AFRICA REGION Assessment of Food Security Early Warning Systems for EastŸand Southern Africa Africa Climate Business Plan Series January 2018 Assessment of Food Security Early Warning Systems for East and Southern Africa AFRICA REGION Assessment of Food Security Early Warning Systems for East and Southern Africa Ademola Braimoh, Bernard Manyena, Grace Obuya, and Francis Muraya © 2018 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http://creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: Braimoh, Ademola; Manyena, Bernard; Obuya, Grace; and Muraya, Francis. 2018. Assessment of Food Security Early Warning Systems for East and Southern Africa. Africa Climate Business Plan series. World Bank, Washington, DC. License: Creative Commons Attribution CC BY 3.0 IGO. Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Cover photo: Woman walking. Photo: © Curt Carnemark/World Bank (Top left); Terraced fields: © A’Melody Lee/World Bank (Top right); A Kenyan farmer uses a mobile phone in the field. Photo: Neil Palmer, CIAT (Bottom left); Vegetables store: © Ademola Braimoh/World Bank (Bottom right). Further permission required for reuse. (This will depend on the final arrangement of the photos) Cover design: Progressive Publishing Services Contents Preface ix Acknowledgments xi Acronyms and Abbreviations xiii Executive Summary xvii Institutional Challenges xviii Technical Challenges xviii Sustainability and Financial Challenges xix Recommendations xix Chapter 1: The Imperative of Early Warning Systems in East and Southern African Regions 1 Background 1 Four Elements of an Effective People-Centered Early Warning 9 Purpose and Scope of the Report 11 Chapter 2: Early Warning Methods, Technical Skills, and Capacity 15 Introduction 15 Seasonal Climate and Weather Forecasting Methods 17 Crop Forecasting and Monitoring Methods 24 Vulnerability and Capacity Assessments 27 Methods for Grain, Market, Cross-Border, Price, and Commodity Monitoring 31 Chapter 3: Performance of EWSs in East and Southern Africa 41 Introduction 41 Performance of EWSs at the National Level 42 Performance of Food Security EWS at the Regional Level 64 Chapter 4: Strategies for Long-Term Sustainability of Investments on Early Warning Systems 75 Introduction 75 The Costs and Benefits of Investing in an EWS 75 Cost-Effective Strategies for Improving Agrometeorological Observation Systems 84 Strengthening Food Security Information Systems at the Regional Level 87 Operational Models and PPPs 88 Chapter 5: Best Practices and Recommendations for Institutional Strengthening 93 Introduction 93 Innovations for Improving Food Security EW: Role of the Private Sector 93 ASEAN Food Security Information System 96 Participatory Scenario Planning for Coproducing User-Based Climate Services 100 Contents v Quality Assurance Measures and Service Improvements through Producer-User Interface Forums 100 Summary of Findings and Recommendations 103 Recommendations 105 References 111 Appendix A—Early Warning Systems and Their Attributes 117 Boxes Box 1.1 Projected Climate Change Impacts in Sub-Saharan Africa 4 Box 2.1 Technical Insight: Functions of the GDPFS 18 Box 2.2 ICPAC Recommends Early Action Following GHACOF’s Rainfall Outlook for March–May 2017 20 Box 2.3 Technical Insights of GIEWS 25 Box 2.4 FAW Monitoring, Impact Assessment, and the EWS 26 Box 2.5 VAA Challenges 30 Box 3.1 “We’ve No Options but to Rely on Indigenous Knowledge” 48 Box 3.2 Seasonal Forecasts Not Packaged according to Language of Users 48 Box 3.3 Risk Assessments Led by International Agencies 56 Box 4.1 Benefits of Early Action 76 Box 4.2 Operating Models of Climate Information Services Providers 90 Box 5.1 Good Practice from EARS 94 Box 5.2 Application of Climate Information through PSP 101 Figures Figure 1.1 Number of Reported Disasters and People Affected by Disaster Type in Sub-Saharan Africa 2 Figure 1.2 Summary of Climate Impacts and Risks in Sub-Saharan Africa 3 Figure 1.3 Rainfall Anomalies in ESA 4 Figure 1.4 Food Security IPC Crisis Phase 3 and Above (January 2017) 5 Figure 1.5 Known and Suspected Distribution of FAW in Africa (April 2017) 7 Figure 1.6 Elements of an EWS 10 Figure 2.1 Characteristic Spatial Scales of Weather Phenomena 17 Figure 2.2 Good Practice from the SADC RVAA System 29 Figure 2.3 IPC Predicts Famine in Somalia (IPC Phase 5) 32 Figure 2.4 Cereal Deficits in the SADC, 2011–16 35 Figure 3.1 Organizational Structure of the EWS in Zambia 53 Figure 3.2 Uganda National Integrated Early Warning System Information Flow 55 Figure 5.1 Contents of AFSIS Database and Benefits to Users 99 Figure B5.2.1 Five Major Steps of the PSP Process 101 vi Contents Tables Table 1.1 Estimated Economic Impact of the FAW 7 Table 1.2 Participating Countries in the Assessment 12 Table 2.1 Common Methods for Generating Risk Information for an EWS 16 Table 2.2 Hydrometeorological Observation Network Density 22 Table 2.3 WMO-Recommended Minimum Densities of Stations (Area in km2 per Station) 23 Table 2.4 Advantages and Disadvantages of AWSs 23 Table 2.5 Actors in ESA Food Security EWS 28 Table 2.6 NVAA Reports Consolidated into RVAA (2012–16) 29 Table 2.7 Progress in Adopting the IPC Tool in East Africa 30 Table 2.8 IPC Phase Descriptions 31 Table 2.9 The SADC Regional Cereal Production (Tonnes) 2010–16 34 Table 2.10 Good Practice on Processes Leading to the NFBS Production in Zambia 35 Table 2.11 Market Performance for February 2017 for Taita Taveta County, Kenya 37 Table 2.12 Role of the SGR in Grain and Price Monitoring 38 Table 3.1 Effectiveness of Risk Assessments at the National Level 42 Table 3.2 Effectiveness of Monitoring and Warning Services at the National