY IS THIS IMPORTANT? The goal of the Dryland Systems research program is to iden-tify and develop resilient, diversified, and more productive combinations of crop, livestock, rangeland, aquatic, and agroforestry systems that increase productivity, reduce hun-ger and malnutrition, and improve quality of life for the rural poor. The research program aims to reduce the vulnerabil-ity of rural communities and entire regions across the world’s dry areas by sustainably improving agricultural productivity. The map provides a starting point for implementing inter-ventions for intermediate development outcomes. It also can help researchers extrapolate from the research outcomes at action sites to target areas and scale up better interventions to target regions over time. WHAT ABOUT THE UNDERLYING DATA? The Remote Sensing (RS)/Geographic Information Systems (GIS) Units of the participating CGIAR centers characterized dryland systems to delineate target areas, action sites, and complementary satellite sites, using various spatial layers, such as aridity index (p. 55), length of growing period (p. 57), access to markets (p. 66), environmental risk, land degradation, and additional criteria from regional and representative target region perspectives (CGIAR 2012). WHERE CAN I LEARN MORE? Dryland Systems: http://drylandsystems.cgiar.org ICARDA Geoinformatics: http://gu.icarda.org Dryland Systems and Other CGIAR Research Programs: http://bit.ly/1eQnJdC TABLE 1 Dryland Systems sites in Africa, 2013 Action Sites IDO1 IDO2 Area (ha) 32,861,151 60,865,568 Population 924,092 18,621,053 Households 184,818 3,724,211 Source: Author. Note: IDO = intermediate development outcomes. 10
- 27. Data source: GeoInformaticsUnit/ICARDA 2013. Note: IDO = intermediate development outcomes. Action sites are representative areas of major widespread agroecosystems where initial intervention takes place to identify best approaches and top priorities for scaling out to large areas (target regions). Satellite sites are complementary (back-up) action sites. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT IDO1 IDO2 IDO1 action site IDO1 satellite site IDO2 action site IDO2 satellite site Target areas Sites MAP 1 Dryland Systems action sites and target research areas 11
- 28. POLITICAL, DEMOGRAPHIC, ANDINSTITUTIONAL CLASSIFICATIONS Works Cited ADMINISTRATIVE BOUNDARIES Brigham, C., S. Gilbert, and Q. Xu. 2013. “Open Geospatial Data: An Assessment of Global Boundary Datasets.” The World Bank. Accessed November 11, 2013. http://bit.ly/1cxvTnE. FAO (Food and Agriculture Organization of the United Nations). 2013. Political Boundaries: Global Administrative Unit Layers (GAUL).” www.fao.org/geonetwork/. STATISTICAL GROUPINGS FAO (Food and Agriculture Organization of the United Nations). 2012a. FAOSTAT data-base. Accessed October 15, 2012. http://faostat.fao.org/site/291/default.aspx. —. 2012b. FAOSTAT database. Accessed December 2, 2012. http://faostat.fao.org/site/291/default.aspx. Lecksell, J./ World Bank. 2013. Personal communication regarding world boundaries for 2013, Nov. 26. Lecksell is lead World Bank cartographer. World Bank. 2013a. World Development Indicators 2013. Washington, DC: World Bank. http://bit.ly/1pW4Rzc. —. 2013b. World Development Indicators (Income Levels). Accessed December 22, 2013. http://bit.ly/1aS5CmL. PUBLIC AGRICULTURE RD INVESTMENTS ASTI (Agricultural Science Technology Indicators). 2013. “ASTI Countries.” Accessed October 14, 2013. www.asti.cgiar.org/countries. AFRICA’S AGRICULTURAL RESEARCH POOL ASTI (Agricultural Science and Technology Indicators). 2013. ASTI database. Accessed January 17, 2013. www.asti.cgiar.org/data/. Beintema, N., and G.-J. Stads. 2011. African Agricultural RD in the New Millennium: Progress for Some, Challenges for Many. Food Policy Report 24. Washington, DC: International Food Policy Research Institute. FAO. 2013. FAOSTAT Database. http://faostat.fao.org/site/291/default.aspx. Sène, L., F. Liebenberg, M. Mwala, F. Murithi, S. Sawadogo, and N. Beintema. 2011. Staff Aging and Turnover in African Agricultural RD: Lessons from Five National Agricultural Research Institutes. Conference Working Paper No. 17. Washington, DC: International Food Policy Research Institute and Forum for Agricultural Research in Africa. CGIAR RESEARCH PROGRAM ON DRYLAND SYSTEMS CGIAR. 2012. “Research Program on Dryland Systems.” Accessed November 19, 2013. http://drylandsystems.cgiar.org. ICARDA (International Center for Agricultural Research in the Dry Areas). 2013. GeoInformatics Unit. http://gu.icarda.org. 12
- 29. ATLAS OF AFRICANAGRICULTURE RESEARCH DEVELOPMENT Footprint of Agriculture Farming Systems of Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Cropland and Pastureland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Irrigated Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Cereal Crops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Root Crops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Livestock and Mixed Crop-Livestock Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Ruminant Livestock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Cropping Intensity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Land Productivity for Staple Food Crops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Works Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 13
- 30. FOOTPRINT OF AGRICULTURE FIGURE 1 Rural poor living on ≤ $1.25/day by farming system, Africa south of the Sahara, 2010 0 10 20 30 40 50 60 Arid pastoral-oases Perennial mixed Irrigated Forest-based Fish-based Pastoral Humid lowland tree crop Highland mixed Cereal-root crop mixed Root and tuber crop Highland perennial Agropastoral Maize mixed People (millions) Data source: Dixon, Boffa, and Garrity 2014; Azzarri et al. 2012; UN 2013. Note: See glossary for definitions of specific farming systems. Poverty data calibrated to 2010. Farming Systems of Africa Christopher Auricht, John Dixon, Jean-Marc Boffa, and Dennis Garrity WHAT IS THIS MAP TELLING US? Populations within the same farming system share similar farming practices and livelihood strategies. As the map shows, many farming systems in Africa exhibit a strong geographical pattern, extending across northern Africa and Africa south of the Sahara (SSA), reflecting a mix of factors, including climate, soils, and markets. In SSA, 16 percent of land area is dominated by the maize mixed farming system, mostly in the eastern, central, and southern regions. This farming system is home to nearly 100 million rural people, of whom 58 million live on less than $1.25 a day (Figure 1), represent-ing 23 percent of the total rural poor in SSA. The highland areas of eastern and southern Africa feature smaller frag-mented systems, such as the highland perennial and high-land mixed systems that cover just 2 percent of the area, but are home to 11 and 6 percent, respectively, of SSA rural poor. A large share of the rural poor live in the agropastoral farm-ing system (18 percent), root and tuber crop system (11 per-cent), and cereal-root crop mixed system (10 percent), which combined cover more than one-third of SSA’s area. WHY IS THIS IMPORTANT? Broadly similar farming systems share recognizable livelihood patterns and similar development pathways, infrastruc-ture, and policy needs. Delineating major farming systems provides a framework to guide the development and tar-geting of strategic agricultural policies and interventions to reduce poverty and promote the adoption of more sustain-able land use practices. This classification can help policy-makers and scientists target institutional innovations and technologies to specific farming systems, thereby focusing planning, policies, and research. In this respect, high poten-tial farming systems with good market access might benefit most from improved maize, cowpeas, and dairy, while drier areas might benefit from improved sorghum, millet, and live-stock, because these contrasting farming systems offer differ-ent ways to improve livelihoods. Similarly, fertilizer policies should take into account the different nutrient requirements and markets of various crops in different farming systems. WHAT ABOUT THE UNDERLYING DATA? Farming systems are defined based on: available natu-ral resources (including water, land area, soils, elevation, and length of growing period); population; cropping and pasture extent; the dominant pattern of farm activities and household livelihoods; and access to markets. The spatial characterization of African farming systems used data on agroecological and socioeconomic variables. The two main spatial variables were length of growing period (FAO/IIASA 2012) and distance to markets (HarvestChoice 2011; Map 1, p. 67), supplemented by data on population and poverty, elevation, soils and irrigation, crop and livestock patterns, productivity, and change over time (Dixon et al. 2014; FAO 2013a–e). A multidisciplinary team of experts for each farm-ing system identified system characteristics, emergent prop-erties, drivers of change and trends, and priorities. This work updates and expands the analysis of the African farming sys-tems in the World Bank and FAO farming systems and pov-erty assessment (Dixon et al. 2001). WHERE CAN I LEARN MORE? Farming Systems and Poverty: Improving Farmers’ Livelihoods in a Changing World. Dixon et al. 2001. http://bit.ly/1dDekBW Understanding African Farming Systems: Science and Policy Implications. Food Security in Africa: Bridging Research into Practice. Garrity et al. 2012. http://bit.ly/1h8lmGJ 14
- 31. Data source: Dixon,Boffa, and Garrity 2014. Note: See glossary for definitions of specific farming systems. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Maize mixed Agropastoral Highland perennial Root and tuber crop Cereal-root crop mixed Highland mixed Humid lowland tree crop Pastoral Fish-based Forest-based Irrigated Perennial mixed Arid pastoral-oasis North Africa dryland mixed North Africa rainfed mixed North Africa highland mixed MAP 1 Farming systems of Africa 15
- 32. FOOTPRINT OF AGRICULTURE Cropland and Pastureland Navin Ramankutty WHAT ARE THESE MAPS TELLING US? Map 1 shows the extent of cropland, and Map 2 shows the extent of pastureland circa 2000. The values are presented as a percentage of each ~100 km2 grid cell. Pastureland cov-ers one-quarter of the African continent (Table 1) and domi-nates the landscape in the Sahel and Sudano-Sahelian regions in the west, the Maghreb, much of eastern and southern Africa, and western Madagascar. The only portions of the continent not grazed are those that are too hot and too dry, such as the Sahara, and the tropical rain forests of the Congo Basin. Cropland covers approximately 7 percent of the con-tinent. Western Africa has the greatest proportion at 39 per-cent. High concentrations of cropland (60≤ percent) can be found along the Mediterranean coast in the Nile Valley, Nigeria, the Ethiopian highlands, the Rift Valley north and west of Lake Victoria, and South Africa near Cape Town and north of Lesotho. Low-to-moderate cropland intensity (20–60 percent) extends from Nigeria to Senegal and can be found in parts of Sudan, and scattered throughout southeast-ern Africa. WHY IS THIS IMPORTANT? These maps of cropland and pastureland provide critical pieces of information used to analyze food security and agriculture’s environmental impact. More accurate assess-ments of the land under cultivation and areas potentially available for expansion could help improve food security. For instance, in Africa south of the Sahara—one of the only regions in the world where increases in food production have not kept pace with population growth—the land area suitable for cultivation is estimated to be nearly five times what is currently in production. Knowledge of pasturelands is similarly vital to food security because livestock provide not only a source of food but also income, insurance, soil nutrients, employment, traction (for instance, plowing), and clothing (Thornton and Herrero 2010). However, both grazing and planting also contribute to environmental deg-radation (Foley et al. 2005) and already have modified a large part of the African continent. Overgrazing contrib-utes to land degradation, further diminishing soil health, plant productivity and diversity, and by extension, livestock production. Grazing is also a significant source of methane emissions, a potent greenhouse gas that contributes to cli-mate change. WHAT ABOUT THE UNDERLYING DATA? The distribution and intensity of croplands and pastures are expressed as a percentage of the area within each ~100 km2 grid cell. The maps represent “arable land and per-manent crops” and “permanent meadows and pastures,” respectively, as defined by the Food and Agriculture Organization of the United Nations (FAO 2013). Data for both maps derive from integrating administrative-level agricultural sta-tistics with global land cover classification data from satel-lite remote sensing using a statistical data fusion method (Ramankutty et al. 2008). The agricultural statistics for Africa came mainly from FAO’s national statistics (FAOSTAT 2012), supplemented with subnational statistics for Nigeria and South Africa. Two different sources of satellite-based land cover classification data were merged: the MODIS land cover dataset from Boston University (Friedl et al. 2010) and the GLC2000 dataset (Bartholomé and Belward 2005) from the European Commission (both at 1 km spatial resolution). WHERE CAN I LEARN MORE? Download crop and pasture data at EarthStat: www.earthstat.org “Farming the Planet. Part 1: The Geographic Distribution of Global Agricultural Lands in the Year 2000.” Ramankutty et al. 2008: http://bit.ly/1ctE7Nf TABLE 1 Cropland and pastureland by region, c. 2000 Crop Area Pasture area Total area Region (000 sq km) Share of total (%) (000 sq km) Share of total (%) (000 sq km) Share of total (%) Eastern Africa 501 23.4 2,404 31.3 6,172 20.9 Middle Africa 249 11.6 1,144 14.9 6,448 21.8 Northern Africa 374 17.5 1,603 20.9 8,266 27.9 Southern Africa 172 8.0 1,380 18.0 2,683 9.1 Western Africa 842 39.4 1,149 15.0 6,031 20.4 Total 2,138 100.0 7,680 100.0 29,599 100.0 Data source: Ramankutty et al. 2008 and FAO 2012. Note: sq km=square kilometers. 16
- 33. Data source (allmaps): Ramankutty et al. 2008. Notes: All values are expressed as a percentage of the area within each ~100 km2 grid cell. Cropland=arable land and permanent crops; pastureland = permanent meadows and pastures, as defined by the Food and Agriculture Organization of the United Nations (FAO 2013). ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT 0 0 20 40 60 80 100 Percent 0 0 20 40 60 80 100 Percent MAP 1 Cropland, c. 2000 MAP 2 Pastureland, c. 2000 17
- 34. FOOTPRINT OF AGRICULTURE Irrigated Areas Stefan Siebert and Karen Frenken WHAT IS THIS MAP TELLING US? Total area equipped for irrigation in Africa is 13.5 million hectares (ha) of which 11.5 million ha are actually under irrigation (Figure 1). The map shows the countries with the largest amount of area equipped for irrigation are Egypt (3.5 million ha), Sudan and South Sudan (1.9 million ha), South Africa (1.5 million ha), and Morocco (1.5 mil-lion ha). All of these countries face arid climate conditions. In Madagascar where it is more humid, rice is cultivated on about 1 million ha of irrigated land. These six coun-tries account for almost 60 percent of the area equipped for irrigation in Africa. The regions with the highest den-sity of irrigated land (50 percent or greater of the grid cell)1 are located mainly in northern Africa in the Nile River Basin (Egypt, Sudan) and in the countries next to the Mediterranean Sea (Morocco, Algeria, Tunisia, Libya). WHY IS THIS IMPORTANT? Since the beginning of crop cultivation, irrigation has been used to compensate for the lack of precipitation. In rice cultivation, irrigation also controls the water level in the fields and suppresses weed growth. Crop yields are higher and the risk of crop failures is lower in irrigated agriculture. Because the risk of drought stress is lower on irrigated land, farmers are more likely to spend on other inputs like fertil-izers. Irrigation may also increase cropping intensity (p. 28), allowing farmers to cultivate several crops per year on the same field. It is important, therefore, when assessing crop productivity and food security, to consider the availability of irrigation infrastructure. Irrigation represents the largest use of freshwater in Africa. Many dams were constructed to improve the sup-ply of irrigation water, thereby modifying river discharge and increasing evaporation from artificial lakes. Extraction of groundwater for irrigation is increasingly of concern, because it has lowered groundwater tables in important aquifers. Use of irrigation results in an increase of evapotranspiration and reduces the land’s surface temperature. Information on the extent of irrigated land is therefore also important for hydro-logical studies and regional climate models. WHAT ABOUT THE UNDERLYING DATA? The map shows the area equipped for irrigation as a per-centage of a 5 arc-minute grid cell. It was derived from version 5 of the Digital Global Map of Irrigation Areas (Siebert et al. 2013a). The map was developed by combin-ing subnational irrigation statistics for 441 administrative units derived from national census surveys and from reports available at the Food and Agriculture Organization of the United Nations and other international organizations with geospatial information on the position and extent of irri-gation schemes. Statistics for the year closest to 2005 were used if data for more than one year were available. Geospatial information on position and extent of irrigated areas was derived by digitizing a large number of irrigation maps derived from inventories based on remote sensing (Siebert et al. 2013b). WHERE CAN I LEARN MORE? Global Map of Irrigation Areas (Version 5): http://bit.ly/1eHpDex Update of the Digital Global Map of Irrigation Areas to Version 5. Siebert et al. 2013b: http://bit.ly/1cM6bip Development and Validation of the Global Map of Irrigation Areas. Siebert et al. 2005. FIGURE 1 Area equipped for irrigation and area actually irrigated per region, c. 2005 0 1 2 3 4 5 6 7 8 9 Area actually irrigated Area equipped for irrigation Western Africa Southern Africa Northern Africa Middle Africa Eastern Africa Irrigated area (million ha) Data source: Siebert et al. 2013a and FAO 2012. 1 Each cell measures approximately 100 km2 or 10,000 hectares at the equator. 18
- 35. Data source: Siebertet al. 2013a. Note: The percent values represent the share of each 100 km2 cell that is equipped for irrigation. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Tunisia Algeria Morocco Libya Egypt Sudan Libya Chad Ethiopia 0 0 1 5 10 20 35 100 50 75 Percent irrigated MAP 1 Extent of irrigated areas, c. 2005 19
- 36. FOOTPRINT OF AGRICULTURE FIGURE 1 Area harvested of top five cereal crops 0 5 10 15 20 25 30 35 1961 1968 1975 1982 1989 1996 2003 2010 Million hectares Rice, paddy Maize Sorghum Millet Wheat Data source: FAO 2012. FIGURE 2 Yield of top five cereal crops 0.0 0.5 1.0 1.5 2.0 2.5 3.0 1961 1968 1975 1982 1989 1996 2003 2010 Metric tons/hectare Rice, paddy Maize Sorghum Millet Wheat Data source: FAO 2012. Note: One metric ton=1,000 kilograms. Cereal Crops Ulrike Wood-Sichra WHAT ARE THESE MAPS TELLING US? Cereals are grown in all of Africa except for desert and forested areas. The cereal area is about 30 percent maize, 23 percent sorghum, 21 percent millet, 9 percent wheat (Maps 1–4), and 9 percent rice. Maps 1–3 show that maize is prevalent throughout Africa and the densest areas for sor-ghum and millet, with more than 3,000 hectares per cell,1 are just south of the Sahel. Wheat (Map 4) is grown in high con-centrations in northern Africa, with sparser areas in eastern and southern Africa. In the last 50 years, the harvested areas of maize, millet, and sorghum each doubled from a base of 10–15 million hectares to 20–30 million hectares (Figure 1). Rice areas have nearly quadrupled, from 2.8 to 9.3 million hectares. Yields have notably climbed for maize and wheat during the same period, rising from 0.7 to 2.3 metric tons per hectare for wheat and doubling from 1.0 to 2.0 for maize (Figure 2). Rice yields have increased by more than half, from about 1.5 to 2.5 metric tons per hectare. Millet and sorghum yields show little change (FAO 2012). WHY IS THIS IMPORTANT? Cereals account for 50 percent of the average daily caloric intake in Africa. Wheat and rice are particularly important, accounting for 30 percent and 16 percent of cereal calories consumed, respectively. Cereal production in Africa is sub-stantial, but it is not enough to meet demand; the continent must import about 55 percent of consumed wheat and more than 30 percent of consumed rice (FAO 2012). Understanding where half of the continent’s calories (both vegetal and ani-mal) are grown, and how intensively, is vital to increasing productivity and enhancing food security. Identifying areas where new or improved rice- and wheat-growing technolo-gies could have the most impact can also aid in making the continent less dependent on imports. WHAT ABOUT THE UNDERLYING DATA? The maps are based on area harvested per cell, calculated using the Spatial Production Allocation Model (SPAM) 2000 (You et al. 2012). The model uses many datasets, including land cover recorded by satellites, crop suitability maps under various water regimes and production systems, irrigation maps (p. 19), subnational crop statistics from each country, country totals from the Food and Agriculture Organization of the United Nations (FAO 2012), and data on production systems within each country. WHERE CAN I LEARN MORE? SPAM: The Spatial Production Allocation Model: http://mapspam.info Food and Agriculture Organization of the United Nations Statistics Division database: http://faostat3.fao.org 1 Each cell measures approximately 100 km2 or 10,000 hectares at the equator. 20
- 37. Data source (allmaps): You et al. 2012. Note: The values on the maps represent the number of hectares harvested per 100 km2 cell. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI 0 0 10 250 500 1,000 3,000 Hectares 0 0 10 250 500 1,000 3,000 Hectares 0 0 10 250 500 1,000 3,000 Hectares 0 0 10 250 500 1,000 3,000 Hectares MAP 1 Maize MAP 2 Sorghum MAP 3 Millet MAP 4 Wheat Cereal crop area harvested, 2000 21
- 38. FOOTPRINT OF AGRICULTURE Root Crops Ulrike Wood-Sichra WHAT ARE THESE MAPS TELLING US? The area devoted to harvest root crops in Africa has grown significantly over the last 50 years. Cassava area has more than doubled, from 5.5 million to 12 million hectares (ha); sweet potato area has more than quintupled, from 600,000 to 3.3 million ha; and potato area has grown more than six-fold, from 250,000 to 1.8 million ha. Cassava and sweet potatoes continue to be among the most import-ant root crops in Africa, with cassava occupying about half of the root crops area and sweet potatoes about 14 per-cent. South of the Sahel, cassava and sweet potatoes are grown in similar areas (Maps 1 and 2). Both are grown inten-sively, with 1,000 or more ha per cell,1 in the southeast cor-ner of Nigeria, in the eastern part of Uganda, and in Rwanda and Burundi. Potatoes are becoming a more important part of Africa’s crop mix, although they currently account for only 8 percent of the harvested area and are grown in just a few African countries (Map 3). While harvested area of root crops has expanded considerably since 1961, yields per hectare have increased significantly for only some crops (Figure 1). Cassava yields have notably improved by about 80 percent, from less than 6 to roughly 10 metric tons per hectare. Potato yields have also fared well, increasing by about half from about 8 to 12 metric tons per hectare. Taro and yam yields grew more modestly, by 41 percent and 26 percent to 6 and 10 metric tons per hectare, respectively. Sweet potato yields, however, have hovered around 5 metric tons per hectare for decades and even shown a slight down-ward trend over the past 30 years (FAO 2012). WHY IS THIS IMPORTANT? Africa needs to improve yields and the share of nutrient-rich roots and tubers in the diet of its growing population. Roots and tubers contribute only about 13 percent of the calories in the average African’s diet, which is a smaller por-tion than other staples. But roots, especially cassava, are “insurance crops” that increase food security because they can be left in the ground until needed. Nearly all the sweet potato crop (85 percent) is destined for human consump-tion. But cassava is also important as fodder, and more than a third produced goes to animal feed. Most of the roots and tubers consumed are grown locally. Thus, policymakers and agricultural experts can use the maps to identify areas that might benefit from larger harvests of roots and tubers, and by extension, improve nutrition at the local level. WHAT ABOUT THE UNDERLYING DATA? The maps are based on area harvested per cell, calculated by the Spatial Production Allocation Model (SPAM) 2000 (You et al. 2012). The model uses many datasets, includ-ing land cover recorded by satellites, crop suitability maps under various water regimes and production systems, irri-gation maps, subnational crop statistics from each country, Food and Agriculture Organization of the United Nations’ country totals (FAO 2012), and data on production systems within each country. WHERE CAN I LEARN MORE? SPAM: The Spatial Production Allocation Model: http://mapspam.info Food and Agriculture Organization of the United Nations statistical database: http://faostat3.fao.org FIGURE 1 Yield of top five root crops 2 4 6 8 10 12 14 1961 1967 1973 1979 1985 1991 1997 2003 2009 Metric tons/hectare Sweet potatoes Cassava Taro (cocoyam) Potatoes Yams Data source: FAO 2012. Note: One metric ton = 1,000 kilograms. 1 Each cell measures approximately 100 km2 or 10,000 hectares at the equator. 22
- 39. Data source (allmaps): You et al. 2012. Note: The values on the maps represent the number of hectares harvested per 100 km2 cell. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT 0 0 10 250 500 1,000 3,000 Hectares 0 0 10 250 500 1,000 3,000 Hectares 0 0 10 250 500 1,000 3,000 Hectares INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI MAP 1 Cassava MAP 2 Sweet potato MAP 3 Potato Root crop area harvested, 2000 23
- 40. FOOTPRINT OF AGRICULTURE Livestock and Mixed Crop-Livestock Systems Philip Thornton WHAT IS THIS MAP TELLING US? Livestock-producing agricultural systems cover 73 percent of Africa and stretch across several climates (Map 1). To some extent, these climates determine what type of farming is practiced. In Africa, livestock-producing systems are bro-ken into two main categories: livestock and mixed crop-live-stock. These systems exist in three common African climates: arid/semiarid, humid/subhumid, and temperate/tropical highlands. Livestock systems are most prevalent on graz-ing lands in arid climates that cover large swaths of Africa. Mixed crop-livestock farming systems are either rain-fed or irrigated. Rainfed systems are much more common (although areas of Sudan and Egypt have important irri-gated mixed systems that present different opportunities and constraints). There are many mixed crop-livestock sys-tems throughout western Africa, eastern Africa, and parts of southern Africa. The Congo Basin, in central Africa, is mostly forest, with some savanna and cropland at its outer edges. As a result, the Basin is home to a small number of livestock systems relative to the rest of the continent and only a smat-tering of mixed crop-livestock systems. WHY IS THIS IMPORTANT? Many studies have found the influences of crop and live-stock production vary considerably, not only regionally but also according to production system (Robinson et al. 2011). Globally, but particularly in Africa and Asia, crops and livestock are often interdependent and influence farmer households and livelihoods in a number of ways. Detailed knowledge of crop and livestock systems and their distribu-tion allows researchers to measure impacts on everything from the environment to livestock disease risk. For exam-ple, viewing the livestock density by type and region helps researchers measure the level of environmental impact (Table 1). Classification of agricultural systems can also pro-vide a framework for predicting the evolution of the agri-cultural sector in response to changing demography and associated shifts in food demand, land use (for example, competition for land from food, feed, and biofuel produc-tion), and climate. WHAT ABOUT THE UNDERLYING DATA? The systems classification is based on Seré and Steinfeld (1996). In livestock systems, more than 90 percent of dry matter fed to animals comes from rangelands, pastures, annual forages, and purchased feeds, and less than 10 per-cent of the total value of production (VoP) comes from nonlivestock farming activities. Mixed crop-livestock farm-ing systems are systems in which more than 10 percent of the dry matter fed to animals comes from crop by-products (for example, stubble) or more than 10 percent of the total VoP comes from nonlivestock farming activi-ties. The systems were mapped using various mapped data sources, including land cover data, irrigated areas, human population density, and length of growing period (LGP). The climate categories are defined as follows: arid/semi-arid has an LGP ≤ 180 days; humid/subhumid has an LGP 180 days; and the temperate/tropical highlands climate is based on specific LGP, elevation, and temperature cri-teria. The systems classifications have several weaknesses, including differences in estimates of the amount of Africa’s cropland, depending on the data used, thus, there is some uncertainty in identifying the mixed crop-livestock sys-tems. Researchers are now using other data sources to break down the mixed systems of the Seré and Steinfeld classification by dominant food and feed crop categories (Robinson et al. 2011). WHERE CAN I LEARN MORE? Farming Systems and Poverty: Improving Farmers’ Livelihoods in a Changing World. Dixon et al. 2001. Global Livestock Production Systems. Robinson et al. 2011. TABLE 1 Livestock density by region, 2005 REGION TYPE OF LIVESTOCK average number/km2 Cattle Sheep Goat Northern Africa 5 10 5 Middle Africa 3 1 2 Eastern Africa 14 6 9 Western Africa 6 10 12 Southern Africa 7 11 4 AFRICA 7 7 7 Data source: Robinson et al. 2011 and FAO 2012. 24
- 41. Data source: Robinsonet al. 2011. Note: The mixed categories represent a mix of crop and livestock systems. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Rainfed-arid/semiarid Rainfed-humid/subhumid Rainfed-temperate/tropical highlands Livestock-grazing Rainfed-arid/semiarid Rainfed-humid/subhumid Rainfed-temperate/tropical highlands Irrigated Mixed crop-livestock Urban areas Nonlivestock vegetated areas Other MAP 1 Livestock production systems by climate zone 25
- 42. FOOTPRINT OF AGRICULTURE Ruminant Livestock Timothy Robinson, William Wint, Giulia Conchedda, Guiseppina Cinardi, and Marius Gilbert WHAT ARE THESE MAPS TELLING US? Ruminant livestock are raised across large parts of Africa where environmental conditions allow (Maps 1–4). Cattle, sheep, and goats are the most widespread, while camels are restricted to drier areas, particularly in the Horn of Africa and the arid parts of western Africa. These maps of ruminant dis-tribution should, however, be used in conjunction with the livestock production systems map (p. 25) to better under-stand the systems and climate zones where ruminant live-stock are found. The role of livestock varies greatly depending on the production system. The heavily forested areas and hyperarid deserts of Africa have very low densities of live-stock. In arid and semiarid regions of Africa, where the poten-tial for crop growth is limited, cattle, sheep, goats, and camels are raised in low productivity, pastoral (extensive livestock grazing) systems in which ambulatory stock can take advan-tage of seasonal, patchy vegetation growth. In these areas, raising livestock is the only viable form of agriculture. In the more settled humid, subhumid, and tropical highland areas, cattle and small ruminants largely live in the same areas as the human population. In these mixed crop-livestock farming systems, livestock can increase crop production by provid-ing draft power and manure, and by enhancing labor pro-ductivity. At the same time, organic material not suited for human consumption can be converted into high-value food and nonfood products, such as traction, manure, leather, and bone. WHY IS THIS IMPORTANT? Poverty in Africa remains widespread (p. 77). One quarter of the world’s estimated 752 million poor livestock keepers live in Africa south of the Sahara (SSA), where more than 85 percent of them live in extreme poverty (Otte et al. 2012). Agricultural productivity gains and diversification into high-value prod-ucts such as livestock are essential ways of raising rural incomes and improving food security in such areas. For three reasons—the large share of the rural poor who keep livestock, the important contributions livestock can make to sustain-able rural development, and the fast-growing demand for live-stock products—diversification into livestock and increased livestock productivity must play an integral role in strategies to reduce poverty and increase agricultural productivity. Progress in poverty reduction will require well-targeted inter-ventions to promote economic growth that the poor can contribute to and from which they can benefit. Livestock maps such as these, along with other information such as pov-erty and production systems, can contribute significantly to better targeting. WHAT ABOUT THE UNDERLYING DATA? The Gridded Livestock of the World database (Wint and Robinson 2007) provided the first modelled livestock den-sities of the world, adjusted to match official national esti-mates for the reference year 2005 (FAO 2007), at a spatial resolution of 3 arc-minutes (about 25 km2 at the equa-tor). Recent methodological improvements have signifi-cantly enhanced these maps. More up-to-date and detailed subnational livestock statistics have been collected; a new, higher resolution set of predictor variables based on multitemporal Moderate Resolution Imaging Spectroradiometer (MODIS) imagery is used; and the analytical procedure has been revised and extended to include a more systematic assessment of the model accuracy. While the observed, sub-national statistics vary in date and resolution, the maps are standardized so that the national totals match the official estimates for 2006 (FAO 2013). WHERE CAN I LEARN MORE? Download the data from the Livestock-Geo-Wiki Project: http://livestock.geo-wiki.org “Mapping the Global Distribution of Livestock.” Robinson et al. 2014. “The Food and Agriculture Organization’s Gridded Livestock of the World.” Robinson, Franceschini, and Wint 2007. Gridded Livestock of the World, 2007. Wint and Robinson 2007. 26
- 43. Data source (allmaps): Robinson et al. 2013; Lecksell and World Bank 2013. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT 0 1 5 10 20 50 100 250 No data Number per km² 0 1 5 10 20 50 100 250 No data Number per km² 0 1 5 10 20 50 100 250 No data Number per km² 0 1 5 10 20 50 100 250 No data Number per km² MAP 1 Cattle MAP 2 Sheep MAP 3 Goats MAP 4 Camels Ruminant livestock distribution, 2006 27
- 44. FOOTPRINT OF AGRICULTURE Cropping Intensity Stefan Siebert, Petra Döll, and Felix T. Portmann WHAT IS THIS MAP TELLING US? The map shows cropping intensity, which is the number of crop harvests per cell per year.1 Cropping intensity is highest in irrigated regions, such as the Nile Delta (p. 19), or in wet-land rice-growing areas, such as southern Nigeria and Côte d’Ivoire, where more than one crop harvest per year is pos-sible. In contrast, many rainfed areas in Africa see less than one harvest per year due to scarce water or nutrient supplies, particularly in drier regions such as the Sahel, South Sudan, Central African Republic, and much of southern Africa. Additionally, shifting cultivation, in which crops are grown every three to ten years on available cropland with fallow periods in between to allow for nutrient regeneration, is common practice in Africa. These limitations and practices lead to low cropping intensity values on average for most regions of Africa (Figure 1). One also can use the map to identify potential target areas for agricultural intensification by identifying regions with low-cropping intensity and com-paring them with areas with fast-growing populations. WHY IS THIS IMPORTANT? The growing demand for agricultural products requires either the cultivation of more land or intensified agricul-tural land use. It would be difficult to increase cropland area, particularly in regions with high population density, sensitive ecosystems, or poor soil quality. In such regions, intensifying agricultural land use may be the only option. Previous research on crop productivity has focused primar-ily on crop yields or yield gaps and therefore strictly on the amount of crop yield per harvest. This works for temperate climate regions where only one harvest is possible per year. In contrast, in tropical or subtropical regions, increasing the number of harvests per year can lead to increases in crop production. Increasing cropping intensity by reducing the length of the fallow period is a traditional way to adapt cul-tivation systems to growing demand for crop products and to shortages in cultivatable land. To be sustainable, increases in cropping intensity must be supplemented with improved water and nutrient management. WHAT ABOUT THE UNDERLYING DATA? Cropping intensity was calculated based on the MIRCA2000 dataset (Portmann, Siebert, and Döll 2010) as a ratio of harvested crop area to cropland extent, which included fallow land. This dataset provides, separately for irrigated and rainfed agriculture, monthly growing areas of 26 crops or crop groups at a 5 arc-minute resolution. It refers to the period around 2000 and was developed by combin-ing global inventories on cropland extent (Ramankutty et al. 2008; Map 1, p. 17); the harvested area of 175 distinct crops (Monfreda, Ramankutty, and Foley 2008); the extent of area equipped for irrigation (Siebert, Hoogeveen, and Frenken 2006); and inventories on irrigated area per crop that used crop calendars derived from FAO and other databases. WHERE CAN I LEARN MORE? Current Opinion in Environmental Sustainability: www.sciencedirect.com/science/journal/18773435/5/5 FAO Irrigated Crop Calendars: http://bit.ly/1c9yLH6 “Global Estimation of Monthly Irrigated and Rainfed Crop Areas”: http://bit.ly/1dc6Gz8 “Global Patterns of Cropland Use Intensity.” Siebert, Portmann, and Döll 2010: http://bit.ly/1rnJ0RT “Increasing Global Crop Harvest Frequency: Recent Trends and Future Directions.” Ray and Foley 2013: http://bit.ly/1gTax8S FIGURE 1 Cropping intensity by region, 2000 Southern Africa Middle Africa Eastern Africa Northern Africa Western Africa 0.0 0.2 0.4 0.6 0.8 1.0 Average number of crop harvests/year Data source: Siebert, Portmann, and Döll 2010 and FAO 2012. Note: Cropping intensity=the number of crop harvests per year. 1 Each cell measures approximately 100 km2 or 10,000 hectares at the equator. 28
- 45. Data source: Siebert,Portmann, and Döll 2010. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT 0.0 0.2 0.4 0.6 0.9 1.1 1.3 1.5 3.0 No data Average number of harvests per year MAP 1 Cropping intensity 29
- 46. FOOTPRINT OF AGRICULTURE Land Productivity for Staple Food Crops Ulrike Wood-Sichra and Stanley Wood WHAT IS THIS MAP TELLING US? Almost three-quarters of Africa’s harvested agricultural land is devoted to the production of staple food crops,1 but only about one-third of that land generates annual output worth more than $5002 from each cropped hectare. With farmers typically cultivating just a half to three hectares of land to support entire families, rural poverty and food insecurity are pervasive, especially where nonfarm employment options are limited. While some areas can produce food crop outputs worth more than $2,500 per hectare (compared to an aver-age of $517 per hectare across all of Africa), such impressive results are concentrated in less than 1 percent of the total harvested area and are likely boosted by access to irrigation. Map 1 shows the distribution of Africa’s average land pro-ductivity for staple crops ranging from $250 or less per hec-tare at the fringes of the Sahel and in parts of eastern Africa to $1,000 or more per hectare in southern Nigeria, parts of Ghana, and along the Nile Valley and Delta. Summarizing values by agroecological zone (p. 34), tropical arid zones, such as on the northern edge of the Sahel and in eastern Africa, have some of the lowest average values of production per hectare; and subtropical arid zones, such as the Nile Delta where irrigation is widely practiced, and subtropical humid zones in southern Africa, have some of the highest average values of production per hectare (Table 1). WHY IS THIS IMPORTANT? Land productivity serves as a compact measure of the gen-eral status of agricultural and rural development. It is an implicit reflection of the status of local environmental con-ditions, input use, and farmer know-how. Its spatial varia-tion, furthermore, provides a picture of the likely relative differences in land rental values. Detailed empirical studies of diversity in land productivity point to a range of associated factors including agroecology; farmers’ access to knowledge; inputs, credit, infrastructure, and markets; land tenure; and cultural preferences that shape crop and technology choices, production practices, and market engagement. WHAT ABOUT THE UNDERLYING DATA? Estimates of land productivity were derived using two core data sources: (1) average annual production (metric tons) and area harvested (hectares) for 14 of the most widely grown food crops during the period 1999–2001 derived for each 5 arc-minute grid cell3 across Africa using the Spatial Production Allocation Model (SPAM) 2000 (You et al. 2012), and (2) prices and national value of production for each crop over the same period (FAO 2012). The total value of food crop production (VoP) for any grid cell is calculated as the sum of the VoPs for each crop, where VoP is a product of crop price and production. Land productivity is derived by dividing the total VoP of the 14 crops by the total harvested area of those same crops for each grid cell. WHERE CAN I LEARN MORE? More information on the SPAM model: http://mapspam.info FAOSTAT database: http://faostat3.fao.org/home/index.html TABLE 1 Average value of staple food crop production (US$) per hectare by region in Africa Agroecological zone Eastern Middle Northern Southern Western AFRICA Subtropic–arid 781 1448 501 1355 Subtropic–semiarid 546 726 295 392 338 Subtropic–subhumid 336 532 349 Subtropic–humid 837 837 Tropic–arid 89 208 186 336 225 184 Tropic–semiarid 336 469 100 351 246 270 Tropic–subhumid 496 479 161 491 1083 760 Tropic–humid 659 661 133 1571 1144 749 Average 480 536 433 398 580 517 Data source: You et al. 2012; FAO 2012; Sebastian 2009. Note: All local prices converted to international dollars at 2004–2006 average purchasing power parity exchange rates. 1 Harvested areas and production values include the following staple food crops: maize, sorghum, millet, rice, wheat, barley, cassava, sweet potatoes and yams, bananas and plantains, Irish potatoes, beans, groundnuts, soybeans, and other pulses. 2 All local prices converted to international dollars at 2004–2006 average purchasing power parity exchange rates. 3 Each cell measures approximately 100 km2 or 10,000 hectares at the equator. 30
- 47. Data source: Youet al. 2012 and FAO 2012. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT 0 250 500 1,000 2,500 Nonstaple food crop area Average value of production in cropland area ($ per hectare) INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI MAP 1 Land productivity for staple food crops, 2000 31
- 48. FOOTPRINT OF AGRICULTURE Works Cited FARMING SYSTEMS OF AFRICA Azzarri, C., S. Wood, G. Hyman, E. Barona, M. Bacou, and Z. Guo. 2012. Sub-national Poverty Map for Sub-Saharan Africa at 2005 International Poverty Lines (r12.12). http://bit.ly/1gq4jLF. Dixon, J., J-M. Boffa, and D. Garrity. 2014. Farming Systems and Food Security in Sub-Saharan Africa: Priorities for Science and Policy. Unpublished. Australian Centre for International Agricultural Research and World Agroforestry: Canberra and Nairobi. Dixon, J., A. Gulliver, and D. Gibbon. 2001. Farming Systems and Poverty: Improving Farmer’s Livelihoods in a Changing World. Rome and Washington, DC: Food and Agriculture Organization of the United Nations and World Bank. http://bit.ly/1dDekBW. FAO (Food and Agriculture Organization of the United Nations). 2013a. Aquastat database. Accessed on February 7, 2014. http://bit.ly/1gNqgEf. —. 2013b. FAOSTAT database. Accessed on Dec. 15, 2013. http://faostat3.fao.org/. —. 2013c. Global Livestock Production and Health Atlas database. Accessed on February 7, 2014. http://bit.ly/NyPMF8. —. 2013d. Gridded Livestock of the World database. Accessed on February 7, 2014. http://bit.ly/1pW6kFE. —. 2013e. State of the World’s Land and Water Resources for Food and Agriculture database. Accessed on February 7, 2014. www.fao.org/nr/solaw/en/. FAO/IIASA (International Institute for Applied Systems Analysis). 2012. GAEZ v3.0 Global Agro-ecological Zones database. Accessed on Feb. 7, 2014. www.gaez.iiasa.ac.at/. Garrity, D., J. Dixon, and J-M. Boffa. 2012. “Understanding African Farming Systems: Science and Policy Implications.” Paper presented at the Food Security in Africa: Bridging Research into Practice Conference, Australian International Food Security Centre/ Australian Centre for International Agriculture Research, Sydney, Australia, November 2012. http://bit.ly/1h8lmGJ. HarvestChoice. 2011. “Average Travel Time to Nearest Town Over 20K (hours) (2000).” Washington, DC and St. Paul, MN: International Food Policy Research Institute and University of Minnesota. http://harvestchoice.org/node/5210. UN (United Nations). 2013. “World Urbanization Prospects, the 2011 Revision.” Department of Economic and Social Affairs, Population Division, Population Estimates and Projections Section. Accessed February 7, 2014. http://esa.un.org/unup/. CROPLAND AND PASTURELAND Bartholomé, E., and A. S. Belward. 2005. “GLC2000: A New Approach to Global Land Cover Mapping from Earth Observation Data.” International Journal of Remote Sensing 26 (9): 1959–1977. FAO. 2012. FAOSTAT database. Accessed October 2012. http://faostat.fao.org/site/291/default.aspx. — . 2013. FAOSTAT: Concepts Definitions: Glossary List. http://faostat.fao.org/site/291/default.aspx. Foley, J. A., R. DeFries, G. P. Asner, C. Barford, G. Bonan, S. R. Carpenter, F. S. Chapin, M. T. Coe, G. C. Daily, H. K. Gibbs, J. H. Helkowski, T. Holloway, E. A. Howard, C. J. Kucharik, C. Monfreda, J. A. Patz, I. C. Prentice, N. Ramankutty, and P. K. Snyder. 2005. “Global Consequences of Land Use.” Science 309 (5734): 570–574. Friedl, M. A., D. Sulla-Menashe, B. Tan, A. Schneider, N. Ramankutty, and X. M. Huang. 2010. “MODIS Collection 5 Global Land Cover: Algorithm Refinements and Characterization of New Datasets.” Remote Sensing of Environment 114 (1): 168–182. Ramankutty, N., A. T. Evan, C. Monfreda, and J. A. Foley. 2008. “Farming the Planet: 1. Geographic Distribution of Global Agricultural Lands in the Year 2000.” Global Biogeochemical Cycles 22 (1). Thornton, P. K., and M. Herrero. 2010. “Potential for Reduced Methane and Carbon Dioxide Emissions from Livestock and Pasture Management in the Tropics.” Proceedings of the National Academy of Sciences of the United States of America 107 (46): 19667–19672. IRRIGATED AREAS FAO. 2012. FAOSTAT database. Accessed October 2012. http://faostat.fao.org/site/291/default.aspx. Siebert, S., P. Döll, J. Hoogeveen, J.-M. Faures, K. Frenken, and S. Feick. 2005. “Development and Validation of the Global Map of Irrigation Areas.” Hydrology and Earth System Sciences 9: 535–547. Siebert, S., V. Henrich, K. Frenken, and J. Burke. 2013a. Global Map of Irrigation Areas Version 5. Bonn, Germany: Rheinische Friedrich-Wilhelms-University; Rome: Food and Agriculture Organization of the United Nations. http://bit.ly/1eHpDex. Siebert, S., V. Henrich, K. Frenken, and J. Burke. 2013b. Update of the Global Map of Irrigation Areas to Version 5. Bonn, Germany: Rheinische Friedrich-Wilhelms-University; Rome: Food and Agriculture Organization of the United Nations. http://bit.ly/1cM6bip. CEREAL CROPS FAO. 2012. FAOSTAT database. Accessed October 15, 2012. http://faostat.fao.org/site/291/default.aspx. You, L., Z. Guo, J. Koo, W. Ojo, K. Sebastian, M. T. Tenorio, S. Wood, and U. Wood-Sichra. 2012. “Spatial Production Allocation Model (SPAM) 2000, Version 3, Release 1.” Accessed December 14, 2012. http://MapSPAM.info. ROOT CROPS FAO. 2012. FAOSTAT database. Accessed October 2012. http://faostat.fao.org/site/291/default.aspx. You, L., Z. Guo, J. Koo, W. Ojo, K. Sebastian, M. T. Tenorio, S. Wood, and U. Wood-Sichra. 2012. “Spatial Production Allocation Model (SPAM) 2000, Version 3, Release 1.” Accessed December 14, 2012. http://MapSPAM.info. LIVESTOCK AND MIXED CROP-LIVESTOCK SYSTEMS Dixon, J., A. Gulliver, and D. Gibbon. 2001. Farming Systems and Poverty: Improving Farmers’ Livelihoods in a Changing World. Rome: Food and Agriculture Organization of the United Nations; Washington, DC: World Bank. FAO. 2012. FAOSTAT database. Accessed October 2012. http://faostat.fao.org/site/291/default.aspx. Robinson, T. P., P. K. Thornton, G. Franceschini, R. L. Kruska, F. Chiozza, A. Notenbaert, G. Cecchi, M. Herrero, M. Epprecht, S. Fritz, L. You, G. Conchedda, and L. See. 2011. Global Livestock Production Systems. Rome: Food and Agriculture Organization of the United Nations and International Livestock Research Institute. Seré, C., and H. Steinfeld. 1996. World Livestock Production Systems: Current Status, Issues and Trends. FAO Animal Production and Health Paper 127. Rome: Food and Agriculture Organization of the United Nations. RUMINANT LIVESTOCK FAO 2007. FAOSTAT database. Accessed in 2007. http://faostat.fao.org. FAO. 2013. FAOSTAT database. Accessed in 2013. http://faostat.fao.org. Lecksell, J., and World Bank. 2013. Personal communication regarding world boundaries for 2013, Nov. 26. Lecksell is lead World Bank cartographer. Otte, J., A. Costales, J. Dijkman, U. Pica-Ciamarra, T. P. Robinson, V. Ahuja, C. Ly, and D. Roland-Holst. 2012. Livestock Sector Development for Poverty Reduction: An Economic and Policy Perspective–Livestock’s Many Virtues. Rome: Food and Agriculture Organization of the United Nations, Animal Production and Health Division. Robinson, T. P., G. Franceschini, and W. Wint. 2007. “The Food and Agriculture Organization’s Gridded Livestock of the World.” Veterinaria Italiana 43: 745–751. Robinson, T. P., G. R. W. Wint, G. Conchedda, T. P. van Boeckel, V. Ercoli, E. Palamara, G. Cinardi, L. D’Aietti, S. I. Hay, and M. Gilbert. 2014. “Mapping the Global Distribution of Livestock.” PLoS ONE, in press. Wint, G. R. W., and T. P. Robinson. 2007. Gridded Livestock of the World, 2007. Rome: Food and Agriculture Organization of the United Nations, Animal Production and Health Division. CROPPING INTENSITY FAO. 2012. FAOSTAT database. Accessed October 2012. http://faostat.fao.org/site/291/default.aspx. Monfreda, C., N. Ramankutty, and J. A. Foley. 2008. “Farming the Planet: 2. Geographic Distribution of Crop Areas, Yields, Physiological Types, and Net Primary Production in the Year 2000.” Global Biogeochemical Cycles 22 (1). Portmann, F. T., S. Siebert, and P. Döll. 2010. “MIRCA2000-Global Monthly Irrigated and Rainfed Crop Areas around the Year 2000: A New High-Resolution Data Set for Agricultural and Hydrological Modeling.” Global Biogeochemical Cycles 24 (1). Ramankutty, N., A. T. Evan, C. Monfreda, and J. A. Foley. 2008. “Farming the Planet: 1. Geographic Distribution of Global Agricultural Lands in the Year 2000.” Global Biogeochemical Cycles 22 (1). Ray, D. K., and J. A. Foley. 2013. “Increasing Global Crop Harvest Frequency: Recent Trends and Future Directions.” Environmental Research Letters 8 (4): 1–10. http://bit.ly/1gTax8S. Siebert, S., J. Hoogeveen, and K. Frenken. 2006. Irrigation in Africa, Europe and Latin America: Update of the Digital Global Map of Irrigation Areas to Version 4. Frankfurt, Germany: Goethe University; Rome: Food and Agriculture Organization of the United Nations. http://bit.ly/1hm1f4Z. Siebert, S., F. T. Portmann, and P. Döll. 2010. “Global Patterns of Cropland Use Intensity.” Remote Sensing 2 (7): 1625–1643. LAND PRODUCTIVITY FOR STAPLE FOOD CROPS FAO. 2012. FAOSTAT database. Accessed October 15, 2012. http://faostat.fao.org/site/291/default.aspx. Sebastian, K. 2009. Agro-ecological Zones of Africa. International Food Policy Research Institute. http://hdl.handle.net/1902.1/22616. You, L., Z. Guo, J. Koo, W. Ojo, K. Sebastian, M. T. Tenorio, S. Wood, and U. Wood-Sichra. 2012. “Spatial Production Allocation Model (SPAM) 2000, Version 3, Release 1.” Accessed December 14, 2012. http://MapSPAM.info. 32