Level 44 Table 3.3 Dissemination and Communication of EWS Information at the National Level 47 Table 3.4 Applying EWS Information in Response at the National Level 49 Table 3.5 Performance of EWS Governance and Investment at the National Level 52 Table 3.6 Effectiveness of Risk Assessments at the Regional Level 65 Table 3.7 Effectiveness of Monitoring and Warning Service at the Regional Level 67 Table 3.8 Effectiveness of EWS Dissemination and Communication at the Regional Level 69 Table 3.9 Effectiveness of Response to Regional Warnings at the Regional Level 70 Table 3.10 Institutional Arrangements and Investment at the Regional Level 72 Table 3.11 ESA RECs and Their Member States Overlaps 73 Table 3.12 Regional EWS Institutional Summary 74 Table B4.1.1 Cost Estimates for Drought Responses in Horn of Africa, Discounted over 20 Years 76 Table 4.1 Examples of Triple Bottom Benefits of Investing in Hydrometeorological Services 77 Table 4.2 Illustrative Economic Assessments of Meteorological/ Hydrometeorological Services 78 Table 4.3 Benefits of Investing in Climate Information and EWSs 79 Contents vii Table 4.4 Financing for Strengthening Hydrometeorological and Climate Services in the Democratic Republic of Congo 80 Table 4.5 Estimated Costs for National Action Plan for Improvement of Hydrometeorological Services (NAPIHMS) Project in Madagascar (US$) 81 Table 4.6 Estimated Costs of the Food Security EWS in Zambia 82 Table 4.7 Options for Investing in Observation Networks 83 Table 4.8 Estimated Annual Costs of an Advanced Network 86 Table 4.9 IGAD RIIS: Estimated Budget for the Formulation Process of the RIIS 88 Table 4.10 Roles of Regional and National Funding Mechanisms 88 Table B4.2.1 Operating Models 90 Table 4.11 Potential for NMHSs to Develop PPPs in Weather Products in Some ESA Countries 91 viii Contents Preface Disasters caused by climate extremes such as tropical cyclones and severe storms, floods, heat waves, and droughts are jeopardizing Africa’s hard-won development achievements toward further growth, food security, and pov- erty reduction. In 2016 the food security situation deteriorated sharply in Africa—especially in East and Southern Africa—as a result of droughts and floods linked in part to El Niño/La Niña–related phenomena. The impacts are particularly felt in countries with the least capacity to respond because of overreliance on rain-fed agriculture, high levels of poverty, inadequate access to financial capital, and poor infrastructure. Since the adoption of the Hyogo Framework for Action in 2005, evidence suggests that early warning systems have generally been effective in alerting countries and stakeholders to impending hazards. However, the recurrence and magnitude of food crises in East and Southern Africa (ESA) underscore the need to improve prevention and response mechanisms in order to address the determinants and the dynamic nature of food insecurity, at both national and regional levels. Many countries in ESA have national platforms, legisla- tion and policies on disaster risk reduction, but few address agriculture, food security, and nutrition with sector-specific disaster risk reduction policies and objectives. Reducing risks and building resilience within agriculture requires a policy environment that is conducive to the full mainstreaming of disaster risk reduction within the sector. There have been cases where inadequate analysis, together with poor communication and ineffective coordination and response mechanisms, has contributed to acute food security emergencies that might have been pre- vented. There are also concerns about the accuracy and reliability of some data, links of food security information to trade policy, and private sector participation. Informing stakeholders and building consensus on the severity of food insecurity is vital, particularly in crisis situations to proactively reduce disaster losses in the sector, enable sector growth and protect the food secu- rity and nutrition of vulnerable populations. In-depth and regular informa- tion and analysis of food security vulnerability and resilience help countries to make better decisions and apply measures to protect and enhance their livelihoods. In order to fulfill the commitments made toward Sendai Framework for Disaster Risk Reduction 2015–2030, it is critical that countries, regional orga- nizations, development partners, and private sector focus their collaborative efforts on creating and strengthening institutional mechanisms that guide the development of the EWS. This will enable EWS to more effectively meet the decision-making needs of their primary users and evolve in a dynamic and sustainable manner. EWS for food security programs should not be per- ceived as part of the emergency response activities. They should become part Preface ix of an expanded food security information and analysis system that can pro- duce viable, relevant, and credible information necessary for responding to short-term emergencies as well as contributing to longer-term development programming. Makhtar Diop Vice President, Africa Region The World Bank x Preface Acknowledgments This work drew from the contributions of a range of experts working in agriculture, food security, climate change, and disaster risk management. We thank everyone who contributed to its richness and multidisciplinary outlook. The analytical work was carried out by a team comprising Ademola Braimoh, Bernard Manyena, Grace Obuya, and Francis Muraya. The report was pre- pared under the overall direction of Juergen Voegele, Ethel Sennhauser, and Mustapha Ndiaye, and under the guidance of Mark Cackler, Dina Umali- Deininger, and