- 49. ATLAS OF AFRICANAGRICULTURE RESEARCH DEVELOPMENT Growing Conditions Agroecological Zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Climate Zones for Crop Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Rainfall and Rainfall Variability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Soil Fertility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Works Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 33
- 50. GROWING CONDITIONS AgroecologicalZones Kate Sebastian WHAT IS THIS MAP TELLING US? Agroecological zones (AEZs) are geographical areas exhib-iting similar climatic conditions that determine their ability to support rainfed agriculture. At a regional scale, AEZs are influenced by latitude, elevation, and temperature, as well as seasonality, and rainfall amounts and distribution during the growing season. The resulting AEZ classifications for Africa have three dimensions: major climate (tropical or sub-tropical conditions), elevation (warmer lowland or cooler upland production areas), and water availability (ranging from arid zones with less than 70 growing days per year to humid zones where moisture is usually sufficient to sup-port crop growth for at least nine months per year) (Fischer et al. 2009). The map shows the broad latitudinal symmetry of major climates and water availability north and south of the equator, disrupted by the influence of highland and lake complexes primarily associated with the East African Rift Valley that extends from Ethiopia to Mozambique. The Sahel—located between the Sahara Desert in the north and the Sudanian Savanna in the south—comprises warm tropical arid and semiarid zones characterized by a strong north-south water availability gradient, while the highlands of East Africa are distinguished by cooler, more humid tropical conditions. The most extensive humid zone is centered on the Congo Basin, stretching from the Rwenzori and Virunga mountains at the borders of Uganda, Rwanda, and the Democratic Republic of the Congo in the east to the Atlantic coast in the west. The continent is pri-marily tropical, but significant subtropical areas with pro-nounced seasonality in temperatures and day length are found in northern and southern Africa (beyond the tropical limits of 23.44 degrees north and south of the equator). WHY IS THIS IMPORTANT? Most African farmers, particularly in tropical areas, rely on rainfed agriculture with very limited use of inputs such as fertilizers. This means that the land’s agricultural produc-tion depends almost solely on the agroecological context. The spatial distribution of Africa’s dominant farming systems (p. 15) is, therefore, closely aligned with the regional pat-tern of AEZs. Local agroecological conditions not only influ-ence the range of feasible agricultural enterprise options but also often strongly predict the feasibility and effectiveness of improved technologies and production practices. For this reason agriculture research and development planners are keen to understand the nature and extent of agroecological variation in the areas where they work. Planners who think in terms of AEZ boundaries rather than country or regional boundaries open up the potential for sharing knowledge and tools with people on the opposite side of the continent who work in similar AEZs. There is also growing interest in the potential consequences of agroecological change. Change might be brought about by mitigating local agroecological constraints through, for example, investments in irrigation or improved soil-water management practices. Or external fac-tors such as climate change may drive agroecological change. The likely negative economic and social implications of shift-ing agroecological patterns in Africa due to climate change are priorities for emerging research and policy research. WHAT ABOUT THE UNDERLYING DATA? The most common approaches to agroecological zone map-ping were originally developed by the Food and Agriculture Organization of the United Nations (FAO) and are still being developed and applied (for example, Fischer et al. 2009 and FAO/IIASA 2012). In Africa, the variable (and in many cases declining) quality and availability of climate-station data needed to generate reliable climatological maps is an ongoing challenge (p. 37), although increased access to satellite-derived weather and land-surface observations could ease the constraints on gathering the data in the future. The map was developed applying the regional AEZ approach using long-term average, spatially interpolated cli-mate data for Africa for the period 1960–1990 (Hijmans et al. 2005; Sebastian 2009). WHERE CAN I LEARN MORE? AEZ maps and underlying data can be downloaded at: http://hdl.handle.net/1902.1/22616 The most comprehensive collection of global AEZ-related data can be found at the FAO/ International Institute for Applied Systems Analysis Global Agro-Ecological Zones website: www.fao.org/nr/gaez/en/ 34
- 51. Data source: Sebastian2009. Note: Moisture classes are defined as follows: Arid = length of growing period (LGP) of less than 70 days; Semiarid = LGP of 70–180 days; Subhumid = LGP of 180–270 days; and Humid = LGP of greater than 270 days. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Tropic of Cancer Equator Tropic of Capricorn Arid Semiarid Humid Subhumid Subtropics - warm Subtropics - cool Tropics - warm Tropics - cool NA INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI MAP 1 Agroecological zones 35
- 52. GROWING CONDITIONS ClimateZones for Crop Management Lieven Claessens and Justin Van Wart WHAT IS THIS MAP TELLING US? Agricultural climate zones represent ecological conditions farmers face based on moisture availability, length of grow-ing period, and seasonality. Zones with little seasonal vari-ation in temperature and in wet conditions are primarily found in central Africa whereas the northernmost and southernmost countries experience high temperature sea-sonality and arid conditions. This map provides informa-tion not only on what growing conditions these agricultural climate zones present, but also on the relative size of each zone. In some countries the climate zones are quite large, such as in Mali or Niger where the weather is homoge-nous across large areas. In other countries, such as Kenya or Ghana, these zones are much smaller as agricultural systems face more diverse climates across space due to topography, proximity to the coast, and/or rainfall variation. The rela-tive size and extent of these zones offer information on the expected diversity of cropping systems within each coun-try and can be used to understand how effectively research and technology can be extrapolated to other regions. Table 1 provides a general understanding of the density and average harvested area of zones within each region. WHY IS THIS IMPORTANT? While agroecological zones (p. 34) help broadly define envi-ronments where specific agricultural systems may thrive, an agriculture climate zone seeks to more adequately dis-tinguish between the diversity of practices for similar agri-cultural systems within the larger agroecological zones, primarily in terms of different climates. A map of agricultural climate zones is a tool that can help scientists, governments, and businesses determine the best areas to boost produc-tion or focus investment. These zones help streamline tech-nology adoption and encourage innovative approaches by providing insights into the size, location, and properties of the climates where such technologies and research have improved productivity. The map also helps identify similar zones where new farming methods could be deployed in the future to increase productivity of existing cropland. Knowing the location of specific agricultural climate zones can help stakeholders target new technologies and approaches to the zones where they can make the most difference, and by extension, help meet the growing demand for food in the future. These agricultural climate zones can also be used to scale up or extrapolate and compare site-specific results, such as those obtained through field experiments or crop simulations, to larger regions or even other countries. For example, new rice management systems being developed by the AfricaRice organization for western Africa (Africa Rice Center 2011) would also be useful in south central India and central Thailand, where rice is grown in similar climate zones. WHAT ABOUT THE UNDERLYING DATA? These observations are based on the Global Yield Gap Atlas Extrapolation Domain (GYGA-ED) approach. The GYGA-ED is constructed from three variables: (1) growing degree days (GDD) with a base temperature of 0°C; (2) temperature sea-sonality (quantified as the standard deviation of monthly average temperatures); and (3) an aridity index (annual total precipitation divided by annual total potential evapotrans-piration). Each grid cell for weather data is approximately 100 km2 at the equator. Growing degree days and tempera-ture seasonality were calculated using climate data from WorldClim (Hijmans et al. 2005); the aridity index values were taken from CGIAR-CSI (Trabucco et al. 2008) (p. 54). A more extensive description and comparison with other zone schemes can be found in van Wart et al. (2013). WHERE CAN I LEARN MORE? Global Yield Gap Atlas: www.yieldgap.org Zone characterizations: “Use of Agro-Climatic Zones to Upscale Simulated Crop Yield Potential.” van Wart et al. 2013. Boosting Africa’s Rice Sector: http://bit.ly/1kBUWO3 TABLE 1 Agricultural climate zones and harvested area by region of Africa Region Number of agricultural climate zones Average harvested area per zone (000 ha) Northern Africa 72 446 Western Africa 39 2,425 Eastern Africa 71 853 Middle Africa 56 370 Southern Africa 77 80 All Africa 126 680 Data source: van Wart et al. 2013 and FAO 2012. Note: ha=hectares. 36
- 53. Data source: vanWart et al. 2013. Note: The gradient on this map reflects three factors: moisture, temperature, and seasonality. Seasonality is based on varia-tions in temperature and is quantified as the standard deviation of monthly average temperatures. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Wet Warmer Low seasonality Cooler Arid High seasonality MAP 1 Agricultural climate zones 37
- 54. GROWING CONDITIONS Rainfalland Rainfall Variability Philip Thornton WHAT ARE THESE MAPS TELLING US? An average of less than 1,000 millimeters of rain falls per year across most of Africa (Map 1). Rainfall tends to decrease with distance from the equator and is negligible in the Sahara (north of about latitude 16°N), in eastern Somalia, and in the southwest of the continent in Namibia and South Africa. Rainfall is most abundant on the eastern seaboard of Madagascar; portions of the highlands in eastern Africa; large areas of the Congo Basin and central Africa; and parts of coastal western Africa including Liberia, Sierra Leone, and Guinea. Northern Africa experiences highly variable rain-fall, except along the coasts of Algeria and Morocco (Map 2). This region’s coefficient of variation—a measure of how much rainfall varies from the annual average—is greater than 45 percent, reflecting the erratic nature of rainfall in a region that gets little precipitation. The story is similar in the extreme southwest of the continent and in pockets of the Horn of Africa. The amount of rainfall in parts of the Congo Basin is much less variable, with a coefficient of varia-tion around 10–15 percent. For most of the continent where rainfed crops are prevalent, the variability is 15–35 percent. WHY IS THIS IMPORTANT? In Africa, where most agriculture is rainfed, crop growth is limited by water availability. Rainfall variability during a grow-ing season generally translates into variability in crop produc-tion. While the seasonality of rainfall in the drier rangelands can play a significant role in productivity, rain-use efficiency (RUE)—the amount of biomass produced (in kilograms of dry matter per hectare) per millimeter of rainfall—also drives production. RUE averages about 3.0 kg of dry matter per hec-tare for every millimeter of rainfall in northern Africa, 2.7 in the Sahel, and 4.0 in eastern Africa, compared with up to 10.0 or so in temperate rangelands (Le Houérou, Bingham, and Sherbek 1988). Estimates of annual rainfall variability in the drier rangeland can offer a rough indication of pos-sible production changes. Figure 1 shows how Ethiopia’s gross domestic product echoed rainfall variability (mea-sured as a percentage variation from the long-term average) from the early 1980s to 2010. The close relationship illus-trates the importance of rainfed agricultural production to the national accounts of Ethiopia during this time period. Ethiopia is one of many countries in Africa where the econ-omy is closely tied to rainfed agriculture. WHAT ABOUT THE UNDERLYING DATA? Rainfall data are from WorldClim (Hijmans et al. 2005), an interpolated product based on average monthly cli-mate data from weather stations from 1960 to 1990. The data were aggregated to a spatial resolution of 5 arc-minutes (grid cells approximately 100 km2 at the equa-tor), and the long-term average monthly rainfall amounts add up to the annual totals (Map 1). To estimate the variability of annual rainfall (Map 2), the weather gen-erator MarkSim (Jones and Thornton 2013) was used to simulate 1,000 years of daily rainfall data for the roughly 420,000 grid cells that make up Africa and the standard deviation of annual rainfall was calculated for each grid cell and converted to the coefficient of variation. MarkSim pre-dicts rainy days and is able to simulate the variation in rain-fall observed in both tropical and temperate regions. WHERE CAN I LEARN MORE? WorldClim data. Hijmans et al. 2005: www.worldclim.org/methods Generating Downscaled Weather Data from a Suite of Climate Models for Agricultural Modelling Applications. Jones and Thornton 2013. “Evidence from Rain-use Efficiencies Does Not Indicate Extensive Sahelian Desertification.” Prince, Brown De Colstoun, and Kravitz 1998. FIGURE 1 Economic growth and rainfall variability in Ethiopia, 1982–2010 −2 −1.5 −1 −0.5 0 0.5 1 1.5 2 −20 −15 −10 −5 0 5 10 15 20 1982 1986 1990 1998 1994 2002 2006 2010 Rainfall variability (WASP) Percentage change in GDP Rainfall variability Ag GDP growth GDP growth Source: Thornton, Ericksen, and Herrero 2013; World Bank 2013; IRI/LDEO 2013. Note: WASP = the 12-month Weighted Anomaly of Standardized Precipitation. 38
- 55. Data source: Map1—WorldClim (Hijmans et al. 2005); Map 2—MarkSim (Jones and Thornton 2013). Note: Rainfall variability is represented by the coefficient of variability (CV), calculated as the standard deviation divided by the mean annual rainfall. It is expressed as a percentage and indi-cates how much rainfall varies from average annual rainfall. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT 30N 20N 10N 0 10S 20S 30S 10W 0 10E 20E 30E 40E 50E 0 500 250 1,000 2,000 1,500 2500 0 or missing value Millimeters per year 0 15 25 35 45 Percent MAP 1 Average annual rainfall MAP 2 Variability in annual rainfall 39
- 56. GROWING CONDITIONS SoilFertility Cindy Cox and Jawoo Koo WHAT ARE THESE MAPS TELLING US? Years of weathering have leached nutrients away from many soils in the cropped areas of Africa south of the Sahara (SSA). The resulting highly acidic soils ( 5.5 pH) are vulnerable to aluminum toxicity, an issue across much of Africa (Map 1), which occurs when aluminum becomes soluble and poisons plants. It is the most common soil constraint across major farming systems in SSA (Figure 1), affecting 32 percent of cropland, followed by low nutrient reserves (20 percent) and high leaching potential (12 percent). The worst soils in SSA are concentrated along the eastern coast, throughout central Africa, and scattered throughout the Sahel (Map 2). The Sahel and central Africa suffer primarily from high-leaching poten-tial and low-nutrient reserves. Some soils along eastern Africa’s coastal edges and in the Horn of Africa are calcareous, containing high levels of calcium carbonate. Such soils can be highly fertile, but extremely calcareous soils can make crops nutritionally deficient by fixing phosphorus (P), which makes it insoluble and therefore not available to plants. SSA is also home to large expanses of fertile soils that are free of constraints. WHY IS THIS IMPORTANT? About 80 percent of SSA’s cropland is not considered highly suitable for agriculture, because the extremely weathered soil limits farmers’ yields. Low-input farming further degrades soils when farmers fail to replenish nutrient reserves mined by crops. To combat poor soil, liming can increase pH and decrease acidity in soils. Breeder selection for crop variet-ies, such as beans, sorghum, and fodder crops that resist aluminum toxicity, is another way to deal with toxic soils. Furthermore, the consequences of poor soil fertility can exac-erbate other constraints, such as water uptake. Understanding where and how soils are constrained is a primary concern for the farmers and stakeholders who depend on less than ideal soil conditions and those who seek to improve their welfare. WHAT ABOUT THE UNDERLYING DATA? The underlying spatial data for major soil constraints was taken from an updated version of the Soil Functional Capacity Classification System (FCC) (HarvestChoice 2010). HarvestChoice updated the FCC by applying ver-sion 4’s methodology (Sanchez, Palm, and Buol 2003) to FAO’s Harmonized World Soil Database version 1.1. The pixel-level FCC data (Palm et al. 2007) was aggregated using HarvestChoice’s cropland extent estimate (You et al. 2012) and FAO’s Farming Systems (Dixon, Gulliver, and Gibbon 2001; p. 14). WHERE CAN I LEARN MORE? “Updating Soil Functional Capacity Classification System,” HarvestChoice 2010: http://harvestchoice.org/node/1435 Africa Soil Information Service data: http://africasoils.net FIGURE 1 Dominant soil constraint by farming system type in Africa south of the Sahara 0 2 4 6 8 10 12 14 16 18 20 22 Percent of total cropped area Agropastoral Maize mixed Cereal-root crop mixed Pastoral Root and tuber crop Humid lowland tree crop Highland mixed Highland perennial Irrigated Perennial mixed Artisanal fishing Aluminum toxicity Calcareou s Cracking clays High-leaching potential High P fixation Low nutrient reserves Poor drainage Volcanic Free of constraints Data source: Dixon, Gulliver, and Gibbon 2001; Sanchez, Palm, and Buol 2003; HarvestChoice 2010. Note: See glossary for definitions of specific soil constraints. 40
- 57. Data source: Map1—Sanchez, Palm, and Buol 2003; HarvestChoice 2010; You et al. 2012; Map 2—HarvestChoice 2010 and You et al. 2012. Note: Grid cells are approximately 100 km2 at the equator. See glossary for definitions of specific soil constraints. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Dominant soil constraint Aluminum toxicity Calcareous High-leaching potential Poor drainage High P fixation Low nutrient reserves Cracking clays Volcanic Free of constraints Outside crop growing area Outside focus area INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI Percent of grid cell 0 50 100 Outside crop growing area Outside focus area MAP 1 Dominant soil constraints within cropped areas of Africa south of the Sahara (SSA) MAP 2 Area of cell affected by soil constraints within cropped areas of SSA 41
- 58. GROWING CONDITIONS WorksCited AGROECOLOGICAL ZONES FAO (Food and Agriculture Organization of the United Nations) and IIASA (International Institute for Applied Systems Analysis). 2012. “Global Agro-ecological Zones.” www.fao.org/nr/gaez/en/. Fischer, G., M. Shah, H. van Velthuizen, and F. Nachtergaele. 2009. “Agro-ecological Zones Assessments.” In Land Use, Land Cover and Soil Sciences, Vol. III, edited by W. H. Verheye. Oxford, UK: United Nations Educational, Scientific, and Cultural Organization/ Encyclopedia of Life Support Systems. Hijmans, R. J., S. E. Cameron, J. L. Parra, P. G. Jones, and A. Jarvis. 2005. “Very High Resolution Interpolated Climate Surfaces for Global Land Areas.” International Journal of Climatology 25: 1965–1978. Sebastian, K. 2009. Agro-ecological Zones of Africa. International Food Policy Research Institute. http://hdl.handle.net/1902.1/22616. CLIMATE ZONES FOR CROP MANAGEMENT Africa Rice Center (AfricaRice). 2011. Boosting Africa’s Rice Sector: A Research for Development Strategy 2011–2020. Accessed February. 12, 2014. http://bit.ly/1kBUWO3. Hijmans, R. J., S. E. Cameron, J. L. Parra, P. G. Jones, and A. Jarvis. 2005. “Very High Resolution Interpolated Climate Surfaces for Global Land Areas.” International Journal of Climatology 25: 1965–1978. Trabucco, A., R. J. Zomer, D. A. Bossio, O. van Straaten, and L. V. Verchot. 2008. “Climate Change Mitigation through Afforestation/Reforestation: A Global Analysis of Hydrologic Impacts with Four Case Studies.” Agriculture, Ecosystems and Environment 126: 81–97. van Wart, J., L. G. J. van Bussel, J. Wolf, R. Licker, P. Grassini, A. Nelson, H. Boogaard, J. Gerber, N. D. Mueller, L. Claessens, M. K. van Ittersum, and K. G. Cassman. 2013. “Use of Agro-Climatic Zones to Upscale Simulated Crop Yield Potential.” Field Crops Research 143: 44–55. RAINFALL AND RAINFALL VARIABILITY Hijmans, R. J., S. E. Cameron, J. L. Parra, P. G. Jones, and A. Jarvis. 2005. “Very High Resolution Interpolated Climate Surfaces for Global Land Areas.” International Journal of Climatology 25: 1965–1978. IRI/LDEO (International Research Institute for Climate and Society/Lamont-Doherty Earth Observatory). 2013. Climate Data Library. http://iridl.ldeo.columbia.edu/index.html. Jones, P. G., and P. K. Thornton. 2013. “Generating Downscaled Weather Data from a Suite of Climate Models for Agricultural Modelling Applications.” Agricultural Systems 114: 1–5. Le Houérou, H. N., R. L. Bingham, and W. Sherbek. 1988. “Relationship between the Variability of Primary Production and the Variability of Annual Precipitation in World Arid Lands.” Journal of Arid Environments 15: 1–18. Prince, S. D., E. Brown De Colstoun, and L. L. Kravitz. 1998. “Evidence from Rain-Use Efficiencies Does Not Indicate Extensive Sahelian Desertification.” Global Change Biology 4: 359–374. http://bit.ly/1g522Ug. Thornton, P. K., P. J. Ericksen, and M. Herrero. 2013. “Climate Variability and Vulnerability to Climate Change: A Review.” Global Change Biology, http://bit.ly/1pIHBBZ. World Bank. 2013. World Development Indicators. http://data.worldbank.org/indicator. SOIL FERTILITY Dixon, J., A. Gulliver, and D. Gibbon. 2001. Farming Systems and Poverty: Improving Farmers Livelihoods in a Changing World. Rome: Food and Agriculture Association of the United Nations; Washington, DC: World Bank. HarvestChoice. 2010. “Updating Soil Functional Capacity Classification System.” HarvestChoice Labs. Accessed December 2, 2012. http://harvestchoice.org/node/1435. Palm, C., P. Sanchez, S. Ahamed, and A. Awiti. 2007. “Soils: A Contemporary Perspective.” Annual Review of Environment and Resources 32 (1): 99–129. Sanchez, P. A., C. A. Palm, and S. W. Buol. 2003. “Fertility Capability Soil Classification: A Tool to Help Assess Soil Quality in the Tropics.” Geoderma 114: 157–185. http://bit.ly/NNLn0L. You, L., Z. Guo, J. Koo, W. Ojo, K. Sebastian, M. T. Tenorio, S. Wood, and U. Wood-Sichra. 2012. “Spatial Production Allocation Model (SPAM) 2000, Version 3, Release 1.” Accessed December 14, 2012. http://MapSPAM.info. 42
- 59. ATLAS OF AFRICANAGRICULTURE RESEARCH DEVELOPMENT Role of Water Effects of Rainfall Variability on Maize Yields . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Blue and Green Virtual Water Flows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 Blue and Green Water Use by Irrigated Crops . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Rainfall Data Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Works Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 43
- 60. ROLE OF WATER Effects of Rainfall Variability on Maize Yields Jawoo Koo and Cindy Cox WHAT ARE THESE MAPS TELLING US? Farmers in SSA depend largely on rainfed crops for food secu-rity and their livelihood. But how reliable is rainfall across Africa, and how does the variability of rainfall from season to season affect crop yields? The following maps indicate where the variability of total rainfall in SSA (Map 1) may influence maize yields (Map 2), depending on the level of inputs such as fertilizer used in maize cultivation (Map 3 and Map 4) and the environment (Figure 1). A comparison of Maps 1 and 2 shows a correlation between rainfall and rainfed maize yields. Yields tend to correspond to seasonal fluctuations in rainfall, although in SSA, yields fluctuate year to year more than rainfall since crops are at the mercy of many other fac-tors, including total rainfall, cultural practices, pests, and soil quality. Maps 3 and 4 show how the variability in yields may be affected by changes in inputs. Figure 1 shows with more inputs, such as hybrid seeds and more fertilizer (50 kilograms of nitrogen per hectare), the probability of achieving accept-able levels of yield variability—assumed to be 25 percent or less—rises, although the effect of increased inputs, or intensification, varies by agroecological zone (p. 34). When shifting from low to high inputs, the share of total maize growing area considered more reliable—that is, exhibiting lower estimated variability in yield—rises from 20 percent to 74 percent in the subhumid and humid regions of SSA. In contrast, high inputs in arid and semi-arid regions of SSA have a smaller impact on crop reliability with a change from 11 percent to 56 percent, as the yield potential in this region, including the southern portions of Mali and Niger and cen-tral Chad, is more limited by water availability than in the humid and subhumid regions of western Africa. In some areas, such as the northern edge of the Sahel, the variability may even rise (Map 4). WHY IS THIS IMPORTANT? While estimates of yearly rainfall averages are important, yield reliability, predicted by fluctuations in growing con-ditions from year to year, concerns farmers worldwide. Knowing how rainfall variability affects yields helps stake-holders make climate-based decisions about what crops to grow, which farming systems and management practices are most suitable at a particular location, and where more invest-ments and resources are needed to improve farm productiv-ity and welfare. These may include decisions related to scaling up technologies such as irrigation, synthetic fertilizers, hybrid maize, and improved crop varieties that are more resistant to or better tolerate moisture fluctuations and drought. WHAT ABOUT THE UNDERLYING DATA? Grid-based historical daily weather and soil databases were used as inputs for the CERES-Maize model in the Decision Support System for Agrotechnology Transfer (DSSAT) v4.5 (Jones et al. 2003). Historical daily weather data for 1980–2010 generated by Elliott et al. (2014) based on the AgMIP Hybrid Baseline Climate Dataset (Ruane and Goldberg 2014) was used to retrieve site-specific solar radi-ation, temperature, and rainfall. The season-to-season vari-ability in rainfall was measured using the coefficient of variation (CV). The CV divides the standard deviation by the mean, thus indicating the likelihood that rain-fall in a given area will vary from the long-term average. A gridded soil database was derived from FAO’s Harmonized World Soil Database v1.1 (FAO et al. 2009) and the ISRIC WISE Global Soil Profile Database v1.1 (Batjes 2002). The CERES-Maize model simulated rainfed maize produc-tion across the region in areas where rainfed maize produc-tion is biophysically possible. The modeling was performed at a resolution of 5 arc-minutes, where a grid cell is approxi-mately 100 km2 at the equator. WHERE CAN I LEARN MORE? Rainfall Variability and Crop Yield Potential: http://bit.ly/1jCMRbN FIGURE 1 Variation in share of total maize growing area under varying input levels by agroecological zone Agroecological zone Arid and semiarid Variability of rainfed maize yield (%) Subhumid and humid 0 10 20 40 30 50 60 70 100 90 80 Total maize growing area (%) 56 11 74 20 Input Level Low High 0 25 50 75 100 125 150 0 25 50 75 100 125 150 Data source: Elliott et al. 2014 and Sebastian 2009. Note: The variability of rainfed maize yield is measured by the coefficient of variation (CV). 44
- 61. Data sources: Map1—Ruane and Goldberg 2014; Elliott et al. 2013; Elliott et al. 2014; Maps 2–4—Authors using DSSAT model in Hoogenboom et al. 2011; Elliott et al. 2013; Elliott et al. 2014. Notes: Rainfall variability based on seasonal total rainfall during maize growing period. Rainfed maize yield variability estimated from simulated seasonal maize yield. Low inputs = open-pollinated seeds with no fertilizer. High inputs = hybrid seeds with 50 kg nitrogen fertilizer per ha. All data simulated for 1950–1990. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Variability (CV) Low High Outside focus area Outside maize growing area Variability (CV) Low High Outside focus area Outside maize growing area Variability (CV) Low High Outside focus area Outside maize growing area Variability (CV) Low High Outside focus area Outside maize growing area INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI MAP 1 Variability of total rainfall during maize growing season MAP 2 Variability of estimated maize yields MAP 3 Variability of maize yield potential under low inputs MAP 4 Variability of maize yield potential under high inputs 45
- 62. ROLE OF WATER Blue and Green Virtual Water Flows Stefan Siebert and Petra Döll WHAT ARE THESE MAPS TELLING US? The term virtual water content refers to the volume of water used by a crop per unit of crop harvest. Virtual water flows are then determined by commodity flows between the loca-tions where crops are produced and consumed. Virtual water flows are further distinguished as flows of blue (irriga-tion) and green (precipitation stored in the soil) water. The maps show blue and green net virtual water flows caused by the production and consumption of 19 major crops (wheat, barley, rye, maize, rice, sorghum, millet, pulses, soy-beans, groundnuts, sunflower, rapeseed, potatoes, cassava, grapes, citrus, dates, cocoa, coffee). Negative values in the maps indicate a net outflow of virtual water and show major production areas where the amount of water used locally to produce crops consumed elsewhere is greater than the amount contained in crops consumed locally. Positive val-ues indicate a net inflow of virtual water to major consump-tion areas. The major irrigation regions are the source regions of blue virtual water flows (blue in Map 1) while concentra-tions of rainfed crop production are the source of green vir-tual water flows (green in Map 2). Cities and other densely populated regions represent the sinks of virtual water flows (red in Maps 1 and 2). In total, northern and southern Africa see a net inflow of both blue and green virtual water while eastern, middle, and western Africa have a net inflow of blue water but a net outflow of green water, indicating that crop imports from irrigated production compensate for exported rainfed crops (Figure 1). WHY IS THIS IMPORTANT? Production and consumption of agricultural commodities used to be local. Now, with the rapid growth in trade and urban areas, food may be produced in one place and con-sumed far away. With globalization, new links and depen-dencies between producers and consumers have formed. Demand from faraway markets for agricultural commod-ities may elevate local resource use. On the other hand, resource shortages in major production regions may result in reduced crop yields and send price signals to commod-ity markets worldwide. Mapping virtual water flows helps policymakers to better understand the importance of links between resource use and trade and of dependencies between producers and consumers of commodities. WHAT ABOUT THE UNDERLYING DATA? Crop production, crop water use, and corresponding blue and green virtual water content were computed by apply-ing the Global Crop Water Model (Siebert and Döll 2010). Crop consumption within each country was computed by adding imports of the respective crop commodity to domes-tic crop production and then subtracting the corresponding commodity exports derived from the Comtrade database for the period 1998–2002 (UN 2009). It was assumed that per capita commodity consumption is similar for all people belonging to the same country. Production surpluses and deficits within each country were leveled out by commod-ity flows (and linked virtual water flows) across increasingly larger distances and finally the whole country, if required (Hoff et al. 2014). The dataset refers to 1998–2002 and has a spatial resolution of 5 arc-minutes.1 WHERE CAN I LEARN MORE? Water Footprint Network: www.waterfootprint.org/ “Water Footprints of Cities: Indicators for Sustainable Consumption and Production.” Hoff et al. 2014: http://bit.ly/1ogjdK1 FIGURE 1 Net virtual water flows, 2000 -40 -30 -20 -10 0 10 20 30 40 -3 -2 -1 0 1 2 3 Eastern Africa Middle Africa Northern Africa Southern Africa Western Africa Net flow of green virtual water (km³ per year) Net flow of blue virtual water (km³ per year) Blue virtual water Green virtual water Data source: Hoff et al. 2014 and FAO 2012. Note: Blue virtual water=irrigation water drawn from groundwater bodies (aquifers) or surface water bodies (rivers, lakes, wetlands, or canals). Green virtual water=precipitation stored in the soil and used by rainfed and irrigated crops. Positive values represent net flows into each region. 1 Each cell measures approximately 100km2 or 10,000 hectares at the equator. 46
- 63. Data source (allmaps): Hoff et al. 2014. Note: Virtual water content refers to the volume of water used by the crop per unit of crop harvest. Virtual water flows are then established by commodity flows between the locations of crop production and crop consumption. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT −100 Outflow Inflow −25 −5 −1 15 25 100 Millimeters per year −100 Outflow Inflow −25 −5 −1 15 25 100 Millimeters per year MAP 1 Net virtual water flow of blue water (irrigation), 2000 MAP 2 Net virtual water flow of green water (precipitation stored in the soil), 2000 47
- 64. ROLE OF WATER Blue and Green Water Use by Irrigated Crops Stefan Siebert and Petra Döll WHAT ARE THESE MAPS TELLING US? In these maps, blue water refers to irrigation water while green water is precipitation stored in the soil that is also used by irrigated crops. The values refer to the amount of water that is evapotranspirated, or converted from soil water to vapor and evaporated off plant stems and leaves. Blue and green water use by irrigated crops is highest in regions with a large extent of irrigated land (p. 18), high cropping intensity (p. 28), and climate conditions causing a high evaporative demand, for example, along the Nile River, in the northern African countries of Morocco, Algeria, Tunisia, and Libya, and in South Africa (Maps 1 and 2). The contribution of blue water to total water use of irrigated crops (Map 3) depends on the aridity of the site because irrigation is mainly used to replace missing precipitation. The staple food crops with the highest irrigation water use are rice (12.1 km3 per year), wheat (11.1 km3 per year), and maize (9.0 km3 per year) (Figure 1). Combined they account for a third of the total blue water used for irrigation in Africa. More than 77 per-cent of the total irrigation water use is in northern Africa. WHY IS THIS IMPORTANT? Although only 9 percent of the harvested crop area in Africa is under irrigation, cereal production would decline by about 24 percent in Africa without the use of irrigation (Siebert and Döll 2010). This highlights the importance of irrigation for food security. On the other hand, irrigation accounts for 86 percent of global consumptive freshwater use (Döll et al. 2012) with contributions of more than 90 percent in many African countries. Availability of freshwater therefore may limit the use of irrigation in many regions. To identify regions where expanding irrigation could increase future crop pro-duction, it is necessary to consider irrigated crops’ blue water use along with freshwater availability (Bruisma 2009). Green water use is also important to consider, because blue and green water can be substituted for each other. WHAT ABOUT THE UNDERLYING DATA? Crop evapotranspiration was calculated by the Global Crop Water Model (GCWM, Siebert and Döll 2008, 2010), distin-guishing blue water use, or the evapotranspiration of irri-gation water (also called consumptive irrigation water use) from green water use (evapotranspiration of precipitation). GCWM is based on the global land use dataset MIRCA2000 (Portmann, Siebert, and Döll 2010), which provides monthly growing areas for 26 irrigated and rainfed crop classes for the period 1998–2002 and also represents multicropping. By computing daily soil water balances, GCWM determines evapotranspiration of blue and green water for each crop and grid cell. GCWM assumes that crop evapotranspira-tion of irrigated crops is always at the potential level and not restricted by water shortage. Water withdrawals for irriga-tion are higher than consumptive use because of losses and water requirements for soil preparation and salt leaching. WHERE CAN I LEARN MORE? FAO Aquastat: http://bit.ly/1dUQWqj The Global Crop Water Model (GCWM): Documentation and First Results for Irrigated Crops. Siebert and Döll 2008. FIGURE 1 Blue water use by irrigated crop and region, 1998–2002 0 2 4 6 km³ per year 8 10 12 14 16 Middle Africa Western Africa Eastern Africa Southern Africa Northern Africa Rice Wheat Maize Sugar cane Citrus Cotton Pulses Date palm Potatoes Sorghum Groundnuts Sugar beets Grapes Barley Sunflower Coffee Soybeans Rapeseed Others annual Others perennial Fodder grasses Data source: Siebert and Döll 2010 and FAO 2012. Note: Blue water use refers to the net irrigation water used by irrigated crops. 48
- 65. Data source (allmaps): Siebert and Döll 2010. Note: Blue water use refers to the net irrigation water used by irrigated crops. Green water use refers to precipitation water stored in the soil and used by irrigated crops. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT 0 0 2 10 20 50 200 500 1,700 Millimeters per year Lakes 0 0 2 10 20 50 200 260 Lakes Millimeters per year 0 20 30 40 50 60 70 80 100 Percent No irrigation Lakes MAP 1 Blue water use by irrigated crops, 2000 MAP 2 Green water use by irrigated crops, 2000 MAP 3 Contribution of blue water to total water use of irrigated crops 49
- 66. ROLE OF WATER Rainfall Data Comparison Jawoo Koo and Cindy Cox WHAT ARE THESE MAPS TELLING US? Regional rainfall data estimates for Africa can look signifi-cantly different depending on which data sources are used and how the data are analyzed. Estimates rely on precipita-tion records from a variety of land-based weather station networks with varying levels of data quality and spatio-temporal coverage. For example, Maps 1 and 2 illustrate average annual rainfall for 2000–2008 in Africa south of the Sahara (SSA) at the same 0.5° spatial resolution, but are derived from different data sources. Data from the University of East Anglia’s Climate Research Unit Time Series (CRU-TS) v3.10.01 (Map 1) shows less pixel-to-pixel variabil-ity than the University of Delaware’s Gridded Monthly Time Series (GMTS) v2.01 data (Map 2). This suggests different modeling algorithms and possibly the use of fewer observa-tions. Map 3 shows the percentage difference between the two, indicating where the rainfall estimation of GMTS is rel-atively higher (green) or lower (red) than CRU-TS. The dif-ferences are particularly evident in areas with low annual rainfall such as the Sahel, since the significance of the differ-ence between averages will be greater when average rainfall values are low. Significant differences in rainfall estimations in areas of southern Africa, particularly in Mozambique, also exist. Compared with the CRU-TS dataset, GMTS calculates a 17 percent higher average rainfall for the entire SSA region (Figure 1). WHY IS THIS IMPORTANT? Gridded climate data allow researchers to compare varia-tions in climate with other phenomena, such as crop yields or areas suitable for crop growth. Variables other than pre-cipitation— including cloud cover, diurnal temperature range, frost day frequency, daily mean temperature, and monthly average daily maximum temperature—are also available and can be used for similar comparisons. Rainfall averages and patterns are important not only to African farmers and stakeholders who rely on rainfed crops for food security and livelihoods, but also to researchers and decision-makers who need climate information to predict patterns of agricultural productivity, effects of water management technologies (such as drought-adapted crop varieties or conservation agriculture), and potential changes in climate projected over the coming decades. Climate-related datasets from different sources are not identical because of limited source data—perhaps because the network of weather sta-tions is not dense enough—and differences in interpolation methods. For this reason, researchers should not rely on just one dataset. Depending on the research questions and geographical areas of interest, the data source chosen may introduce bias to the results. If possible, researchers should compare data across multiple datasets to better understand the range of uncertainties and to avoid reaching conclusions that may inflate or understate the truth. WHAT ABOUT THE UNDERLYING DATA? Historic gridded climate databases from two sources, the University of East Anglia’s CRU-TS v3.10.01 (2013) and University of Delaware’s GMTS v2.01 (2009), were used in the mapping and intercomparison analysis for the years 2000–2008. Both datasets are based on the same 0.5 degree spatial resolution (~3,600 km2 at the equator). Annual rain-fall data were computed for each grid cell, and their average values across the years were mapped and compared with each other. WHERE CAN I LEARN MORE? University of East Anglia CRU climate data: www.cru.uea.ac.uk University of Delaware’s Gridded Monthly Time Series (GMTS) v2.01 data: http://bit.ly/1m7HvHk FIGURE 1 Distribution of average annual total rainfall from two climate data sources CRU-TS v3.10.01 GMTS v2.01 0 1,000 2,000 3,000 4,000 Average annual total rainfall (mm) 180 1,180 233 1,362 680 797 Source: University of East Anglia 2013 and University of Delaware 2009. Note: Each bar indicates a grid-cell-level value. The dotted black line indicates the average across the SSA region, and the gray area shows +1/-1 standard deviation. The y-axis precipitation totals are nine-year averages (2000–2008) at 0.5° grid cells. 50
- 67. Data sources: Map1—University of East Anglia 2013; Map 2—University of Delaware 2009; Map 3—Calculation based on University of East Anglia 2013 and University of Delaware 2009. Note: Each grid cell measures 0.5 degrees or ~3,600 km2 at the equator. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT mm per year per grid cell Low: 0 High: 2,655 Outside focus area mm per year per grid cell Low: 0 High: 3,911 Outside focus area −50 −25 −10 0 10 25 50 75 100 Percent Outside focus area INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI MAP 1 Average annual rainfall for 2000–2008 from the University of East Anglia CRU-TS MAP 2 Average annual rainfall for 2000–2008 from the University of Delaware GMTS MAP 3 Difference between CRU-TS and GMTS average annual rainfall, based on CRU-TS 51
- 68. ROLE OF WATER Works Cited EFFECTS OF RAINFALL VARIABILITY ON MAIZE YIELDS Batjes, N. H. 2002. A Homogenized Soil Profile Data Set for Global and Regional Environmental Research (WISE, Version 1.1). Report 2002/01. Wageningen, The Netherlands: International Soil Reference and Information Centre. http://bit.ly/1gNrrDR. Elliott, J., M. Glotter, N. Best, J. Chryssanthacopoulos, D. Kelly, M. Wilde, and I. Foster. 2014. “The Parallel System for Integrating Impact Models and Sectors (pSIMS).” Prepared for a special issue of Environmental Modeling Software, forthcoming. Elliott, J., D. Kelly, N. Best, M. Wilde, M. Glotter, and I. Foster. 2013. “The Parallel System for Integrating Impact Models and Sectors (pSIMS).” In Proceedings of the XSEDE13 Conference: Gateway to Discovery, chaired by N. Wilkins-Diehr, San Diego, CA, July 22–25. New York: Association for Computing Machinery. FAO (Food and Agriculture Organization of the United Nations), IIASA (International Institute for Applied Systems Analysis), ISRIC-World Soil Information, ISSCAS (Institute of Soil Science-Chinese Academy of Sciences), and JRC (Joint Research Centre of the European Commission). 2009. Harmonized World Soil Database (Version 1.1). Rome: FAO; Laxenburg, Austria: IIASA. http://bit.ly/1ljrHQy. Hoogenboom, G., J. W. Jones, P. W. Wilkens, C. H. Porter, K. J. Boote, L. A. Hunt, U. Singh, J. L. Lizaso, J. W. White, O. Uryasev, F. S. Royce, R. Ogoshi, A. J. Gijsman, G. Y. Tsuji, and J. Koo. 2011. Decision Support System for Agrotechology Transfer (DSSAT) Version 4.5.1.023. http://bit.ly/1i7FvgF. Jones, J. W., G. Hoogenboom, C. H. Porter, K. J. Boote, W. D. Batchelor, L. A. Hunt, P. W. Wilkens, U. Singh, A. J. Gijsman, and J. T. Ritchie. 2003. “The DSSAT Cropping System Model.” European Journal of Agronomy 18: 235–265. Ruane, A. C., and R. Goldberg. 2014. “AgMIP Hybrid Baseline Climate Datasets: Shifted Reanalyses for Gap-filling and Historical Climate Series Estimation.” Unpublished, National Aeronautics and Space Administration, Washington, DC. Sebastian, K. 2009. Agro-ecological Zones of Africa. International Food Policy Research Institute. http://hdl.handle.net/1902.1/22616. BLUE AND GREN VIRTUAL WATER FLOWS FAO (Food and Agriculture Organization of the United Nations). 2012. FAOSTAT Database. http://faostat.fao.org/site/291/default.aspx. Hoff, H., P. Döll, M. Fader, D. Gerten, S. Hauser, and S. Siebert. 2014. “Water Footprints of Cities: Indicators for Sustainable Consumption and Production.” Hydrology and Earth System Sciences 18: 213–226. http://bit.ly/1ogjdK1. Siebert, S., and P. Döll. 2010. “Quantifying Blue and Green Virtual Water Contents in Global Crop Production as Well as Potential Production Losses without Irrigation.” Journal of Hydrology 384 (3–4): 198–217. UN (United Nations). 2009. United Nations Commodity Trade Statistics Database. Accessed on January 21, 2009. http://comtrade.un.org. BLUE AND GREN WATER USE BY IRRIGATED CROPS Bruinsma, J. 2009. The Resource Outlook to 2050: By How Much Do Land, Water and Crop Yields Need to Increase by 2050? Rome: Food and Agriculture Organization of the United Nations. http://bit.ly/NNLJEy. Döll, P., H. Hoffmann-Dobrev, F. T. Portmann, S. Siebert, A. Eicker, M. Rodell, G. Strassberg, and B. R.Scanlon. 2012. “Impact of Water Withdrawals from Groundwater and Surface Water on Continental Water Storage Variations.” Journal of Geodynamics 59–60: 143–156. FAO (Food and Agriculture Organization of the United Nations). 2012. FAOSTAT Database. http://faostat.fao.org/site/291/default.aspx. Portmann, F. T., S. Siebert, and P. Döll. 2010. “MIRCA2000-Global Monthly Irrigated and Rainfed Crop Areas around the Year 2000: A New High-Resolution Data Set for Agricultural and Hydrological Modeling.” Global Biogeochemical Cycles 24 (1). Siebert, S., and P. Döll. 2008. The Global Crop Water Model (GCWM): Documentation and First Results for Irrigated Crops. Frankfurt, Germany: Goethe University. http://bit.ly/1kOQHvS. Siebert, S., and P. Döll. 2010. “Quantifying Blue and Green Virtual Water Contents in Global Crop Production as Well as Potential Production Losses Without Irrigation.” Journal of Hydrology 384 (3–4): 198–217. RAINFALL DATA COMPARISON University of East Anglia Climatic Research Unit. 2013. CRU TS3.10: Climatic Research Unit (CRU) Time-Series (TS) Version 3.10 of High Resolution Gridded Data of Month-by- month Variation in Climate (Jan. 1901–Dec. 2009). www.cru.uea.ac.uk/. University of Delaware, Center for Climatic Research, Department of Geography. 2009. “Terrestrial Precipitation: 1900–2008 Gridded Monthly Time Series (v2.01).” Accessed on September 13, 2011. http://bit.ly/1m7HvHk. 52
- 69. ATLAS OF AFRICANAGRICULTURE RESEARCH DEVELOPMENT Drivers of Change Influence of Aridity on Vegetation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Impacts of Climate Change on Length of Growing Period . . . . . . . . . . . . . . 56 Maize Yield Potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 Wheat Stem Rust Vulnerability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Benefits of Trypanosomosis Control in the Horn of Africa . . . . . . . . . . . . . 62 Works Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 53