Preeti Ahuja. We are grateful for the contribution of the following colleagues: Stephen Njoroge (WMO), Ahmed Habbane (IGAD Secretariat), Bradwell Garanganga, (CSC/SADC), Peter Ambenje (Kenya Meteorological Department), Bruno Sekoli (Meteorological Association of Southern Africa), AbdiShakur Othowai (IGAD/ICPAC), Elliot Vhurumuku (WFP), Emily Massawa-Ojo (USAID/ PREPARED), Gideon Galu (FEWS NET), Kim Mhando (East Africa Grain Council), MaryLucy Oronje (Centre for Agriculture and Biosciences International), Nigist Biru (FEWS NET), Peter Muhangi (IFRC), Priscilla Muiruri (Kenya Agricultural Productivity & Agribusiness Project), Stanley Chabvunguma, (Department of Climate Change and Meteorological Services, Malawi), Susan Chomba (World Agroforestry Center), Venty Thierry (National Bureau for DRR, Madagascar), Yazan Elhadi (Adaptation Consortium), Wagayehu Bekele (Agricultural Transformation Agency, Ethiopia), Walter Nganyi (Kenya Meteorological Department), Birhanu Woldemikael (Ministry of Agriculture and Food Security, Ethiopia), Titus Ng’andu (Disaster Management and Mitigation Unit, Zambia), Generose Nziguheba (International Institute of Tropical Agriculture), David Nyamai (Ministry of Agriculture, Livestock and Fisheries, Kenya), John Mulenga (Ministry of Agriculture, Zambia), Joseph Intsiful (African Climate Policy Centre), and all those who provided substan- tive feedback to the EWS questionnaire. We also acknowledge the input of those who attended the workshop to deliberate on the findings of the field surveys. The report benefited greatly from invaluable suggestions from peer review- ers. We would like to thank Makoto Suwa and Vikas Choudhary (World Bank), and Jasper Mwesigwa (IGAD/ICPAC) for their insightful comments and suggestions. We thank Yisgedullish Amde, Deo Ndikumana, Holger Kray, and Stephen D’Alessandro for their assistance, and Pauline Zwaans, Joab Osumba, Rhoda Rubaiza, Alex Mwanakasale for their inputs toward a success- ful validation workshop. We also acknowledge the efforts of Priya Thomas, Tabrez Ahmed, Damalie Nyanja, Sophie Rabuku, Marie Lolo Sow, and Srilatha Shankar (World Bank) for assistance rendered at various stages of the project. Acknowledgments xi Acronyms and Abbreviations ACCRA Africa Climate Change Resilience Alliance AFSIS ASEAN Food Security Information System AGRHYMET Agriculture, Hydrology, and Meteorology ARC African Risk Capacity ASEAN Association of Southeast Asian Nations ASFR ASEAN Food Security Reserve AU African Union AWS automatic weather station BCR benefit-cost ratio CAADP Comprehensive Africa Agriculture Development Program CFSAM Crop and Food Security Assessment Mission CILSS Permanent Inter-State Committee for Drought in the Sahel COMESA Common Market for Eastern and Southern Africa CSC Climate Services Centre DEWS Karamoja Drought Early Warning System DGM General Directorate of Meteorology (Direction Generale de la Meteorologie) DMMU Disaster Management and Mitigation Unit DRF Disaster Risk Financing DRR disaster risk reduction EAC East African Community EAGC East African Grain Council EARS Environmental Analysis and Remote Sensing ECCAS Economic Community of Central African States ECMWF European Centre for Medium Range Weather Forecasts ENSO El Niño-Southern Oscillation EOC emergency operation center ESA East and Southern Africa EU European Union EW early warning EWS early warning system FAO Food and Agriculture Organization of the UN FAW fall armyworm FEWS NET Famine Early Warning Systems Network FPM focal point meeting FPMA Food Price Monitoring and Analysis FRA Food Reserve Agency FSNWG Food Security and Nutrition Working Group GDP Gross domestic product GDPFS Global Data Processing and Forecasting System GFCS Global Framework for Climate Services GFDRR Global Facility for Disaster Reduction and Recovery Acronyms and Abbreviations xiii GHACOF Greater Horn of Africa Climate Outlook Forum GIEWS Global Information and Early Warning System on Food and Agriculture GLOFAS Global Flood Awareness System GMB Grain Marketing Board HDI Human Development Index HEWS Humanitarian Early Warning Service ICPAC IGAD Climate Prediction and Applications Centre IFRC International Federation of Red Cross and Red Crescent Societies IGAD Intergovernmental Authority for Development INFORM Index for Risk Management IOC Indian Ocean Commission IPC Integrated Food Security Phase Classification KI key informant KII key informant interview MAAIF Ministry of Agriculture, Animal Industries and Fisheries MAM March, April, and May MIS Market Information System MoU memorandum of understanding MSD Meteorological Service Department NDVI Normalized Difference Vegetation Index NECOC Uganda National Emergency Coordination and Operations Centre NFBS national food balance sheet NGO nongovernmental organization NMHS National Meteorological and Hydrological Service NHS National Hydrological Service NMS National Meteorological Service NPV net present value NVAA National Vulnerability Assessment and Analysis NVAC National Vulnerability Assessment Committee O&M operation and maintenance ODK Open Data Kit PAD Project Appraisal Document PPP public-private partnership PSP participatory scenario planning RATIN Regional Agriculture Trade Intelligence Network REC Regional Economic Community RFBS regional food balance sheet RIIS Regional Integrated Information System RVAA Regional Vulnerability Assessment and Analysis RVAC Regional Vulnerability Assessment Committee SACU Southern African Customs Union SADC Southern African Development Community SARCOF Southern African Regional Climate Outlook Forum xiv Acronyms and Abbreviations SDG sustainable development goal SFDRR Sendai Framework