- 70. DRIVERS OF CHANGE Influence of Aridity on Vegetation Antonio Trabucco and Robert Zomer WHAT IS THIS MAP TELLING US? The aridity index measures the adequacy of the precipita-tion to satisfy vegetation water requirements. Large areas of northern and southern Africa are dry with an aridity index of less than 0.65. In contrast, central Africa is more humid, with an aridity index that exceeds 0.65. Variations in dry-ness reflect Africa’s geography and topography. For example, hyperarid zones, such as the Sahara and Namibia deserts, which receive less than 100 mm of precipitation annually, correspond to prevailing high pressure systems preventing cloud formation over the western edges of subtropical areas. Equatorial areas are more humid than other parts of Africa, because low pressure systems and strong air convection con-dense the moisture into clouds, which lead to high precipi-tation. Dry northeast monsoon winds blowing in from the Arabian Desert make eastern Africa less humid than other equatorial regions, such as central Africa and the Gulf of Guinea, to the west. Mountains, such as Mt. Kenya and Mt. Kilimanjaro, block the passage of rain-producing weather systems, creating more humid conditions in highland areas and drier conditions on the shielded side of these highlands. WHY IS THIS IMPORTANT? More than half of Africa’s population lives in arid, semiarid, or dry subhumid areas. This means nearly 600 million peo-ple spread across 75 percent of the continent’s land area live under ecological conditions where subsistence agriculture may be only partially suitable. In such regions, people may find it difficult to increase incomes from agriculture and improve food security. In fact, there is a direct correlation between aridity and prevailing vegetation and land use (Figure 1). While humid conditions encourage plant growth, arid conditions do not. One way plants adapt to the lack of rain is by limiting their growth. Figure 1 shows the natural process where ecosys-tems evolve from bare land to herbaceous areas, shrub land, and forests, as more humid conditions prevail. Land use, in turn, also reflects human needs. In particular, agriculture fol-lows specific patterns according to aridity. In semiarid areas farmers rely mainly on rainfed subsistence agriculture, which limits crop yields unless irrigation is adopted. In contrast, highly productive agriculture systems are found in places with more humid conditions, such as in southern Nigeria. WHAT ABOUT THE UNDERLYING DATA? Because precipitation alone does not properly character-ize vegetation water stresses across large regions, an aridity index is calculated as the ratio of annual precipitation to potential evapotranspiration (PET). Thus, the aridity index measures how much rainfall is available to satisfy the water demand of a type of vegetation. Using this formula, arid-ity index values increase with more humid conditions and decrease with more arid conditions. Annual precipitation was derived from the WorldClim database (Hijmans 2005). PET was calculated using the Hargreaves method applied to temperature parameter layers from the WorldClim database and extraterrestrial radiation (Allen et al. 1998; Trabucco and Zomer 2009). Although the aridity index map reflects average conditions between 1950 and 2000, rainfall in arid and semiarid regions is highly variable across space and time (Map 2, p. 39). This variability relates to the randomness of prevailing convective rains in arid regions, where short, heavy storms can either hit or miss an area. WHERE CAN I LEARN MORE? Global Aridity and PET (Potential Evapo-Transpiration) Database: http://bit.ly/1hYD3Iv “The Climatology of Sub-Saharan Africa.” Nicholson 1983. Crop Evapotranspiration: Guidelines for Computing Crop Water Requirements: http://bit.ly/1kCFdzq “Carbon Sequestration in Dryland Soil,” Chapter 2 in The World’s Drylands: http://bit.ly/13HBTpc FIGURE 1 Land cover types, by aridity 0 20 40 60 80 100 Area share by land cover (%) Aridity index Bare land Cropland Herbaceous Shrub land Forest 0.05 0.05–0.09 0.10–0.19 0.20–0.34 0.35–0.49 0.50–0.64 0.65–0.79 0.80–0.99 1.00–1.24 1.25 Source: Trabucco and Zomer 2009. Note: Aridity index = precipitation (mm)/potential evapotranspiration (PET mm). 54
- 71. Data source: Trabuccoand Zomer 2009. Note: Aridity Index=precipitation (mm)/potential evapotranspiration (PET mm). The aridity index classes are based on United Nations Environment Programme classifications (UNEP 1997). ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Tropic of Cancer Equator Tropic of Capricorn 0.05 hyperarid arid semiarid dry subhumid humid 0.10 0.20 0.35 0.50 0.65 0.80 1.00 1.25 0 Aridity Index MAP 1 Aridity index 55
- 72. DRIVERS OF CHANGE Impacts of Climate Change on Length of Growing Period Philip Thornton WHAT ARE THESE MAPS TELLING US? Projections show that climate changes between now and the 2050s may significantly affect the length of growing periods (LGP) in Africa. LGP, expressed as number of days per year, is a metric that integrates rainfall, temperature, and some soil conditions to determine when crops grow in cer-tain areas (Map 1). It is a useful proxy for season type in the water-limited conditions that prevail in many parts of the tropics. LGP ignores intervening drought periods and so it is not always a good indicator of cropping success, but it is often highly correlated with yields. Map 2 shows the pro-jected percentage change in LGP in the 2050s compared with current conditions, using a scenario of high greenhouse gas emissions and several global climate models. Most of the continent will see reductions in LGP, some of them severe. Parts of eastern Africa, particularly the Horn of Africa, may see some increases, but in these areas, current LGP is low (90 days or less, Map 1). The climate models used to proj-ect LGP do not all agree on how the climate may change by 2050. Map 3 shows the variability in projections for LGP esti-mated from several climate models. Since areas with lower values, such as much of central Africa, show more agree-ment between the various climate models, one can have more confidence in projected LGP changes in these areas. In areas with higher values, the climate models agree less, meaning the projections of LGP change are less reliable. WHY IS THIS IMPORTANT? To effectively adapt to climate change, farmers, governments, and other stakeholders must understand the potential effects on crop and livestock production. A contracted growing sea-son can impact crop and livestock productivity, particularly in areas where growing seasons are already short. Temperature increases and rainfall changes could push some of these areas to a point where cropping may fail in most years. Some farm-ers may be able to adapt to shorter growing seasons by plant-ing varieties that mature more quickly; other farmers may need to change to more drought- and heat-tolerant crops. Increase in LGP may present more growing opportunities, but it is uncertain how the change in growing time would impact soil moisture. As climate changes, the distribution of crop pests and diseases may change, too. Of course, LGP is only one metric; the information shown here can be combined with or compared to other aspects of projected climate change—such as temperature changes—to create a more detailed picture of how climatic shifts could affect crop growth and development. WHAT ABOUT THE UNDERLYING DATA? The data are from downscaled climate projections. Because differences between climate models may be quite large, par-ticularly for projected changes in rainfall patterns and quan-tities, the means of six climate models (Table 1) form the basis for generating daily weather data sequences plausible for future climatologies. Jones and Thornton (2013) provide details of the models used and the methods applied. LGP is calculated daily using a water balance model that calcu-lates available soil water, runoff, water deficiency, and the ratio of actual to potential evapotranspiration (Ea/Et). The growing period begins with 5 consecutive growing days and ends with 12 consecutive nongrowing days; a growing day has an average air temperature greater than 6⁰C and Ea/Et exceeding 0.35. WHERE CAN I LEARN MORE? Methods used to develop this data and create these maps: www.ccafs-climate.org/pattern_scaling/ More information on the effects of climate change: Easterling et al. 2007. Details on models used and methods applied: Jones and Thornton 2013; and Jones, Thornton, and Heinke 2009. TABLE 1 Atmosphere–Ocean General Circulation Models used to estimate LGP changes to the 2050s Model Name (Vintage) Institution Resolution (degrees) BCCR_BCM2.0 (2005) Bjerknes Centre for Climate Research 1.9 × 1.9 CNRM-CM3 (2004) Météo-France/Centre National de Recherches Météorologiques, France 1.9 × 1.9 CSIRO-Mk3_5 (2005) Commonwealth Scientific and Industrial Research Organisation Atmospheric Research 1.9 × 1.9 ECHam5 (2005) Max Planck Institute for Meteorology 1.9 × 1.9 INM-CM3_0 (2004) Institute Numerical Mathematics 4.0 × 5.0 MIROC3.2 (medres) (2004) Center for Climate System Research, National Institute for Environmental Studies, and Frontier Research Center for Global Change 2.8 × 2.8 Ensemble average Average climatology of the above models Source: For model details, see Randall et al. 2007. 56
- 73. Data source (allmaps): Jones et al. 2009. Note: LGP variability is represented by the coefficient of variation (CV), calculated as the standard deviation divided by the average LGP, expressed as a percentage. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT 1 30 60 90 120 150 180 210 240 0 or missing value Number of days per year −20 −75 −5 0 5 Percent 1 30 60 90 120 0 or missing value Percent MAP 1 Average length of growing period (LGP) for current conditions, 2000s MAP 2 Projected mean change in length of growing period (LGP) in 2050 MAP 3 Variability among length of growing period (LGP) projected values for 2050 57
- 74. DRIVERS OF CHANGE Maize Yield Potential Jawoo Koo WHAT ARE THESE MAPS TELLING US? Map 1 portrays the broad spatial distribution of farm-level rainfed maize production in Africa south of the Sahara. While South Africa, the region’s largest producer, consistent-ly achieves national average yields in excess of 4 tons per hectare (t/ha), the best performing of the remaining countries, including major producers such as Ethiopia, Malawi, and Zambia, typically average only around 2 t/ha. Farmers in other large-producer nations, notably Nigeria, Tanzania, and Kenya, have lower yields, around the regional norms of 1.3–1.7 t/ha. Map 2 shows potential rainfed maize yield, or the modeled patterns of achievable yields if key yield constraints, in this case soil nutrient deficiencies, could be overcome. If concerted development efforts helped to achieve this goal, approximately 55 percent of the cur-rent maize production area could attain yields in excess of 3 t/ha, a threshold that signals the basic subsistence cereal needs of smallholder families can likely be met, assuming typical farm holdings and family size (UN Millennium Project 2005). The gap between actual and potential yields tends to vary systematically by production environ-ment (Figure 1). In drier regions (those areas with less than 500 millimeters of rainfall per year, such as the Sahel), estimated yield gaps are relatively modest, because the lack of rainfall remains a key limiting factor to increased yields even if soil fertility is improved. WHY IS THIS IMPORTANT? The maize yield analysis and mapping shown here helps tar-get and prioritize specific geographic areas where research-ers and farmers can work to overcome common sets of production constraints to enhance local livelihoods and food security. Yield potential in many areas is much higher than what farmers now achieve. In areas with higher levels of rainfall, improving soil quality can provide much bigger payoffs for farmers. However, reducing soil nutrient deficien-cies is not a cure-all; other challenges, such as the increas-ing prevalence of pests and weed competition need to be addressed. Also, some production areas are inherently less suited to maize production using existing technologies and practices. Particularly in drier areas, farmers may already be achieving the potential that current varieties can sup-port without further investments in small-scale irrigation or more drought-tolerant varieties. WHAT ABOUT THE UNDERLYING DATA? Historical daily weather and soil databases generated by HarvestChoice were used as inputs to the DSSAT v4.5 CERES -Maize model (Jones et al. 2003; Hoogenboom et al. 2012) in order to simulate yields across a 5 arc-minute grid (with ~100km2 grid cells at the equator) covering Africa. Historical monthly rainfall data for 1950–90 were extracted from the University of East Anglia CRU-TS v3.10 database (UEA 2011) and temporally downscaled to daily weather by applying satellite-observed daily rainfall patterns for 1997–2008 retrieved from the NASA-POWER Agroclimatic Database. Gridded soil texture classes (sandy, loamy, and clayey) were extracted from the updated Soil Functional Capacity Classification (FCC) System (HarvestChoice 2010a). Achieved yields were extracted from the SPAM database (You et al. 2012). WHERE CAN I LEARN MORE? Spatial Production Allocation Model: http://mapspam.info Synthesized 100-Year Weather Data. HarvestChoice 2010b: http://harvestchoice.org/node/1441 Updating Soil Functional Capacity Classification Systems HarvestChoice 2010: http://harvestchoice.org/node/1435 FIGURE 1 Actual (2000) vs. potential maize yields, Africa south of the Sahara Average rainfall: millimeters/year (Percent of maize area) Maize yield (tons per hectare) 0 500 (7%) 500–1000 (49%) 1001–1500 (37%) 1501–2000 (6%) 2000 (1%) 2 4 6 8 10 Actual yield Potential yield Data source: Actual yield–You et al. 2012; potential yield–author’s calculations. Note: Bars indicate the average yield in each annual rainfall category weighted with maize harvest area, and the error bars indicate one standard deviation. Per-centages in parentheses indicate the approximate share of maize production area in each rainfall category. 58
- 75. Data sources: Map1—You et al. 2012; Map 2—Author. Note: t/ha=tons per hectare. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT High (16 t/ha) Low (0 t/ha) Outside focus area Outside potential growing area High (16 t/ha) Low (0 t/ha) Outside focus area Outside actual growing area INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI MAP 1 Actual rainfed maize yield, c. 2000 MAP 2 Potential rainfed maize yield 59
- 76. DRIVERS OF CHANGE Wheat Stem Rust Vulnerability Yuan Chai and Jason Beddow WHAT IS THIS MAP TELLING US? Much of the wheat-growing area of Africa is susceptible to stem rust, a fungal disease of wheat. The map shows where crops might be vulnerable to stem rust infection. Almost all African areas where wheat production is relatively concen-trated are vulnerable to the disease, including the north-ern growing areas in Morocco, Tunisia, Algeria, and the Nile Valley, along with major growing areas in central Ethiopia, southern Kenya, and South Africa. The map shows the dis-ease’s potential to pose a problem if wheat were grown throughout the continent, although it is not grown in all of the colored areas. In a typical year, the pathogen can per-sist year-round in the red areas, infecting wheat, rye, and bar-ley. The climate of the blue areas is also hospitable to the pathogen, but it cannot survive the entire year in those loca-tions, usually because they become too hot, cold, or dry. For infection to occur in these areas, the pathogen must be trans-ported (primarily by wind) to the area each year. WHY IS THIS IMPORTANT? Stem rust negatively affects food security by limiting wheat production, which increases food prices. Though much of the wheat grown worldwide is somewhat resis-tant to the disease, most of the older cultivars used by many low-input farmers in Africa and elsewhere offer lit-tle resistance. Further, most of the world’s wheat varieties have little resistance to new strains of the stem rust patho-gen, collectively known as Ug99, that were first discov-ered in Uganda in 1998. These new strains could severely shrink global wheat supplies. From its emergence in Uganda, Ug99 has spread to infect wheat crops grown in other African countries, including major wheat-producing coun-tries such as Kenya, Ethiopia, and South Africa. On average, Africa’s wheat-growing areas are highly susceptible to stem rust compared with global norms (Table 1). Based on these estimates along with the cereal crop distributions (p. 20), about 64 percent of the world’s wheat area, representing 71 percent of global wheat output, is climatically vulnerable to stem rust infec-tion, and the disease can persist year-round in about 13 percent of that area. By contrast, 90 percent of Africa’s wheat-growing area, representing 87 percent of its wheat output, is susceptible to stem rust, and the dis-ease can persist year-round in about 71 percent of the con-tinent’s wheat-growing area, representing 67 percent of Africa’s wheat output. Thus, not only is Africa’s wheat crop more vulnerable to stem rust infection, the disease is more likely to be present every year. WHAT ABOUT THE UNDERLYING DATA? Global estimates of climatic suitability were derived by modeling the response of the stem rust pathogen, Puccinia graminis, to climatic factors such as soil moisture and temperature as described by Beddow et al. (2013a). For each 10 arc-minute pixel (~344 km2 at the equator) globally, the model was used to estimate the relative cli-matic suitability for the pathogen to infect a crop host during the growing season (vulnerability) and to sur-vive year-round (persistence). WHERE CAN I LEARN MORE? Puccinia graminis. Beddow et al. 2013a. Measuring the Global Occurrence and Probabilistic Consequences of Wheat Stem Rust. Beddow et al. 2013b. Potential Global Pest Distributions Using Climex: HarvestChoice Applications. Beddow et al. 2010. Right-Sizing Stem-Rust Research. Pardey et al. 2013. Tracking the movement of Ug99—CIMMYT Rust Tracker: http://rusttracker.cimmyt.org TABLE 1 Stem rust vulnerability and persistence in Africa and major wheat-growing areas of the world Region Vulnerable to stem rust Persistent year-round Area (%) Output (%) Area (%) Output (%) China 91.6 90.6 6.6 3.9 India 60.6 63.0 2.8 1.2 United States 53.5 56.6 0.7 1.1 Russia 22.8 21.9 0.0 0.0 Africa 90.0 86.9 70.6 66.6 Global 63.8 71.2 12.6 9.4 Data source: Calculated based on Beddow et al. 2013b and You et al. 2012. Note: The percentages show the portions of the wheat harvested area (area %) and wheat produced (output %) that are susceptible to stem rust infection. Au-thors’ calculations based on stem rust potential and harvested area and annual production for wheat. The climate in vulnerable areas allows the pathogen to infect a host during the growing season. Persistent year-round=areas where the pathogen can become established and survive year-round. 60
- 77. Data source: Beddowet al. 2013b. Notes: Seasonally vulnerable = areas in which the pathogen can grow during the favorable season but cannot survive year-round. Persistently vulnerable = areas where the pathogen can become established and survive year-round. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Seasonally vulnerable Persistently vulnerable MAP 1 Areas vulnerable to wheat stem rust 61
- 78. DRIVERS OF CHANGE Benefits of Trypanosomosis Control in the Horn of Africa Timothy Robinson, Giuliano Cecchi, William Wint, Raffaele Mattioli, and Alexandra Shaw WHAT ARE THESE MAPS TELLING US? Using the Horn of Africa as an example, the maps illustrate different steps in a methodology developed to estimate and map the economic benefits to livestock keepers of con-trolling a disease (Shaw et al. 2014). Cattle are first assigned to different production systems as shown in Map 1, illus-trating for example, where mixed farming is heavily depen-dent on the use of draft oxen in Ethiopia, areas of Sudan and South Sudan where oxen use is much lower, and the strictly pastoral areas of Somalia and Kenya. Information on the location of cattle and production systems is combined with the distribution of tsetse fly species in the area (Map 2) to estimate the presence and absence of trypanosomosis, a par-asitic disease transmitted by the tsetse fly. Herd growth and spread is modelled for the current situation, and for the sim-ulated removal of trypanosomosis. The outputs of the model are then presented as a map of the financial benefits to live-stock keepers that would be realized from trypanosomosis removal, expressed as US$ per km2 (Map 3). The estimated total maximum benefit to livestock keepers, interpreted also as the maximum level of losses avoided, in the Horn of Africa amounts to nearly $2.5 billion, discounted at 10 percent over 20 years to account for the opportunity cost of funds— an average of approximately $3,300 per square kilometer of tsetse-infested area (Table 1). Map 3 shows how these benefits vary spatially. WHY IS THIS IMPORTANT? African animal trypanosomosis reduces the productivity of livestock, especially cattle, when it sickens or kills them. It also affects rural development and livelihoods more gener-ally by limiting options for mixed farming and hindering a balanced use of natural resources. Moreover, in many areas the parasite causes sleeping sickness in people; a highly debilitating disease which if not treated is lethal. Deciding where and how to intervene against this disease requires knowledge of relevant socioeconomic dimensions, such as poverty levels (p. 76) and the role of livestock in people’s livelihoods. The map of potential benefits from trypanoso-mosis removal in the Horn of Africa can help decisionmak-ers prioritize interventions by highlighting areas, such as Ethiopia, South Sudan and Kenya, where the financial return on investments to control the disease would be highest (Table 1). WHAT ABOUT THE UNDERLYING DATA? The model used information on cattle densities and produc-tion systems to account for herd growth and spatial spread of cattle over a 20-year period. For this analysis, pastoral, agro-pastoral, and mixed farming systems, as described in Cecchi et al. (2010), were further characterized to measure dairy and draft power in the Horn of Africa, using reported statistics on improved cattle that were cross-bred with higher yield variet-ies and on the use of draft oxen. The cattle distribution map used for the analysis was an earlier version of that presented for the whole of Africa (Map 1, p. 27). The predicted presence of six tsetse fly species of veterinary importance in eastern Africa at one kilometer resolution (Wint 2001) were com-bined into a single regional map that predicts the absence or presence of the genus Glossina (tsetse fly). Shaw et al. (2014) describe the herd model used and the detailed data on herd parameters with and without trypanosomosis in the region. WHERE CAN I LEARN MORE? “Mapping the Economic Benefits to Livestock Keepers from Intervening Against Bovine Trypanosomosis in Eastern Africa.” Shaw et al. 2014. “Geographic Distribution and Environmental Characterization of Livestock Production Systems in Eastern Africa.” Cecchi et al. 2010. TABLE 1 Projected maximum benefits (US$) over 20 years of eliminating bovine trypanosomosis Country Area of tsetse infestation (000 km2) Total benefit from absence of trypanosomosis (US$ million) Average benefit per km2 infested (US$) Ethiopia 157 834 5,317 Kenya 129 590 4,576 Somalia 38 158 4,181 South Sudan and Sudan 310 485 1,564 Uganda 103 390 3,786 Total 737 2,457 3,335 Source: Shaw et al. 2014. Note: The total benefit represents the cumulative amount of money accrued over 20 years, discounted at 10 percent to account for the opportunity cost of funds, if trypanosomosis were removed in these countries over a five-year period. 62
- 79. Data source: Maps1 and 2—Shaw et al. 2014; Map 3—Calculation based on Wint 2001. Note: Map 3—The benefit=total amount of money accrued over 20 years, discounted at 10 percent to account for the opportunity cost of funds if trypanosomosis were eliminated in a five-year period. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Sudan Eritrea Ethiopia Somalia Kenya Uganda South Sudan 15N 0 30E 45E Djibouti 0 500 1,000 Kilometers Sudan Eritrea Ethiopia Somalia Kenya Uganda South Sudan 15N 0 30E 45E Djibouti !! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! 0 500 1,000 Kilometers 30E 45E 0 250 500 Kilometers 0 Sudan Ethiopia Somalia Uganda Kenya South Sudan Djibouti Pastoral Areas unsuitable for ruminants Mixed farming (general) High dairy High oxen Medium oxen Low oxen Agropastoral High dairy High oxen Medium oxen Low oxen Mixed farming (Ethiopia) Low oxen High oxen Medium oxen Outside study area Absence Outside study area Presence Predicted distributon of tsetse flies 0 500 10 1,000 5,000 2,500 7,500 10,000 12,500 Unsuitable for ruminants Outside study area US$ per km² over a 20-year period MAP 3 Potential benefits of eliminating bovine trypanosomosis MAP 1 Cattle production systems MAP 2 Tsetse fly distribution 63