for Disaster Risk Reduction SGR strategic grain reserve SOP standard operating procedure SMS Short Message Service SWC Severe Weather Consult TAHMO Trans-African Hydro-Meteorological Observatory TOT terms of trade UNDP United Nations Development Programme UNECA United Nations Economic Commission for Africa UNMA Uganda National Meteorological Authority UNISDR United Nations International Strategy for Disaster Reduction USAID U.S. Agency for International Development VAA Vulnerability Assessment and Analysis VAC Vulnerability Assessment Committee WFP World Food Programme WMO World Meteorological Organization ZFU Zimbabwe Farmers’ Union ZMD Zambia Meteorological Department Acronyms and Abbreviations xv Executive Summary The risk1 of the El Niño-induced food insecurity in southern Africa in 2016; the recent risk of famine in northern Kenya, Somalia, Ethiopia, and South Sudan; and the recent outbreak of the fall armyworm (FAW) in East and Southern Africa (ESA) all demonstrate that responses are still largely reactive than proactive. Inadequate early warning systems (EWSs), coupled with lim- ited investment and weak institutional and technical capacity, are implicated in contributing to food insecurity–related emergencies in ESA. Yet over the years, strong evidence has emerged on the benefits of investing in EWSs. In Ethiopia, investing in a drought EWS, which would reduce livelihood losses and dependence on assistance, has a benefit-cost ratio (BCR) of between 3:1 and 6:1. Similarly, the BCR of improving national hydrometeorological ser- vices in developing countries ranges from 4:1 to 36:1. Consistent with one of the goals of the Sendai Framework for Disaster Risk Reduction (SFDRR), increasing investment in EWSs would contribute to a substantial increase in the availability of, and access to multihazard2 and disaster risk information, one of the key inputs in achieving the sustainable development goals (SDGs). In supporting these efforts, an assessment of food security EWSs was con- ducted to improve food security and resilience3 in East and Southern Africa. The study aimed at assessing “bottlenecks” and opportunities for improving food security EWSs for enhanced resilience in ESA. The performance and capacity of EWSs at the Regional Economic Cooperation level and at sampled member states were assessed. The study drew lessons from the Association of Southeast Asian Nations (ASEAN) region, particularly in relation to EWS pol- icies, investments, and technical capacities. Because the assessment attempted to be as comprehensive as possible, it was essential to also draw lessons from programs and projects being piloted on EWSs, including those being imple- mented by the World Bank in Malawi, Zambia, Zimbabwe, Kenya, Somalia, Tanzania, and Mozambique. Although progress has been made, there are challenges that recur across the Regional Economic Communities (RECs) and member states, and these challenges fall into three main categories: 1 Risk is used here to mean the potential for consequences where something of value is at stake and where the outcome is uncertain, recognizing the diversity of values. Risk is often represented as the probability of occurrence of hazardous events or trends multiplied by the consequences if the events occur. Risk results from the interaction of vulnerability, exposure, and hazard (IPCC 2014). 2 Hazard is the potential occurrence of a natural or human-induced physical event or trend, or physical impact, that may cause loss of life, injury, or other health impacts, as well as damage and loss to property, infrastructure, livelihoods, service provision, and environmental resources (IPCC 2014). 3 Resilience means the capacity of a social-ecological system to cope with a hazardous event or disturbance, responding or reorganizing in ways that maintain its essential function, identity, and structure, while also maintaining the capacity for adaptation, learning, and transformation (IPCC 2014). Executive Summary xvii Institutional Challenges • Lack of EWS working groups in both RECs to coordinate EWS activities. In the Intergovernmental Authority for Development (IGAD) region, the institutionalization of the Food Security and Nutrition Working Group (FSNWG) has been slow because the FSNWG has yet to be endorsed as an institution of IGAD. • Policies that outline roles and responsibilities for EWS actors at both regional and national levels are generally weak. Although many ESA coun- tries have sector policies, the sectors still operate in silos because of lack of overarching EWS policies. In addition, although several tools such as the Integrated Food Security Phase Classification (IPC) have manuals that guide users, these have not been endorsed by regions and member states to provide guidance on systematic data collection, data sharing, monitor- ing, and agreed action triggers. • Although the Greater Horn of Africa Climate Outlook Forum (GHACOF) report focuses more on the actions that users need to consider compared with the Southern African Regional Climate Outlook Forum (SARCOF) reports, which are still expressed in probabilities, some users such as pas- toralists are still excluded because the reports are in English, which they do not understand. Whereas the start and end dates for the season are useful, there is still a gap in intensity and frequency of the information and it has yet to reach some of the intended users. • There is a lack of regular updates of the regional food balance sheets (RFBSs) and national food balance sheets (NFBSs) and weak monitoring of grain markets, cross-border trade, and commodity price monitoring. Although the Famine Early Warning Systems Network (FEWS NET) and the World Food Programme (WFP) actively monitor these activities, inte- gration of some of these into regional and national systems is still limited. Technical Challenges • Although the SARCOF and GHACOF processes are well established in providing regional climate forecasts, there are still challenges to down- scaling these forecasts to local levels, such as districts or villages. Another limitation of GHACOF and SARCOF is that they tend to focus on rain and pay little attention to other weather parameters. • Limited coverage of the weather observation networks and challenges in crop production forecasts make agrometeorology data less reliable. The capacity of national meteorological and hydrological services4 (NMHSs) in Africa is not adequate and has considerably degraded in some countries during the past 20 to 25 years. 4National Meteorological and Hydrological Service refers to a National Meteorological Service (NMS) or National Hydrological Service (NHS), or an organization that combines the functions of both (WMO 2012a). The plural, NMHSs, refers to multiple organizations (NMHS, NMS, and NHS). xviii Executive Summary • In some countries, there is a lack of technically qualified professionals such as meteorologists, agrometeorologists, and hydrometeorologists to ensure quality hydrometeorological products. • Although the Vulnerability Assessment Committee (VAC) system has become one of the most useful and reliable EWS tools in the Southern African Development Community (SADC), there are multiple method- ologies that need harmonization. Similarly, the IPC, which continues to gain currency there, has challenges in the comparability of outcome indi- cators because some countries use actual food security outcome indica- tors whereas others make inferences without using the actual data. • Weak food security information systems and absence of a framework for sharing data at both the regional and national levels make EWS informa- tion slow to reach users. Sustainability and Financial Challenges • There are no clear funding mechanisms for EWSs because EWS programs take a reactive rather than a proactive approach. Consequently, because early warning (EW) is considered an emergency response activity, the funding tends to be ad hoc and therefore competes with development funds during an emergency response. In addition, because EWSs rely on international assistance, which tends to be project based, they often face the problem of financial sustainability once external funding ceases. Instead, EWSs should be considered as part of regular development because EWS data are used for planning interventions. • There is limited public-private partnership (PPP) with private climate services providers, yet such partnerships would reduce reliance on donors. There is uncertainty over the nature of the relationship between the NMHSs and the private sector. There is also a perception that the pri- vate sector may be a threat to job security, data security and ownership, and government obligations to supply public goods5 such as EWs. Recommendations Key priority recommendations for targeted investment are outlined as fol- lows: • Develop and strengthen the food security information system at both national and regional levels to meet the RECs’, African Union’s (AU), and member states’ agendas, including the Comprehensive Africa Agriculture 5 A public good is a good that no consumer can be excluded from using if it is supplied and for which consumption by one consumer does not reduce the quantity available for consumption by any other. The first property is referred to as nonexcludability, whereas the latter is termed nonrivalry (Black, Hashimzade, and Myles 2017). Executive Summary xix Development Program (CAADP). The information should contribute to ongoing development programs, as well as to improving the effectiveness of the EWSs, emergency preparedness, and response capacity. The RECs in the ESA region should consider establishing a food security informa- tion system, taking into consideration lessons learned from similar initia- tives, including the ASEAN Food Security Information System (AFSIS) to strengthen food security EWSs. This effort should be supported by a data-sharing framework and a one-stop food security information hub such as an emergency operation center (EOC) that is accessible to rele- vant stakeholders, including government agencies, and the international community. The regional food security information system should also be replicated in each of the member states. • Support the strengthening of the EWS legal, regulatory, and institu- tional frameworks as well as improving coordination and ensuring clarity of roles and responsibilities within and across the four compo- nents of effective EWSs. This will include developing common methodol- ogies and procedures for data collection, management, and data sharing across geographical borders, as well as developing effective strategies for the timely dissemination of actionable warnings. • Promote south-south knowledge exchanges, for example, exchange of information between AFSIS and RECs in the process of developing their food security information systems. Inter-REC knowledge exchanges such as Agriculture, Hydrology, and Meteorology (AGRHYMET) Regional Centre of the countries of the Permanent Inter-State Committee for Drought in the Sahel (CILSS) could also be considered. • Invest in technical capacity development to enable the collection of high-quality agrometeorological crop production forecasts and vulner- ability data. EWSs require (a) improved capacity to downscale global and regional climate forecasts to