- 80. DRIVERS OF CHANGE Works Cited INFLUENCE OF ARIDITY ON VEGETATION Allen, R. G., L. S. Pereira, D. Raes, and M. Smith. 1998. Crop Evapotranspiration—Guidelines for Computing Crop Water Requirements. FAO Irrigation and Drainage Paper 56. Rome: Food and Agriculture Organization of the United Nations. Hijmans, R. J., S. E. Cameron, J. L. Parra, P. G. Jones, and A. Jarvis. 2005. “Very High Resolution Interpolated Climate Surfaces for Global Land Areas.” International Journal of Climatology 25: 1965–1978. Nicholson, S. 1983. “The Climatology of Sub-Saharan Africa.” In Environmental Change in the West African Sahel. Advisory Committee on the Sahel, 71–92. Washington, DC: National Academy Press. Trabucco, A., and R. J. Zomer. 2009. Global Aridity and PET (Potential Evapotranspiration) Database. CGIAR Consortium for Spatial Information. Accessed on September 25, 2013. http://bit.ly/1hYD3Iv. UNEP (United Nations Environment Programme). 1997. World Atlas of Desertification, 2nd ed. London: UNEP. IMPACTS OF CLIMATE CHANGE ON LENGTH OF GROWING PERIOD Easterling, W. E., P. K. Aggarwal, P. Batima, K. M. Brander, L. Erda, S. M. Howden, A. Kirilenko, J. Morton, J.-F. Soussana, J. Schmidhuber, and F. N. Tubiello. 2007. “Food, Fibre and Forest Products.” In Climate Change 2007: Impacts, Adaptation and Vulnerability: Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, edited by M. L. Parry, O. F. Canziani, J. P. Palutikof, P. J. van der Linden, and C. E. Hanson, 273-313. Cambridge, UK, and New York: Cambridge University Press. Jones, P. G., and P. K. Thornton. 2013. “Generating Downscaled Weather Data from a Suite of Climate Models for Agricultural Modelling Applications.” Agricultural Systems 114: 1–5. Jones, P. G., P. K. Thornton, and J. Heinke. 2009. Generating Characteristic Daily Weather Data Using Downscaled Climate Model Data from the IPCC’s Fourth Assessment. Project Report. Nairobi: International Livestock Research Institute. Randall, D. A., R. A. Wood, S. Bony, R. Colman, T. Fichefet, J. Fyfe, V. Kattsov, A. Pitman, J. Shukla, J. Srinivasan, R. J. Stouffer, A. Sumi, and K. E. Taylor. 2007. “Climate Models and Their Evaluation.” In Climate Change 2007: The Physical Science Basis: Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, edited by S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor, and H. L. Mille. Cambridge, UK, and New York: Cambridge University Press. MAIZE YIELD POTENTIAL HarvestChoice. 2010a. “Updating Soil Functional Capacity Classification System.” HarvestChoice Labs. Accessed December 2, 2012. http://harvestchoice.org/node/1435. —. 2010b. “SLATE: Synthesized 100-Year Weather Data for Sub-Saharan Africa.” Washington, DC: International Food Policy Research Institute; St. Paul, MN: University of Minnesota. http://harvestchoice.org/node/1441. Hoogenboom, G., J. W. Jones, P. W. Wilkens, C. H. Porter, K. J. Boote, L. A. Hunt, U. Singh, J. L. Lizaso, J. W. White, O. Uryasev, F. S. Royce, R. Ogoshi, A. J. Gijsman, G. Y. Tsuji, and J. Koo. 2012. Decision Support System for Agrotechnology Transfer (DSSAT) Version 4.5.1.023. Honolulu: University of Hawaii. CD-ROM. Jones, J. W., G. Hoogenboom, C. H. Porter, K. J. Boote, W. D. Batchelor, L. A. Hunt, P. W. Wilkens, U. Singh, A. J. Gijsman, and J. T. Ritchie. 2003. “The DSSAT Cropping System Model.” European Journal of Agronomy 18: 235–265. UN Millennium Project. 2005. Investing in Development: A Practical Plan to Achieve the Millennium Development Goals. London and Sterling, VA, US: Earthscan. University of East Anglia Climatic Research Unit. 2011. CRU TS3.10: Climatic Research Unit (CRU) Time-Series (TS) Version 3.10 of High Resolution Gridded Data of Month-by- month Variation in Climate (Jan. 1901–Dec. 2009). Accessed on September 13, 2011. http://badc.nerc.ac.uk/home/index.html. You, L., Z. Guo, J. Koo, W. Ojo, K. Sebastian, M. T. Tenorio, S. Wood, and U. Wood-Sichra. 2012. “Spatial Production Allocation Model (SPAM) 2000, Version 3, Release 1.” Accessed December 14, 2012. http://MapSPAM.info. WHEAT STEM RUST VULNERABILITY Beddow, J. M., T. M. Hurley, D. J. Kriticos, and P. G. Pardey. 2013b. Measuring the Global Occurrence and Probabilistic Consequences of Wheat Stem Rust. Washington, DC and St. Paul, MN, US: HarvestChoice. http://bit.ly/1eXggG0. Beddow, J. M., D. Kriticos, P. G. Pardey, and R. W. Sutherst. 2010. Potential Global Crop Pest Distributions Using Climex: HarvestChoice Applications. Washington, DC, and St. Paul, MN, US: HarvestChoice. http://bit.ly/1hCwlGh. Beddow, J., R. W. Sutherst, D. Kriticos, E. Duveiller, and Y. Chai. Oct. 2013a. Puccinia gram-inis (Stem Rust). Pest Geography. St. Paul, MN, US: HarvestChoice-InSTePP (International Science Technology Practice Policy), University of Minnesota; and Canberra: Commonwealth Scientific and Industrial Research Organisation. Pardey, P. G., J. M. Beddow, D. J. Kriticos, T. M. Hurley, R. F. Park, E. Duveiller, R. W. Sutherst, J. J. Burdon, and D. Hodson. 2013. “Right-Sizing Stem-Rust Research.” Science 340 (6129): 147–148. You, L., Z. Guo, J. Koo, W. Ojo, K. Sebastian, M. T. Tenorio, S. Wood, and U. Wood-Sichra. 2012. “Spatial Production Allocation Model (SPAM) 2000, Version 3, Release 1.” Accessed December 14, 2012. http://MapSPAM.info. BENEFITS OF TRYPANOSOMOSIS CONTROL IN THE HORN OF AFRICA Cecchi, C., W. Wint, A. Shaw, A. Marletta, R. Mattioli, and T. P. Robinson. 2010. “Geographic Distribution and Environmental Characterization of Livestock Production Systems in Eastern Africa.” Agriculture, Ecosystems and Environment 135: 98–110. Shaw, A. P. M., G. Cecchi, G. R. W. Wint, R. C. Mattioli, and T. P. Robinson. 2014. “Mapping the Economic Benefits to Livestock Keepers from Intervening against Bovine Trypanosomosis in Eastern Africa.” Preventative Veterinary Medicine 113 (2): 197–210. http://bit.ly/1c52W7T. Wint, G. R. W. 2001. Kilometre Resolution Tsetse Fly Distribution Maps for the Lake Victoria Basin and West Africa. Report to the Joint Food and Agriculture Organization of the United Nations (FAO)/International Atomic Energy Agency (IAEA) Programme. Vienna, Austria: FAOIAEA Joint Division. 64
- 81. ATLAS OF AFRICANAGRICULTURE RESEARCH DEVELOPMENT Access to Trade Market Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Accessing Local Markets: Marketsheds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 Accessing International Markets: Ports and Portsheds . . . . . . . . . . . . . . . . . 70 Works Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 65
- 82. ACCESS TO TRADE Market Access Zhe Guo and Cindy Cox WHAT ARE THESE MAPS TELLING US? Most Africans do not have easy access to markets. To reach a city of 50,000 people, a farmer in western Africa may only have to travel 1 to 2 hours, whereas farmers in less densely populated areas such as eastern Angola may need to travel 8 hours or more. The maps show travel time to major settle-ments with populations of 20,000 or more (Map 1), 50,000 or more (Map 2), 100,000 or more (Map 3), and 250,000 or more (Map 4). Travel time is a proxy for accessibility and shows how likely farming households are to be physically integrated with or isolated from markets. Travel time is influ-enced not only by distance but also by infrastructure qual-ity and road conditions. For example, because South Africa has better infrastructure and more well-maintained roads than the Democratic Republic of the Congo, it would take a South African farmer less time to travel the same distance to a market than a Congolese farmer. Another factor in deter-mining market accessibility is the density of large cities in a country. A country with many large cities, like Nigeria, has highly accessible markets. WHY IS THIS IMPORTANT? Improved market access for the poorest countries is widely regarded as necessary to support agricultural and rural development. In Africa the practice of trading agricultural products is highly constrained by agricultural policies and poor transportation networks. Challenging road condi-tions, long distances, and inadequate road infrastructure add to travel times and transportation costs and there-fore limit opportunities for farmers to sell their goods. Poor market access can also negatively impact farm production, because the accessibility of critical agricultural inputs such as fertilizer, pesticides, and seed is also limited. Compared to urban households and those with easy access to mar-kets, rural farm households without market access typi-cally rely on their own production for most of their calorie intake. Inadequate market access, therefore, puts these households at greater risk of food insecurity. The more accessible markets are, the greater the population’s ability to remain economically self-sufficient and maintain food security. A comparison of the maps, which express travel time to different-sized cities (market centers), can help stake-holders better understand factors that determine farm per-formance. A simple cost-benefit analysis reveals whether it is more profitable to travel longer distances to larger markets or travel shorter distances to reach the nearest market. WHAT ABOUT THE UNDERLYING DATA? Accessibility was determined using a cost-distance function to measure the “cost” in hours to the nearest market cen-ter for each location, or 1 km2 grid cell. Market centers and their size were determined using population estimates from Global Rural Urban Mapping Project data for the year 2000 (CIESIN et al. 2011). Travel time was estimated based on the combination of global spatial data layers, including road and river networks, assessed in terms of their “friction” or kilo-meters per hour travel time. Travel time was adjusted based on a number of input variables, including road location, road type, elevation, slope, country boundaries, bodies of water, coastline, and land cover. Each input variable was converted to a value representing the time it takes to travel 1 km. In the case of road type, for example, paved roads were given a value of 60 km per hour, while gravel roads were given a value of 15 km per hour. Bodies of water, land cover, slope, country boundaries, and elevation were also used to modify the speed of travel. For example, steeper areas were assigned slower speeds and time delays were factored into travel that crossed borders. The results are not meant to be accurate travel times but to estimate accessibility. WHERE CAN I LEARN MORE? Market access: http://harvestchoice.org/topics/market-access Market access data for SSA: http://harvestchoice.org/products/data/218 Global Rural-Urban Mapping Project population data: http://bit.ly/KbKxJD 66
- 83. Data source: Map1—HarvestChoice 2011a; Map 2—HarvestChoice 2011b; Map 3—HarvestChoice 2011c; Map 4—HarvestChoice 2011d. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT 0 1 2 4 6 8 12 No data Travel time to markets (hours) 0 1 2 4 6 8 12 No data Travel time to markets (hours) 0 1 2 4 6 8 12 No data Travel time to markets (hours) 0 1 2 4 6 8 12 No data Travel time to markets (hours) INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI MAP 1 Population 20,000 ≤ MAP 2 Population 50,000 ≤ MAP 3 Population 100,000 ≤ MAP 4 Population 250,000 ≤ Market access based on population size of market centers 67
- 84. ACCESS TO TRADE Accessing Local Markets: Marketsheds Zhe Guo and Cindy Cox WHAT ARE THESE MAPS TELLING US? Across Africa buying and selling connects people. For a small-scale farmer, this trade takes place primarily within a limited geographic area based on access to market cen-ters of a given size. The maps illustrate these areas using different colors to represent marketsheds—geographi-cal areas and associated populations that are part of real or potential trade networks with a given market. From any location within a marketshed, it takes less time to travel to the corresponding market compared to any neighbor-ing markets. In theory, farmers within a marketshed prefer to trade their commodities at the corresponding market, which minimizes travel cost (p. 66). The maps show that the density of marketsheds in Nigeria is high compared to that of other countries, because the country has many large cities. The high concentration of marketsheds also shows that it takes less time to travel to markets in Nigeria com-pared to neighboring countries. This suggests a denser and perhaps higher-quality infrastructure. The progression of Maps 1–4 shows that as the size of market centers, based on population, increases, there are fewer markets across the continent. Farmers thus have to travel farther, often across country boundaries, to reach larger market centers which may represent more lucrative trade opportunities. WHY IS THIS IMPORTANT? When analyzing factors that influence current and future farm performance, development planners and researchers need to know which markets are closest to agricultural pro-ducers. Farmers customarily select markets close to them so they can get to the market in the least amount of time to trade their goods; buy critical agricultural inputs, such as fer-tilizer, seed, and pesticides; or tap into a range of public and private services (extension, credit, and veterinary services being prime examples). A relatively large marketshed could mean that the population density for that shed is so low that few markets exist, and therefore that farmers have limited opportunities to sell their products (such as in Namibia). Or it might mean that the market within the shed serves a large population most likely due to adequate investments in road infrastructure. The maps show that the marketsheds are not restricted by country borders, which means that a farm-er’s preferred market of a given size may be in a neighboring country. In that case, restrictions posed by border crossings and trade laws need to be considered when determining the optimal market for a farmer. Because each map is based on market centers of different sizes, they can be used to deter-mine the best markets for selling a farmer’s goods. Farmers with an abundance of high-value goods will often prefer to sell or trade at larger commercial markets where demand and prices are higher than at smaller local markets. WHAT ABOUT THE UNDERLYING DATA? Marketsheds are based on the cost of travel to a market center of a given size. The number of marketsheds in a country indicates the number of market centers of that size within the country (for example, Map 1 is based on a market-center population of 50,000 or greater). The popu-lation cutoffs used in the maps are based on population esti-mates from Global Rural-Urban Mapping Project (GRUMP) data for the year 2000 (CIESIN et al. 2011). Proximity to a market was determined by measuring the lowest accumu-lated cost, or travel time, to each market location. Every mar-ket is surrounded by a marketshed. All points within the marketshed area offer the shortest travel time to the corre-sponding market center. Points along the boundary between two sheds have equal travel time to both of the centers. Travel time is estimated based on a combination of spa-tial data layers and variables that affect the time required to travel to the cities or market centers. These variables include elevation, slope, land cover, roads, road types, rivers, borders, and major bodies of water (Guo 2010). WHERE CAN I LEARN MORE? Marketsheds for Africa south of the Sahara (SSA): http://harvestchoice.org/labs/market-sheds Market access: http://harvestchoice.org/topics/market-access Marketshed data for SSA: http://bit.ly/1oFyB1B 68
- 85. Data source (allmaps): HarvestChoice 2012. Note: Population data used are for the year 2000 (CIESIN 2011). The different colors represent marketsheds. A marketshed is the total area surrounding a market center of a given size. From any point within the marketshed, it is quicker to travel to that market center than to any neighboring marketshed’s main market. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI Marketsheds based on population size of market centers MAP 1 Population 50,000 ≤ MAP 2 Population 100,000 ≤ MAP 3 Population 250,000 ≤ MAP 4 Population 500,000 ≤ 69
- 86. ACCESS TO TRADE Accessing International Markets: Ports and Portsheds Zhe Guo WHAT ARE THESE MAPS TELLING US? More than 300 million Africans, about 30 percent of the total population, live more than one day away from the near-est port. Even when ports lie within a few hundred miles, typically sparse road networks, poor maintenance, and lim-ited transportation infrastructure translate into high access costs. The larger map illustrates cost-of-travel accessibility to 63 major African ports, based on port type, size, and capac-ity in terms of the estimated total number of hours, both off and on the road network, required to travel from any loca-tion in Africa to the nearest port. The populations, traders, and haulage operations of countries such as South Africa and Egypt that maintain more and better ports as well as better transportation infrastructure have significantly bet-ter port access than those in landlocked countries such as Chad and South Sudan or large countries such as Democratic Republic of the Congo where infrastructure is limited. The travel time analysis underpinning the map is further sum-marized in Map 2, which shows portsheds. A portshed is a port’s catchment area. Each portshed includes all the loca-tions that are closer to a given port in terms of travel time than to any other port. Ports with large catchment areas, such as Mombasa in Kenya, have few competing ports and are connected to more extensive road networks. Ideally each port should be endowed with transportation corridors, infrastructure, and port facilities that maximize the trading opportunities within its specific portshed. WHY IS THIS IMPORTANT? Seaports play a significant role in enabling both export oppor-tunities for agricultural products and import potential for new technologies and production inputs. Indeed, more than 90 percent of the international trade in African countries is conducted using maritime transport. Most African countries import vital agricultural inputs such as fertilizers, seeds, pesti-cides, and herbicides. Crops (especially cash crops) and live-stock products (including skins and hides and, in the Horn of Africa, live animals) are primary agricultural exports. Map 1, showing travel time, provides a picture of how isolated many Africans are from such import and export hubs that could connect them with world markets and expand their earning potential. This information is valuable to policymakers and investors, both public and private. It allows them to identify intervention priorities that will, assuming sufficient competi-tion in the transportation sector, reduce transaction costs and increase the capacity and efficiency of transportation systems. This ultimately improves production incentives for farmers and raises farm-level productivity and profitability by lower-ing input costs and increasing output prices. WHAT ABOUT THE UNDERLYING DATA? Travel time was estimated using a combination of spatial data layers and variables that influence accessibility, includ-ing elevation, slope, land cover, the road network, road types, rivers, borders, and major bodies of water. Esri’s ArcGIS Spatial Analyst was used to develop spatial indicators of travel time to 63 major ports in Africa, which were selected based on port size and regional distribution (Table 1). The continent was then divided into portsheds, each defining the area associated with the closest corresponding port. The closest port was determined by estimating the lowest accumulated travel time (or cost) from a geographic loca-tion to the port. Using this approach, port A is the closest port for any geographic location within portshed A. WHERE CAN I LEARN MORE? Portshed data: http://bit.ly/1eRgKkI World Port Source: www.worldportsource.com TABLE 1 Distribution of major African ports by region and size REGION PORT SIZE TOTAL Large Medium Small Eastern Africa 1 7 6 14 Middle Africa - 4 4 8 Northern Africa 3 15 4 22 Southern Africa 1 4 1 6 Western Africa - 9 7 16 Total 5 39 22 66 Data source: World Port Source 2012 and FAO 2012. Note: The classification of harbor size is based on several applicable factors, including area, facilities, and wharf space. It is not based on area alone, nor any other single factor (National Geospatial-Intelligence Agency 2012). 70
- 87. Data source: Map1—HarvestChoice 2012 and World Port Source 2012; Map 2—HarvestChoice 2012. Note: Map 2—The different colors represent portsheds based on access to a major port. A portshed is the total area surrounding a major port for which the given port is closer in terms of travel time than any other port. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT 0 4 12 24 48 Major African port Hours INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI Major African port MAP 1 Travel time to major African ports MAP 2 Portsheds for 63 major African ports 71
- 88. ACCESS TO TRADE Works Cited MARKET ACCESS CIESIN (Center for International Earth Science Information Network)/Columbia University, International Food Policy Research Institute, the World Bank, and Centro International de Agricultura Tropical. 2011. Global Rural-Urban Mapping Project, Version 1 (GRUMPv1): Settlement Points. http://bit.ly/1fgDVBC. HarvestChoice. 2011a. “Average Travel Time to Nearest Town Over 20K (hours) (2000).” International Food Policy Research Institute and University of Minnesota. Accessed January 27, 2014. http://harvestchoice.org/node/5210. —. 2011b. “Average Travel Time to Nearest Town Over 50K (hours) (2000).” International Food Policy Research Institute and University of Minnesota. Accessed January 27, 2014. http://harvestchoice.org/node/5211. —. 2011c. “Average Travel Time to Nearest Town Over 100K (hours) (2000).” International Food Policy Research Institute and University of Minnesota. Accessed January 27, 2014. http://harvestchoice.org/node/5213. —. 2011d. “Average Travel Time to Nearest Town over 250K (hours) (2000).” International Food Policy Research Institute and University of Minnesota. Accessed January 27, 2014. http://harvestchoice.org/node/5214. ACCESSING LOCAL MARKETS: MARKETSHEDS HarvestChoice. 2012. Market Sheds. http://harvestchoice.org/labs/market-sheds. CIESIN (Center for International Earth Science Information Network)/Columbia University, International Food Policy Research Institute, the World Bank, and Centro International de Agricultura Tropical. 2011. Global Rural-Urban Mapping Project, Version 1 (GRUMPv1): Settlement Points. http://bit.ly/1fgDVBC. ACCESSING INTERNATIONAL MARKETS: PORTS AND PORTSHEDS FAO (Food and Agriculture Organization of the United Nations). 2012. FAOSTAT database. http://bit.ly/1dUsYWv. HarvestChoice. 2012. “Infrastructure and Transportation.” http://bit.ly/1dhJ7s9. National Geospatial-Intelligence Agency. 2012. World Port Index, 22nd ed. Pub. 150. Springfield, VA, US. World Port Source. 2012. www.worldportsource.com/index.php. 72
- 89. ATLAS OF AFRICANAGRICULTURE RESEARCH DEVELOPMENT Human Welfare Severity of Hunger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Poverty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Early Childhood Nutrition and Health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 Works Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 73
- 90. HUMAN WELFARE Severityof Hunger Klaus von Grebmer, Tolulope Olofinbiyi, Doris Wiesmann, Heidi Fritschel, Sandra Yin, and Yisehac Yohannes WHAT ARE THESE MAPS TELLING US? Map 1 shows the severity of hunger in Africa by catego-ries— ranging from low to extremely alarming. These cate-gories are associated with Global Hunger Index (GHI) scores. Higher scores indicate greater hunger; the lower the score, the better a country’s situation. Of the 19 countries world-wide with alarming or extremely alarming levels of hunger, most (15) are in Africa south of the Sahara. Map 2 shows country progress in reducing GHI scores since 1990—that is, the percentage change in the 2013 GHI compared with the 1990 GHI. An increase in the GHI indicates a country’s hun-ger situation is deteriorating. A decrease in the GHI indicates an improvement. Overall, from the 1990 GHI to the 2013 GHI, six coun-tries in Africa were able to reduce their scores by 50 percent or more. Twenty countries made modest progress, reducing their GHI scores by 25.0 to 49.9 percent, and 17 countries decreased their GHI scores by 0.0 to 24.9 percent. Hunger grew worse in Burundi, Comoros, and Swaziland (Map 2). Increased hunger in Burundi and Comoros can be attributed to prolonged conflict and political instability. For Burundi, the share of undernourished people in the population rose from 49 to 73 percent between the 1990 GHI and 2013 GHI. In Swaziland (Figure 1), the HIV and AIDS epidemic, along with high unemployment and adverse macroeconomic con-ditions, likely undermined food security. Ghana, the top performer in Africa in terms of improved GHI scores since 1990 (Figure 1), is the only country in Africa to appear on the top 10 list worldwide. Significant drops in the share of undernourished population and in the prevalence of under-weight in children under five (p. 78) contributed to Ghana’s 2013 GHI of 8.2, down from the 1990 GHI of 25.5 (Figure 1). WHY IS THIS IMPORTANT? The GHI is designed to comprehensively measure and track hunger globally, by country, and by region. It highlights suc-cesses and failures in reducing hunger and provides insights into its drivers. By highlighting regional and country differ-ences, the GHI aims to trigger actions to reduce hunger. The GHI is a multidimensional index of hunger that combines three equally weighted indicators (undernourishment, child underweight, and child mortality) in one number. This mul-tidimensional approach takes into account the nutrition sit-uation not only of the population as a whole, but also of a physiologically vulnerable group—infants and young chil-dren— for whom a lack of nutrients (p. 78) creates a high risk of illness, poor physical and cognitive development, and death. WHAT ABOUT THE UNDERLYING DATA? The 2013 GHI was calculated for 120 countries globally for which data were available and where measuring hunger is considered most relevant. The GHI is only as current as the data for the three component indicators: undernourishment, child underweight, and child mortality. Source data for the 2013 GHI are from 2008 to 2012 (von Grebmer et al. 2013). Therefore, the GHI is a snapshot of the recent past, not the present. More up-to-date and extensive country data on hunger are urgently needed. The Democratic Republic of the Congo, for example, had the worst score in past GHI reports. But due to political instability and ongoing conflict, reliable data are no longer available to calculate its GHI. WHERE CAN I LEARN MORE? 2013 Global Hunger Index: http://bit.ly/KaKqhr FIGURE 1 Trends in GHI scores for two countries 0 5 10 15 20 25 30 GHI 2013 GHI 2005 GHI 2000 GHI 1995 GHI 1990 Ghana Swaziland Source: von Grebmer et al. 2013. 74