high resolution for the forecasts to be meaningful at the local level; (b) strong weather observation networks with a wider coverage; and (c) improved data collection for crop assess- ments, livestock assessments, and vulnerability assessments. The SADC’s Vulnerability Assessment and Analysis (VAA) as well as the IPC method- ologies should be harmonized or at least agree on minimum indicators to ensure quality assurance and comparison between countries. • Strengthen public commitment and mainstream EWS considerations into agricultural/food security policies, budgetary allocations, and plan- ning frameworks. This will require evidence-based advocacy to national and regional leaders and cooperation development partners on the eco- nomic benefits of EWSs. • Support the development of tools to support vulnerable households and communities to establish household community systems that can respond to emergencies. xx Executive Summary CHAPTER 1 The Imperative of Early Warning6 Systems in East and Southern African Regions Background A common understanding has emerged over the decades, particularly since 2000, that disasters mainly triggered by hydrometeorological hazards have become a constant occurrence in Sub-Saharan Africa (figure 1.1). These hazards are manifested in the El Niño-Southern Oscillation (ENSO), the largest mode of interannual variability in the climate system (Murphy et al. 2001), whose frequency has increased in the countries of East and Southern Africa (ESA).7 There is strong evidence that higher temperatures, droughts, floods, and changing weather8 patterns expected from climate change will exacerbate the disaster risks associated with hydrometeorological hazards (figure 1.2 and box 1.1). The adverse impact of ENSO events is exacerbated by chronic environmental and socioeconomic conditions that compound to worsen food  insecurity conditions across ESA. Improved regional and national weather, water, and climate-related monitoring and forecasting capabilities 6 Early warning (EW) is the provision of timely and effective information, through identified institutions, that allows individuals exposed to a hazard to take action to avoid or reduce their risk and prepare for effective response (UNISDR 2009). In this study, although EWSs are adopted, experience shows that EWSs alone do not prevent hazards from turning into disasters. Early action is essential, particularly given the increasing accuracy of seasonal forecasts. However, even with timely EWs and planned early action, people suffer the disastrous consequences of natural hazards. 7 The east African countries, under the IGAD region, are Burundi, Eritrea, Ethiopia, Djibouti, Kenya, Rwanda, Somalia, South Sudan, Sudan, and Uganda. The southern African countries under the SADC are Angola, Botswana, the Democratic Republic of Congo, Lesotho, Madagascar, Malawi, Mauritius, Mozambique, Namibia, Seychelles, South Africa, Swaziland, Tanzania, Zambia, and Zimbabwe. 8 Weather refers to atmospheric phenomena that have timescales that range from hours to days to 1 or 2 months, whereas climate refers to atmospheric conditions that have timescales that range from a few months to a season to a year to a decade or more, or even longer. In this sense, the terms weather and cli- mate identify regions along a continuous spectrum of atmospheric conditions, weather describing rapidly changing events, and climate describing slowly changing ones. Climate can be represented in terms of nor- mal, long-term average, and year-to-year fluctuations—the interannual variability—on which that average when viewed over a period of a few hundreds of years has fallen within a bounded “range” of values. Common drivers of climate variability are the oscillations that occur in Earth’s coupled ocean-atmosphere system. An example is the El Niño and La Niña (ENSO) events, shifts of warm, tropical Pacific Ocean cur- rents that can dramatically affect world seasonal weather patterns. Other drivers include volcanic eruptions and solar phenomena. Sometimes climate varies in ways that suggest a component of randomness being inherent in Earth’s climate system. Climate change is a long-term continuous change (either increase or decrease) in a climate normal (for example, an increase in the long-term average temperature) and/or the range of climate variability (for example, more frequent, more intense thunderstorms together with fewer small showers). As the range increases, the year-to-year variations in a variable such as temperature or precipitation should be expected to be greater, and so new extreme values are likely. CHAPTER 1—The Imperative of Early Warning 1 FIGURE 1.1 Number of Reported Disasters and People Affected by Disaster Type in Sub-Saharan Africa 40 140 35 Total no. disaster events per year 120 People affected (millions) 30 100 25 80 20 60 15 40 10 5 20 0 0 1970 1975 1980 1985 1990 1995 2000 2005 2010 Drought Flood Storms Earthquake Volcano No of events Source: GFDRR 2016, 1. contribute to enhancing EWSs, including those focused on food security. Adverse weather and climate conditions, such as those triggered by ENSO events, can affect entire regions. Governments can use improved monitor- ing and forecasting for improving and protecting agricultural productivity and production, contributing to protecting lives and the livelihoods of those households engaged in agriculture and other climate-sensitive sectors. The observed shifting patterns in weather and climate have caused rainfall anomalies in ESA countries (figure 1.3), with these shifts having their impact on agriculture, health, migration, and conflict, exacerbating existing vulner- abilities. Weather- and