- 91. Data source (allmaps): von Grebmer et al. 2013. Note: The 2013 Global Hunger Index score could only be calculated for former Sudan, because separate undernourishment estimates for 2010–2012 were not available for (north) Sudan or South Sudan, which became independent in 2011. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Algeria Libya Egypt Mauritania Mali Morocco Tunisia Niger Nigeria Chad Ghana Senegal Guinea Sudan Liberia Ethiopia Eritrea Djibouti Somalia Kenya Uganda Tanzania Gabon Cameroon Angola Zambia Zimbabwe Namibia Botswana South Africa Mozambique Madagascar Comoros Democratic Republic of the Congo Seychelles Malawi Sao Tome and Principe Côte Sierra d'Ivoire Leone Guinea- Bissau Western Sahara Central African Republic Burkina Faso (former) Rwanda Burundi Lesotho Swaziland Togo Benin e Gambia Republic of Congo Equatorial Guinea Algeria Libya Egypt Mauritania Mali Morocco Tunisia Niger Nigeria Chad Togo Benin Ghana Senegal Guinea Sudan Liberia Ethiopia Eritrea Somalia Kenya Uganda Tanzania Rwanda Burundi Gabon Cameroon Angola Zambia Zimbabwe Namibia Botswana South Africa Mozambique Madagascar Seychelles Sao Tome and Côte Sierra d'Ivoire Leone Western Sahara Central African Republic Burkina Faso (former) Djibouti Lesotho Swaziland Comoros Principe Democratic Republic of the Congo Guinea- Bissau e Gambia Republic of Congo Equatorial Guinea Malawi Low: 4.9 Moderate: 5.0–9.9 Serious: 10.0–19.9 Alarming: 20.0–29.9 Extremely alarming: 30.0 No data Severity: score Decrease by 50.0% or more Decrease by 25.0–49.9% Decrease by 0.0–24.9% Increase 1990 2013 GHI 5 No data INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI MAP 1 2013 Global Hunger Index scores MAP 2 Percentage change in 2013 GHI compared with 1990 GHI 75
- 92. HUMAN WELFARE Poverty Carlo Azzarri WHAT ARE THESE MAPS TELLING US? Almost half of the population of Africa south of the Sahara (SSA) lives in extreme poverty, on less than $1.25 per cap-ita per day.1 Map 1 shows the distribution of the poor and highlights areas where over 80 percent of the population is extremely poor (for example, parts of Liberia, Nigeria, Tanzania, and Zambia). Map 2 shows the density of extremely poor across the continent, highlighting regions that are home to more than 100 extremely poor people per square kilome-ter. Moderate poverty is defined as living on a daily per cap-ita expenditure between $1.25 and $2.00. Map 3 shows the distribution of poor using the $2.00 per day threshold, thus including both the moderately and extremely poor. This map shows a more even distribution of poor across Africa and consistently higher shares of the total population. Map 4 rein-forces that the most densely populated poor areas are con-centrated along the coast of western Africa, in much of Nigeria, in Malawi, in Ethiopia, and in the countries bordering or near Lake Victoria. Figure 1 shows that extreme poverty is also highly correlated with certain agroecological zones (p. 34). For example, poverty levels are highest in the warm semiarid and subhumid tropical areas immediately south of the Sahara and in the tropical warm humid forests of the Democratic Republic of the Congo. And, overall, poverty lev-els are lower in the subtropical zones of southern Africa (for example, Namibia and South Africa). WHY IS THIS IMPORTANT? Poverty prevalence (Maps 1 and 3) is crucial information for policymakers and international donors who are setting priorities for intervention and investment. Poverty density complements prevalence by showing the number of poor people per square kilometer (Maps 2 and 4). These maps together answer two important questions: Where is poverty a serious problem? Where might investments have the great-est impact on the highest number of people? Combining insights on both prevalence and density allows policymakers to more effectively target interventions to reach the great-est number of the poorest people. Once target populations are identified, information on the dominant types of exist-ing livelihoods and agriculture-related opportunities can be helpful in formulating appropriate interventions. WHAT ABOUT THE UNDERLYING DATA? Subnational poverty rates were extracted from 24 nationally representative household surveys conducted in various years. For countries without survey data, national average poverty prevalence extracted from PovcalNet (World Bank 2012) for the closest year to 2005 was uniformly applied to the entire country. As such, subnational poverty rate distributions reflect the relative ranking in the actual survey year, although all val-ues are expressed in terms of 2005 average purchasing power parity exchange rates. Poverty ratios are therefore compara-ble across countries. Not all current data points refer to 2005, with a maximum variance of plus or minus two years for a limited number of countries (HarvestChoice 2012). WHERE CAN I LEARN MORE? Poverty analysis at the World Bank: www.worldbank.org/en/topic/poverty “Poverty Comparisons over Time and Across Countries in Africa.” Sahn and Stifel 2000. “Where Will the World’s Poor Live?: An Update on Global Poverty and the New Bottom Billion.” Sumner 2012. FIGURE 1 Poverty headcount ratio by agroecological zone Warm arid Cool subhumid Cool semiarid Warm semiarid Warm subhumid Cool arid Warm humid Cool humid Warm semiarid Warm subhumid Warm humid Cool subhumid Cool humid Cool semiarid Warm arid Cool arid Tropic Subtropic 0 20 Poverty headcount ratio (%) 40 60 Data source: HarvestChoice 2012 and HarvestChoice 2011. Note: Poverty headcount ratio=the percentage of a population living in house-holds where consumption or income per person is below the poverty line. 1 The $1.25 and $2.00 poverty lines are the level of total household per capita consumption expenditure (a synthetic indicator of household welfare) expressed in terms of 2005 average purchasing power parity exchange rates. 76
- 93. Data source (allmaps): HarvestChoice 2012. Note: All values are expressed in terms of average 2005 purchasing power parity rates. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT 1 10 20 30 40 50 60 70 80 90 100 No data Outside focus area Prevalence (percent) 1 10 20 30 40 50 60 70 80 90 100 No data Outside focus area Prevalence (percent) 0 5 10 25 50 100 500 No data Number of poor/km² Outside focus area 0 5 10 25 50 100 500 No data Number of poor/km² Outside focus area INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI MAP 1 Share of population living at ≤ $1.25/day (extremely poor) MAP 2 Poverty density at ≤ $1.25/day (extremely poor) MAP 3 Share of population living at ≤ $2.00/day (includes moderately and extremely poor) MAP 4 Poverty density at ≤ $2.00/day (includes moderately and extremely poor) 77
- 94. HUMAN WELFARE EarlyChildhood Nutrition and Health Carlo Azzarri WHAT ARE THESE MAPS TELLING US? High levels of stunting, or lower than average height in children younger than five, are more widespread in Africa south of the Sahara (SSA) than high levels of wasting (low-er- than-average weight for height) or underweight (low weight for age) in children under age five (Maps 1, 2, and 3). This reflects a longstanding nutritional problem that has proven difficult to eradicate in this region. Even with improved living conditions in SSA, the prevalence of stunting has not yet been sufficiently reduced. Stunting and under-weight are manifestations of undernutrition—food energy deprivation that occurs when food intake is below standard nutritional requirements for a prolonged period and/or lev-els of food absorption are low. Wasting usually reflects an acute weight loss due to a recent period of hunger or disease and is often associated with shorter term limitations to food supplies. The maps show that high rates of undernutrition do not always correspond to high rates of diarrhea (Map 4), which contribute to undernutrition by interfering with the absorption of food consumed. This suggests that poor infra-structure and lack of access to clean water (the main causes of diarrhea) are just two of many reasons behind the severe undernutrition in SSA. The red areas of the maps reflect undernutrition levels classified as “very high”—40 percent or above for stunting; 15 percent or above for wasting; 30 per-cent and above for underweight (WHO 2006); and 20 per-cent or above for diarrhea—and highlight the key areas for concern across the continent. WHY IS THIS IMPORTANT? The information on these maps is crucial to policymak-ers and national and international donors who seek to direct resources to the most food-insecure regions of the world. Child nutrition is often used as an indicator of an area’s nutrition security. According to the World Health Organization (WHO), child undernutrition is directly or indirectly responsible for one-third of the deaths among children under age five, and it is also related to other ill-nesses common in children, such as diarrhea and measles. Undernutrition carries long-term consequences for children, impairing their cognitive development and affecting their performance once they are adults. Better nutrition trans-lates into a stronger and healthier population with greater opportunities to break the cycle of poverty and achieve bet-ter quality of life. Improving children’s nutritional status is therefore fundamental to realizing a country’s development potential, especially in nations in SSA where nearly half of the population is less than 15 years old. WHAT ABOUT THE UNDERLYING DATA? Measurements are usually taken from children from birth up to 60 months, as this captures the impact of possible deficiencies incurred during gestation, and it is when chil-dren are most vulnerable as they rapidly grow and develop. After the 1,000-day window of opportunity (from the start of a woman’s pregnancy until her child’s second birthday), any impaired height development or cognitive function is largely irreversible. To obtain anthropometric measures, we used the children’s weight, height, and age information col-lected in the Demographic and Health Survey (DHS) Phase 5 (2003–2008) and Phase 6 (2008–2013). The DHS surveys are regularly conducted in many developing countries in differ-ent years, and these maps show the values for countries with survey years ranging from 2003 to 2011 (Measure DHS 2013). WHERE CAN I LEARN MORE? Measure DHS online: www.statcompiler.com/ WHO Child Growth Standards Publications: www.who.int/childgrowth/publications/en/ Explaining Child Malnutrition in Developing Countries: A Cross-Country Analysis. Smith and Haddad 2000. Poverty and Undernutrition: Theory, Measurements, and Policy. Svedberg 2000. “Worldwide Timing of Growth Faltering: Revisiting Implications for Interventions.” Victora et al. 2010. 78
- 95. Data source (allmaps): Measure DHS 2013 and WHO 2006. Note: The maps are based on DHS surveys conducted over the period 2003 to 2011. The maps show prevalence classes and corresponding undernutrition levels (as a share of total children under age five) as designated by the World Health Organization. ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT Low: 20 Medium: 20–29 High: 30–39 Very high: 40 No data Outside focus area Prevalence: percent Low: 10 Medium: 10–19 High: 20–29 Very high: 30 No data Outside focus area Prevalence: percent Low: 5 Medium: 5–9 High: 10–14 Very high: 15 No data Outside focus area Prevalence: percent Low: 11 Medium: 11–15 High: 15–20 Very high: 20 No data Outside focus area Prevalence: percent INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI MAP 1 Stunting MAP 2 Wasting MAP 3 Underweight MAP 4 Diarrhea Nutrition and health among children under age five 79
- 96. HUMAN WELFARE WorksCited SEVERITY OF HUNGER von Grebmer, K., D. Headey, C. Béné, L. Haddad, T. Olofinbiyi, D. Wiesmann, H. Fritschel, S. Yin, Y. Yohannes, C. Foley, C. von Oppeln, and B. Iseli. 2013. 2013 Global Hunger Index: The Challenge of Hunger: Building Resilience to Achieve Food and Nutrition Security. Bonn, Germany: Welthungerhilfe; Washington, DC: International Food Policy Research Institute; Dublin: Concern Worldwide. POVERTY HarvestChoice. 2011. AEZ (16-class). http://harvestchoice.org/data/aez-16-class. —. 2012. “Sub-national and Extreme Poverty Prevalence.” International Food Policy Research Institute and University of Minnesota. Accessed January 27, 2014. http://harvestchoice.org/node/4751. Sahn, D. E., and D. Stifel. 2000. “Poverty Comparisons Over Time and Across Countries in Africa.” World Development 28 (12): 2123–2155. Sumner, A. 2012. “Where Will the World’s Poor Live? An Update on Global Poverty and the New Bottom Billion.” World Development 40 (5): 865–877. World Bank. 2012. “PovcalNet: An Online Poverty Analysis Tool.” http://iresearch. worldbank.org/PovcalNet/index.htm. EARLY CHILDHOOD NUTRITION AND HEALTH Measure DHS (Demographic and Health Surveys). 2013. STATcompiler. Accessed November 11, 2013. http://www.statcompiler.com/. Smith, L. C., and L. Haddad. 2000. Explaining Child Malnutrition in Developing Countries: A Cross-Country Analysis. Washington, DC: International Food Policy Research Institute. Svedberg, P. 2000. Poverty and Undernutrition: Theory, Measurements, and Policy. Oxford, UK: Oxford University Press. Victora, C. G., M. de Onis, P. C. Hallal, M. Blössner, and R. Shrimpton. 2010. Worldwide Timing of Growth Faltering: Revisiting Implications for Interventions. Pediatrics 125 (3): 473–480. WHO Multicentre Growth Reference Study Group. 2006. WHO Child Growth Standards: Length/Height-for-Age, Weight-for-Age, Weight-for-Length, Weight-for-Height and Body Mass Index-for-Age: Methods and Development. Geneva: World Health Organization. http://bit.ly/1i7JPfX. 80
- 97. About the Authors Christopher Auricht ([email protected]) is managing director of Auricht Projects, a consultancy in Adelaide, South Australia. His background is in applied science in natural resource management. He has extensive international experience in the development and implementation of programs and frameworks that help provide quality data and information for planning for natural resources and sustainable management and policymaking. Carlo Azzarri ([email protected]) is a research fellow on the HarvestChoice research team at the International Food Policy Research Institute (IFPRI), Washington, DC. His research involves micro- and macroeconomic analysis, using quantitative (statistical and econometric) and qualitative methods, of the interrelationships among poverty, nutrition, food security, agriculture, and migration. Before joining IFPRI, Azzarri worked in the research group on poverty and inequality at the World Bank and was a member of the Rural Income Generating Activities team at the Food and Agriculture Organization of the United Nations (FAO). Jason Beddow ([email protected]) is an assistant professor of international agriculture at the University of Minnesota (UMN), Minneapolis–St. Paul. His work focuses on how research investment patterns and aspects of the natural environment (including pests, diseases, climate, and the weather) affect agricultural production and productivity, the spatial dynamics of production, and efforts to bridge the divide between biology and economics. Nienke Beintema ([email protected]) is program head of the Agricultural Science and Technology Indicators (ASTI) initiative at IFPRI, Washington, DC. She has close to 20 years of experience doing agricultural research and development in low- and middle-income countries. Her work has involved collecting and analyzing financial and human resource data and examining institutional developments. Chandrashekhar Biradar ([email protected]) is an agroecosystem ecologist who heads Geoinformatics and is the principal scientist at the International Center for Agricultural Research in the Dry Areas (ICARDA). His current research interests include applying geoinformatics to agroecosystem research and the remote sensing of global food and environmental security in dry areas of the world. Jean-Marc Boffa ([email protected]) is a farming systems project manager and consultant at the World Agroforestry Centre (ICRAF), Nairobi. An experienced systems agronomist with wide experience in western, central, and eastern African farming systems, he has worked for ICRAF in various roles and is a recognized authority on parkland agroforestry systems. Giuliano Cecchi ([email protected]) is a project officer for FAO, Rome. He was trained as an environmental engineer. He specializes in geospatial analysis, data management, and remote sensing. Over the last eight years with FAO, he has focused on the epidemiology and biogeography of tsetse-transmitted African trypanosomoses. Yuan Chai ([email protected]) is a PhD student in the department of applied economics at UMN, Minneapolis–St. Paul. Chai holds an MS in plant pathology and has extensive experience working with rust diseases. His research focuses on the bioeconomics of agricultural pests and diseases. Giuseppina Cinardi ([email protected]) works as a livestock information analyst for the Livestock Information, Sector Analysis and Policy Branch of FAO, Rome. She organizes, updates, and maintains a comprehensive database of livestock information and statistics. About the Authors 81
- 98. Lieven Claessens ([email protected])is a principal scientist specializing in natural resources (soil and water) at the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Nairobi, and an assistant professor in soil geography and landscape at Wageningen University, the Netherlands. His current work focuses on spatial analysis, integrated assessment, and modeling of agricultural systems—with an emphasis on smallholder systems in Africa south of the Sahara and on adaptation and mitigation strategies in the context of climate change. Giulia Conchedda ([email protected]) is a livestock and livelihoods information analyst in the Livestock Information, Sector Analysis and Policy Branch of FAO, Rome. She has many years of experience as a spatial data analyst and is particularly interested in socioecological approaches to studying livestock production systems and associated livelihoods. Cindy Cox ([email protected]) is a technical writer on the HarvestChoice team at IFPRI, Washington, DC. Cox holds an MS and PhD in plant pathology. Before joining IFPRI, she worked as a Peace Corps volunteer in the Central African Republic and as an agricultural scientist at The Land Institute. John Dixon ([email protected]) is a principal advisor in the research division of the Australian Centre for International Agricultural Research, Canberra, Australia. He has over 40 years’ experience working for the CGIAR system and FAO on agricultural research and development, including cropping systems, economics, and natural resource management in South, Southeast, and East Asia; Africa; Latin America; and the Middle East. He has served as director of Impacts, Targeting and Assessment at the International Maize and Wheat Improvement Center (CIMMYT), leading activities on impact assessment, value chains, impact knowledge sharing, systems agronomy, and conservation agriculture. He also served in various capacities with FAO in its global, regional, and country programs. Petra Döll ([email protected]) is a professor of hydrology at the Goethe University Frankfurt in Frankfurt am Main, Germany. Her research focuses on the global freshwater system and on interdisciplinary research in natural resources management. Döll is co-developer of WaterGAP, a model of global water resources and their use that has been used to assess the anthropogenic impact on the global freshwater system, including the impact of climate change, reservoirs, and water withdrawals from groundwater and surface water. Kathleen Flaherty ([email protected]) is a senior research analyst with the ASTI initiative at IFPRI, Washington, DC. Her work focuses on collecting and analyzing human and financial resource data related to agricultural research and development in low- and middle-income countries. Karen Frenken ([email protected]) is the senior officer for Water Resource Man agement at the Land and Water Division of FAO, Rome. She worked for almost 20 years as an agricultural engineer in different countries in South Asia, the Near East, and Africa south of the Sahara, mainly on irrigation and water management for agricultural purposes. A particular focus of her work was fragile ecosystems with special attention to women and irrigated agriculture. In 2003, she joined FAO headquarters, where she is responsible for managing the AQUASTAT Programme, FAO’s global water information system. Heidi Fritschel ([email protected]) is an editor in the Communications and Knowledge Management Division of IFPRI, Washington, DC. 82 Atlas of African Agriculture Research Development
- 99. Dennis Garrity ([email protected])is a senior fellow at the World Agroforestry Centre (ICRAF), Nairobi. He is a systems agronomist and research leader whose career has focused on the development of small-scale farming systems in the tropics. He is currently serving as drylands ambassador for the United Nations Convention to Combat Desertification, raising awareness about the importance of combating desertification and land degradation and mitigating the effect of drought. He is also involved in a global effort to reconsider the future of agriculture in the 21st century by examining unconventional ways of creating more productive and environmentally sound farming. Marius Gilbert ([email protected]) is a fellow of the Fonds National de la Recherche Scientifique, which is based in the Department of Biological Control and Spatial Ecology at the Université Libre de Bruxelles in Brussels, Belgium. He has many years of experience in spatial ecology and epidemiology, and his recent research has focused on modeling of livestock distributions, production systems, and both the distributions’ and systems’ associations with disease risk. Zhe Guo ([email protected]) is a geographic information system coordinator in the Environ ment and Production Technology Division of IFPRI, Washington, DC, and is a member of the HarvestChoice team. His research interests include using spatial data and remotely sensed data for spatial modeling, spatial statistics, and evaluating land cover and land use changes. Jawoo Koo ([email protected]) is a research fellow in the Environment and Production Technology Division of IFPRI, Washington, DC, and is on the HarvestChoice team. He specializes in crop modeling. His current work explores strategic questions about agricultural technologies in Africa, including which areas to target with the technologies and their potential impacts. Raffaele Mattioli ([email protected]) works as a senior officer in the Animal Health Division of FAO, Rome. He is a veterinarian with more than 25 years of experience in tropical animal health and production, with a focus on tsetse and trypanosomosis interventions and other programs that control vectors and vector-borne diseases. He now leads the disease ecology group. Tolulope Olofinbiyi ([email protected]) is a program manager in the Director General’s Office of IFPRI, Washington, DC. She is also a PhD candidate studying development economics and political economy at the Fletcher School at Tufts University. She holds a master of arts in law and diplomacy in international affairs (development economics) and a master of agribusiness and has extensive experience working in the agribusiness sector in Nigeria. Her research focuses on the political economy of public finance in the context of fiscal federalism and decentralization. Felix T. Portmann ([email protected]) is a postdoctoral research associate at the Biodiversity and Climate Research Centre linked to the Senckenberg Research Institute and Natural History Museum and Goethe University Frankfurt in Frankfurt am Main, Germany. Before obtaining his PhD from Goethe University, he was a scientist at the German Federal Institute of Hydrology. His research interests include environmental modeling of the hydrological cycle, coupled climate and ocean modeling, statistical analysis related to past and present climates, integrated water resources management, and remote sensing. Navin Ramankutty ([email protected]) is an associate professor of geography at McGill University, Montreal. His research program addresses issues at the intersection of global food security, land use change, and global environmental change. About the Authors 83
- 100. Timothy Robinson ([email protected])works as a senior spatial analyst at the International Livestock Research Institute (ILRI), Nairobi. His interests include spatial analytical techniques and their application to understanding current and future distributions of livestock and farming systems. He is particularly interested in exploring the social, environmental, and epidemiological risks and opportunities associated with an evolving livestock sector. Kate Sebastian ([email protected]) is a consultant with the Bill Melinda Gates Foundation, Seattle, and the project manager of the eAtlas initiative. She has worked in the field of geographic information systems and agriculture research for a number of organizations including IFPRI, the US Agency for International Development, the World Bank, the HarvestChoice team, and the CGIAR Consortium. Her focus is on mapping and spatial analyses of data related to agricultural land use, poverty, and food security. Alexandra Shaw ([email protected]) is an independent consultant, currently associated with the Division of Pathway Medicine and Centre for Infectious Diseases at the University of Edinburgh, UK. She is an economist with many years of experience in the health economics of the livestock sector, particularly in the study of tsetse-transmitted trypanosomosis and, more recently, of other neglected zoonotic diseases. Stefan Siebert ([email protected]) is a senior scientist at the Institute of Crop Science and Resource Conservation of the University of Bonn, Germany. His research focuses on the analysis of large-scale datasets and the development of large-scale model components to assess the interaction among resource use, crop management, and crop productivity. This model development includes the generation and improvement of datasets required as model inputs. Siebert’s research interests also include general crop modeling (phenology, drought and heat stress, and crop yields), modeling of virtual water flows, and the alteration of ecosystems by anthropogenic impacts. Gert-Jan Stads ([email protected]) is senior program manager of the ASTI initiative at IFPRI, Washington, DC. His focus is on conducting ongoing analysis of agricultural research and development datasets, communicating the results of this analysis to promote advocacy and support policymaking, and building in-country capacity for data collection and analysis. Philip Thornton ([email protected]) leads the “Integration for Decision Making” theme for the CGIAR Research Program on Climate Change, Agriculture and Food Security, which is hosted by ILRI in Nairobi. Thornton is based in Edinburgh, UK. He works mostly in integrated modeling at different scales, looking at the impacts of climate change on smallholder farming systems in the tropics and subtropics. Antonio Trabucco ([email protected]) is a researcher at the Euro-Mediterranean Center on Climate Change, Lecce, Italy. He is a landscape ecologist with a background in geography and a scientific interest in the interactions among climate, society, agriculture, and natural ecosystems. Trabucco is currently investigating the impact of climate and climate change on agricultural production and ecosystem services. Justin Van Wart ([email protected]) is a postdoctoral research associate in the Department of Agronomy and Horticulture of the University of Nebraska, Lincoln. His work focuses on understanding spatial and temporal patterns of agricultural management and crop yields, estimating crop yield potential on the regional and national scales, and deriving global climate data for use in agroclimatology research as well as for field crop simulation models. 84 Atlas of African Agriculture Research Development
- 101. Klaus von Grebmer([email protected]) is a research fellow and strategic advisor in the Director General’s Office of IFPRI, Washington, DC. He served as director of IFPRI’s Communications Division for over 12 years. His work focuses on communicating complex issues to selected stakeholder groups and on communication strategies, science communications, and issues management. Doris Wiesmann ([email protected]) is an independent food security and nutrition expert based in Halendorf, Germany. She is a nutritionist with extensive experience in measuring food security, designing and reviewing international indexes, and developing proxy indicators of household food security and micronutrient adequacy. Wiesmann designed the Global Hunger Index. More recently, her work has focused on dietary assessment methods and strategies to improve micronutrient adequacy in developing countries. William Wint ([email protected]) is an independent consultant and director of the Environmental Research Group Oxford, which is based at the Department of Zoology, University of Oxford, UK. He has worked extensively in the spatial analysis of many aspects of the livestock sector, in particular livestock distribution mapping and the modeling of disease and vector distributions. Stanley Wood ([email protected]) is a senior program officer in the Agricultural Development Program of the Bill Melinda Gates Foundation, Seattle, Wash ington. Earlier, he was a senior research fellow in the Environment and Production Tech nology Division of IFPRI, where he led IFPRI’s research on spatial analysis in a policy context and served as coordinator of CGIAR’s Consortium for Spatial Information. Wood was also one of the principal investigators for the HarvestChoice team. Ulrike Wood-Sichra ([email protected]) has been a research analyst on the HarvestChoice team at IFPRI, Washington, DC, since the team’s inception. She deals with crop-related data and models and the assembly of the HarvestChoice data matrix. Before that, Wood-Sichra was a consultant on other projects at IFPRI and developed databases and software. She also worked on agriculture and natural resources projects for earlier employers, such as FAO, the Asian Development Bank, and the Inter-American Development Bank, sometimes living in the countries the projects related to. Sandra Yin ([email protected]) is an editor in the Communications and Knowledge Management Division of IFPRI, Washington, DC. Yisehac Yohannes ([email protected]) is a research analyst with the Poverty, Health, and Nutrition Division of IFPRI, Washington, DC. He holds a master of science in food and resource economics and a master of science in statistics. He has extensive experience in analyzing multifaceted large-scale primary household surveys that examine topics such as poverty and hunger alleviation, social protection, income, food consumption, and nutrition. His current research involves analyzing agricultural growth and nutrition linkages and the impacts of safety net programs. Robert Zomer ([email protected]) is a landscape ecologist with a broad background in plant community and forest and agricultural ecology and advanced skills in statistical analysis, geographic information systems, remote sensing, environmental modeling, and landscape-level spatial analysis. He is currently a visiting professor at the Kunming Institute of Botany, China, and a senior landscape ecologist at the World Agroforestry Centre (ICRAF)-China. About the Authors 85
- 103. Glossary agricultural systems:crop management schemes selected by farmers to optimize the yield of a particular crop given sociological, economic, biological, and political constraints. agroecological zone: geographical areas that exhibit similar climatic conditions that determine their ability to support rainfed agriculture. These zones broadly define environments where specific agricultural systems thrive. agropastoral farming systems: farming systems located in semiarid areas of western, eastern, and southern Africa, dominated by sorghum, millet, and livestock. Livelihoods are derived from maize, pearl millet, pulses, sesame, sorghum, cattle, goats, poultry, sheep, and off-farm activities. aluminum toxicity: occurs in weathered soils that have become highly acidic, making aluminum soluble and thus toxic to plants. Aluminum toxicity is the most common soil constraint in Africa south of the Sahara (SSA). arable land: the land under temporary agricultural crops (multiple-cropped areas are counted only once), temporary meadows for mowing or pasture, land under market and kitchen gardens, and land temporarily fallow (less than five years). This category does not include abandoned land resulting from shifting cultivation. arid: an area where the length of growing period (LGP) is less than 70 days per year. arid pastoral oasis farming systems: farming systems in scattered communities in arid areas with average length of growing period less than 30 days, and located primarily in northwest, northeast, and southern Africa. Livelihoods are based on cattle, small ruminants, date palms, and off-farm activities. aridity index: the ratio of annual total precipitation to annual total potential evapotranspiration (PET). Aridity index values increase with more humid conditions and decrease with more arid conditions. The aridity index measures how much rainfall is available to satisfy the evapotranspiration water requirements for different reference vegetation types. blue water: water withdrawn from groundwater bodies (aquifers) or surface water bodies (rivers, lakes, wetlands, canals) and used for irrigation of agricultural land, for drinking water, or by the industrial sector for processing and cooling. calcareous: a kind of soil that contains high levels of calcium carbonate. Calcareous soils can be highly fertile, but extremely calcareous soils can lead to crop nutrient deficiencies by fixing phosphorus (see P fixation). cereal-root crop mixed farming systems: farming systems located in subhumid areas of western and central Africa, distinguished by cereal crops along with roots and tubers. Livelihoods are based on cassava, cattle, legumes, maize, millet, sorghum, yams, and off-farm activities. coefficient of variation: a measure of variability from an average calculated as the standard deviation divided by the mean and expressed as a percentage, such as year- to-year rainfall variability. Comprehensive Africa Agriculture Development Program (CAADP): an Africa-led program designed to promote increasing investments in agricultural growth in Africa through research, extension, education, and training. CAADP is a program of the New Partnership for Africa’s Development (NEPAD). consumptive water use: in agriculture, typically refers to crop evapotranspiration only and excludes return flows. cracking clay: soils with high amounts of clay that shrink and swell upon wetting and drying—also called expansive clay. These soils can be difficult to manage because they can be too wet (reducing gas exchange in the soil) for good plant growth. When wet, cracking clay can greatly expand in volume and create additional soil problems. Extensive soil cracking can disturb plant roots, and crusting can reduce water infiltration, when dry. crop evapotranspiration: the sum of evaporation from the soil and transpiration of the plants. dryland systems (also known as dryland agricultural production systems): agroecosystems characterized by low and erratic precipitation, persistent water scarcity, extreme climatic variability, high susceptibility to land degradation— including desertification—and higher loss rates for natural resources, including biodiversity. In dryland systems, the lack of water is the key factor that limits profitable agricultural production. Ea/Et: ratio of actual to potential evapotranspiration. Glossary 87
- 104. evapotranspiration: the conversionof soil water into water vapor. Estimating evapotranspiration rates is important when planning irrigation schemes. farming system: population of farm households that have broadly similar resource and livelihood patterns, face similar constraints and opportunities, and could benefit from similar development strategies and interventions. Household livelihoods are based on farm production as well as off- farm activities. fish-based farming systems: farming systems that are close to major inland or coastal water bodies with fish as a major source of livelihoods. Although located throughout Africa, these fish-based farming systems are concentrated along the coast and around major lakes. Livelihoods are based on fish, bananas, cashews, coconuts, fruit, yams, poultry, goats, and off-farm activities. forest-based farming systems: farming systems in humid lowland, heavily forested areas of central Africa. Livelihoods are based on subsistence food crops, including beans, cassava, cocoyams, maize, taro, and off-farm activities. free of constraints: soils free from fertility constraints. Global Hunger Index (GHI): a multidimensional measure of hunger that combines three equally weighted indicators (undernourishment, child underweight, and child mortality) in one index number. It takes into account the nutrition situation of not only the population as a whole, but also of a physiologically vulnerable group—children—for whom a lack of nutrients creates a high risk of illness, poor physical and cognitive development, and/or death. Global Yield Gap Atlas Extrapolation Domain (GYGA-ED): a climate zone scheme or domain based on three variables: (1) growing degree days with base temperature of 0°C; (2) temperature seasonality (quantified as the standard deviation of monthly average temperatures); and (3) an aridity index (annual total precipitation divided by annual total potential evapotranspiration). (See aridity index, potential evapotranspiration, and transpiration). green water: precipitation stored in the soil and used by rainfed and irrigated crops. growing degree days (GDD): a measure of heat accumulation used to estimate plant development rates. GDD are calculated as the difference between current temperatures and a minimum base threshold temperature (where growth rate=0). Plant growth rates can be measured through the accumulation of GDD, with different species requiring different numbers of accumulated GDD to reach maturity. highland mixed farming systems: farming systems in cool highland areas (above 1,600 meters), dominated by temperate cereals and livestock, located in eastern and southern Africa. Livelihoods are based on broadbeans, goats, lentils, peas, potatoes, rape, teff, wheat barley, poultry, sheep, and off-farm activities. highland perennial farming systems: farming systems in moist highland areas (above 1,400 meters) of eastern Africa, with relatively good market access and with a dominant perennial crop, either food or commercial. Livelihoods are based on diverse activities, including bananas, beans, cassava, coffee, enset (or false banana, Enset ventricosum, in Ethiopia), maize, sweet potatoes, tea, livestock (including dairy), and off-farm activities. humid: an area where the length of growing period (LGP) is greater than 270 days per year. humid lowland tree crop farming systems: farming systems located in western and central Africa that appear in humid lowland areas where commercial tree crops have replaced forest and provide more than one-quarter of household cash income. Livelihoods are based on coffee, cocoa, oil palm, and rubber, as well as cassava, maize, yams, and off-farm activities. insurance crops: crops that increase food security because they can be left in the ground until needed. Roots and tubers, including cassava, fall into this category. intensity ratio of investment: public RD investment measured as a share of agricultural output. irrigated farming systems: large-scale contiguous irrigation schemes, with almost no rain-fed agriculture. Located mostly in areas with low rainfall. Livelihoods are largely based on irrigated commercial crops, notably rice, cotton, and vegetables, as well as cattle and small ruminants. land cover: the physical material at the surface of the earth, such as crops, pasture, trees, bare rock, water, and urban areas. leaching: occurs when water percolating through the soil moves soluble nutrients below the crop root zones. Over time leaching can reduce the availability of nutrients to crops. Old, highly weathered soils in areas of moderate to high precipitation are typically nutrient depleted and acidic as a result of nutrient leaching. 88 Atlas of African Agriculture Research Development
- 105. length of growingperiod (LGP): generally calculated as the period (in days) during a year when precipitation exceeds half the potential evapotranspiration, while also taking into account soil moisture holding capacity. It is used to determine the number of days per year that are suitable for crop growth in a given location. livestock system: a farming system where more than 90 percent of dry matter fed to animals comes from rangelands, pastures, annual forages, and purchased feed and less than 10 percent of the total value of production comes from nonlivestock farming activities. low nutrient reserves: soils with less than 10 percent reserves of weatherable minerals that naturally supply phosphorus, potassium, calcium, magnesium and micronutrients. major climate divisions: major latitudinal thermal or temperature shifts in climate zones. maize mixed farming systems: farming systems located in subhumid and humid areas of eastern, middle, and southern Africa, dominated by maize with legumes. Livelihoods are based mainly on maize, tobacco, cotton, legumes, cassava, cattle, goats, poultry, and off-farm activities. marketshed: geographical area and associated population that has real or potential trade relationships with a market center. Each market shed is associated with the closest corresponding market in terms of the least-cost travel time to that market. MarkSim: a statistical weather generator that produces weather records (rainfall, maximum and minimum air temperature, and solar radiation) on a daily basis. It is able to simulate the variation in rainfall observed in both tropical and temperate regions. mixed crop-livestock farming systems: a farming system in which more than 10 percent of the dry matter fed to animals comes from crop by-products (for example, stubble) or more than 10 percent of the total value of production comes from nonlivestock farming activities. Livestock convert organic material not fit for human consumption into high-value food products (meat, milk) and nonfood products (traction, manure, leather, bone). net primary production (NPP): the amount of biomass produced by a plant or ecosystem, excluding the energy it uses for the process of respiration. This typically corresponds to the rate of photosynthesis, minus respiration by the photosynthesizers. New Partnership for Africa’s Development (NEPAD): a vision and a policy framework of the African Union for pan-African socioeconomic development in the 21st century. North Africa dryland mixed farming systems: farming systems in dry semi-arid areas with rainfall of 150–300 mm, based on rainfed barley and wheat grown in a rotation with one- or two-year fallows and a strong small ruminant component. Livelihoods also include off-farm activities. North Africa highland mixed farming systems: farming systems dominated by rainfed cereal and legume cropping with tree crops, fruits, and olives on terraces, together with vines and/or raising livestock (mostly sheep) on communally managed lands and characterized by moderately high population densities. Livelihoods also include off-farm activities. North Africa rainfed mixed farming systems: farming systems in subhumid areas characterized by tree crops (olive and fruit), melons, grapes, irrigated vegetables, and flowers as well as rainfed wheat, barley, chickpea, lentil, and fodder crops. Livelihoods are supplemented by dry-season grazing of sheep migrating from the steppe areas and off-farm activities. P fixation: occurs when phosphorus (P) becomes insoluble and therefore is not available to plants. Extremely calcareous soils, which contain high levels of calcium carbonate, and soils that are rich in iron and aluminum oxides fix phosphorus and can lead to nutrient deficiencies in a crop. pastoral farming systems: farming systems with low population density in arid areas of western, eastern, and southern Africa, dominated by livestock. Livelihoods are based on camels, cattle, goats, sheep, some cereal crops, and off-farm activities. perennial mixed farming systems: commercially oriented farming systems predominantly found in South Africa and comprising deciduous fruits and vineyards in the Western Cape and eucalyptus, pines, and wattle as well as sugarcane in the southeastern region (KwaZulu-Natal, Mpumalanga and Eastern Cape Provinces) interspersed with cereals, oilseeds, and pulses. Livelihoods include off-farm activities. permanent crops: crops—such as cocoa, coffee, and rubber— that are sown or planted once and then occupy the land for several years and do not need to be replanted after each annual harvest. This category includes flowering shrubs, fruit trees, nut trees, and vines, but excludes trees grown for wood or timber. Glossary 89
- 106. permanent meadows andpastures: land used five years or more to grow herbaceous forage crops, either cultivated or growing wild (wild prairie or grazing land). poor drainage: soils characterized by the inability to properly drain. portshed: an area associated with the closest corresponding port in terms of the least-cost travel time to that port. potential evapotranspiration (PET): the energy available in the system to remove water through the processes of evaporation and transpiration. It is generally associated with a reference crop, namely short grass completely covering the ground, and assumes no limitation on water availability. rain-use efficiency (RUE): the amount of biomass produced (kilograms of dry matter per hectare) per millimeter of rainfall calculated as the ratio of net primary production (NPP) over rainfall. root and tuber crop farming systems: farming systems located in lowland areas of western and middle Africa where systems are dominated by roots and tubers without a major tree crop. Livelihoods are based mainly on cassava, legumes, yams, and off-farm activities. seasonality: the way in which climate (such as rainfall or temperature) varies regularly through the year in a particular place. semiarid: an area where the length of growing period (LGP) is 70–180 days per year. Spatial Production Allocation Model (SPAM): a model that produces estimates of crop distribution and can be used to generate maps showing area harvested per cell by crop and production system (technology). The model draws on many datasets, including land cover imagery, crop suitability maps, irrigation maps, subnational crop statistics, FAO country totals of crop production and area, and data on production systems in each country. stem rust: a fungal disease that affects wheat. stunting: low height for age in children (under age five). Stunting reflects a sustained past episode or episodes of chronic undernutrition. subhumid: an area where the length of growing period (LGP) is 180–270 days per year. subtropics: areas where mean monthly temperature adjusted to sea-level is less than 18° C for one or more months in a year. transpiration: the evaporation of water from the leaves and stems of plants. tropics: areas where the monthly temperature adjusted to sea-level is greater than 18° C for all months. trypanosomosis: a parasitic disease transmitted by the tsetse fly. The African animal form of the disease reduces the productivity of livestock, especially cattle, when it sickens or kills them. Ug99: the collective name for new strains of stem rust pathogen, first discovered in Uganda in 1998. Most of the world’s wheat varieties offer little resistance to Ug99 (see stem rust). undernutrition: a measure of food energy deprivation. Undernutrition results when prolonged food energy intake is below standard nutritional requirements and/or low levels of absorption of food consumed. underweight: low weight for age in children (under age five). Underweight reflects a current condition resulting from inadequate food intake, past episodes of undernutrition, and/or poor health conditions. virtual water: the water needed to produce a product. If a country exports such a product, it exports water in virtual form. virtual water content: the volume of water used by a crop per unit of crop harvest. volcanic: amorphous soils characterized by large reserves of weatherable minerals (which are unstable in humid climates) and soil organic matter making them highly fertile. Volcanic soils also have a high phosphorus fixation capacity which can slightly limit their fertility. wasting: low weight for height in children under age five. Wasting generally reflects an acute weight loss associated with a recent period of hunger or disease. 90 Atlas of African Agriculture Research Development
- 108. 2033 K Street,NW, Washington, DC 20006-1002 USA Tel.: +1.202.862.5600 • SKYPE: ifprihomeoffice Fax: +1.202.467.4439 • Email: [email protected] www.ifpri.org CGIARCSI Consortium for Spatial Information INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE IFPRI ATLAS OF AFRICAN AGRICULTURE RESEARCH DEVELOPMENT The work of agricultural researchers and development workers in Africa has the potential to significantly improve the lives of the poor. But that potential can only be realized with easy access to high-quality data and information. The Atlas of African Agriculture Research Development highlights the ubiquitous role of smallholder agriculture in Africa; the many factors shaping the location, nature, and performance of agricultural enterprises; and the strong interdependencies among farming, natural-resource stocks and flows, and the well-being of the poor. Organized around 7 themes, the atlas covers more than 30 topics, each providing mapped geospatial data and supporting text that answers four fundamental questions: What is this map telling us? Why is this important? What about the underlying data? Where can I learn more? The atlas is part of a wide-ranging eAtlas initiative that will showcase, through print and online resources, a variety of spatial data and tools generated and maintained by a community of research scientists, development analysts, and practitioners working in and for Africa. The initiative will serve as a guide, with references and links to online resources to introduce readers to a wealth of data that can inform efforts to improve the livelihoods of Africa’s rural poor. To learn more about the eAtlas initiative, visit http://agatlas.org. R180 G46 B52 R154 G58 B32 R255 G204 B104 R52 G51 B22 R227 G137 B27 R198 G113 B41