climate-related hazards combined with social vulner- ability9 drivers pose a threat to lives, livelihoods, economic activities, and socioeconomic assets. The major agricultural vulnerability drivers include poor agricultural production, loss of livestock, high food prices, cross-border trade barriers, growing economic interdependence, poverty, and civil con- flict. The drought and floods of the 2015–16 El Niño phenomenon crippled agricultural production in a number of ESA countries, exacerbated chronic food insecurity in some regions, and triggered acute food insecurity in others, threatening millions of vulnerable households (figure 1.4). The climate anomalies have wide-ranging impacts in ESA. Although glob- ally 108 million people from 18 countries faced crises-level food insecurity or worse in 2016 (Integrated Food Security Phase Classification [IPC] Phase 3 and above), 10 of these countries were from ESA (figure 1.4). What distin- guishes the ESA countries from the rest of the world is the extent to which 9Vulnerability is the propensity or predisposition to be adversely affected. It encompasses a variety of con- cepts, including sensitivity or susceptibility to harm and lack of capacity to cope and adapt (IPCC 2014). 2 CHAPTER 1—The Imperative of Early Warning FIGURE 1.2 Summary of Climate Impacts and Risks in Sub-Saharan Africa Observed Approximately vulnerability 1.5°Ca,b Approximately Approximately Approximately Risk/impact or change (≈2030sc) 2°C (≈2040s) 3°C (≈2060s) 4°C (≈2080s) Heat Unusual heat Virtually 20–25 percent 45 percent of 70 percent of >85 percent of extremed (in extremes absent of land land land land the Southern Unprecedented absent <5 percent of 15 percent of 35 percent of >85 percent of Hemisphere heat extreme land land land land summer) Drought Increasing Likely risk of Likely risk of Likely risk of Likely risk of drought risk severe drought severe extreme extreme drought in Southern, in Southern drought in drought in in Southern Central and and Central Southern and Southern Africa Africa and West Africa, Africa, Central Africa, and severe severe drought decrease in increased risk increased risk drought in in Central Africa, East Africa, in West Africa, in West Africa, Central Africa, increased risk but West decrease in decrease in increased risk in West Africa, and East East Africa but East Africa but in West Africa, decrease in African West Africa West Africa decrease in East Africa, projections and East and East East Africa, but but West and are African African West and East East African uncertain. projections are projections are African projections are uncertain uncertain projections are uncertain uncertain Aridity Increased Less change Area of Area of drying expected hyper-arid and hyper-arid and arid regions arid regions grows by grows by 10 3 percent percent. Total arid and semi-arid area increases by 5 percent Sea-level rise above present About 21cm 30 cmf by 30 cm – 2040s 30 cm by 30 cm by 2040s (1985–2005) to 2009e 2040s 2040s 50 cm by 2070 50 cm – 2070 50 cm by 2060 50 cm by 2060 70 cm by 70 cm by 90 cm by 105 cm by 2080–2100 2080–2100 2080–2100 2080–2100 Source: UNDP, 2016, 6. a Refers to the global mean increase above pre-industrial temperatures. b Years indicate the decade during which warming levels are exceeded in a business-as-usual scenario exceeding 4°C by the 2080s. c Years indicate the decade during which warming levels are exceeded with a 50 percent or greater change (generally at the start of the decade) in a business-as-usual scenario (RCP8.5 scenario). Exceedance with a likely chance (>66 percent) generally occurs in the second half of the decade cited. d Mean heat extremes across climate model projections are given. Illustrative uncertainty range across the models (minimum to maximum) for 4°C warming are 70–100 percent for unusual extremes, and 30–100 percent for unprecedented extremes. The maximum frequency of heat extreme occurrence in both cases is close to 100 percent, as indicator values saturate at this level. e Above 1880 estimated global mean sea level. f Add 20 cm to get an approximate estimate above the pre-industrial sea level. CHAPTER 1—The Imperative of Early Warning 3 BOX 1.1 Projected Climate Change Impacts in Sub-Saharan Africa With an increase of 4°C of global warming by the end of the century, sea levels are projected to rise up to 100 cm, droughts are expected to become increasingly likely in central and southern Africa, and never-before-experienced heat extremes are projected to affect increasing proportions of the region. Projections also show a growing probability of increased annual precipitation in the Horn of Africa and parts of east Africa, which is likely to be concentrated in heavy downpours and, thereby, increase the risk of flooding. Sub-Saharan Africa is particularly vulnerable to impacts on agriculture. Most of the region’s agricultural crop production is rain fed and therefore highly susceptible to shifts in precipitation and temperature. A net expansion of the overall area classified as arid or hyperarid is projected for the region as a whole, with likely adverse consequences for crop and livestock production. Savannah grasslands may be reduced in area, with potential impacts on livelihoods and pasto- ral systems. By the time global warming reaches 3°C, savannahs are projected to decrease from about a quarter at present to approximately one-seventh of the total land area, reducing the avail- ability of food for grazing animals. Source: UNDP 2016, 5–6. FIGURE 1.3 Rainfall Anomalies in ESA October 2015–February 2016 Rainfall Anomalies October–December 2016 Rainfall Anomalies (Percentage of the 1982–2011 Average) (Percentage of the 1981–2010 Average) for Southern Africa for East Africa Source: FEWS NET 2016, 2017. 4 CHAPTER 1—The Imperative of Early Warning FIGURE 1.4 Food Security IPC Crisis Phase 3 and Above (January 2017) 15 Percentage of population in IPC phase 3+ Population in IPC phase 3+ in millions 12 50% 49% 48% 46% 41% 40% 38% 9 32% 25% 22% 6 15% 15% 12% 3 9% 9% 8% 8% 4% 0 Yemen Ethiopia Afghanistan North Nigeria including northeast Nigeria Syria Malawi DRC South Sudan Sudan Zimbabwe Somalia Burundi CAR Mozambique Iraq Haiti Chad Southern Madagascar Source: FSIN 2017, 17. food security10 has been undermined by a combination of El Niño-induced drought conditions and conflict. Climate variability and change, along with weather extremes and conflict, have put pressures on food production, mar- keting, and humanitarian systems at a time when resources and capacity are already strained. Projections for early 2017 indicated an increase in the severity of food insecurity in these regions, particularly in Ethiopia, Kenya, and Somalia. In Ethiopia, parts of the Somali Region are expected to remain in Emergency (IPC Phase 4) through January 2018—particularly Dollo, Korahe, and Afder zones—in the absence of humanitarian assistance. In Somalia, an estimated 2.5 to 3 million people will remain in need of humanitarian assistance through the end of 2017, with large areas expected to be in Emergency (IPC Phase 4) in the absence of assistance. According to the findings of the July 2017 Long 10 For consistency with the global conceptualization of food security, the 1996 World Food Summit defi- nition of food security is adopted as the following: “Food security at the individual, household, national, regional and global levels [is] when all people, at all times, have physical and economic access to sufficient, safe and nutritious food to meet their dietary needs and food preferences for an active and healthy life” (FAO 1996). In this case, food security is the absence of, among others, food insecurity, hunger, starva- tion, and nutrition insecurity. Food security metrics, which also form the basis for EWSs, should at mini- mum consider understanding the hazards vulnerabilities, which affect food availability, access, utilization, or some combination of these metrics, considering social and cultural dimensions. For this assessment, a combination of food balance sheet approaches, physical and economic access, utilization (in terms of nutrition), and the social and cultural aspects (in terms of governance) are considered. CHAPTER 1—The Imperative of Early Warning 5 Rains Assessment carried out by the Kenya Food Security Steering Group (KFSSG), 800,000 people are in Stressed (IPC Phase 2) and 2.6 million are in Crisis (IPC Phase 3) and in need of urgent humanitarian assistance. KFSSG indicated that the Stressed (IPC Phase 2) population is likely to move to Crisis (IPC Phase 3) over the next few months. The recent food crisis in ESA demonstrates that responses are still largely reactive rather than proactive. In 2016, the southern African region experi- enced a historic El Niño-induced drought, which triggered a food crisis that has been described as the worst in 35 years. Approximately 40 million people were in need of humanitarian assistance, and about 2.7 million children were suffering from malnutrition in the subregion. There was a cereal production shortfall of approximately 9.3 million tons. As a result, the South African Development Community (SADC) declared a state of regional disaster in June 2016 and appealed for US$2.4 billion for the support of the humanitar- ian needs of the affected population (SADC 2016a). The pressure on food security in Sub-Saharan Africa is likely to be further exacerbated by the recent threats of the fall armyworm (FAW) (Spodoptera frugiperda) outbreak (figure 1.5), which may be related to warmer global tem- peratures over the past few years (CIMMYT 2017). Environmental and cli- matic analyses of Africa show that the FAW is likely to build permanent and significant populations in west, central, and southern Africa and spread to other regions when weather or temperatures are favorable. Prediction mod- els so far present much uncertainty because scientific institutions are still learning of the pest’s habits and environmental suitability on the continent (Abrahams et al. 2017). Notwithstanding the limited scientific knowledge on the FAW in Africa, in 2016 the FAW was identified in 11 countries and was suspected in at least 14 other countries. The economic impact of the FAW on maize, sorghum, rice, and sugarcane in Africa is estimated to be approximately US$13,383 million (table 1.1). This does not consider up to 80 other crops the insect has been known to feed on, as well as subsequent seeds lost for future growing seasons. If the FAW threat materializes, this will likely put further pressure on food security. The observed and projected impacts of a changing climate on food security in the ESA countries, exacerbated by evolving threats, for example, pest infes- tations such as FAW, are projected to worsen existing social vulnerabilities, highlighting the urgent need for improving the effectiveness and efficiency of early warning systems (EWSs). Inadequate EWSs and existing gaps in the effective flow of information across agencies and between levels of govern- ment administration hinder the governments’ capacity to prepare for and respond to food insecurity-related emergencies in ESA countries (Tadesse et al. 2008). It is hypothesized that had there been effective EWS and adequate communication systems in place for the dissemination of actionable warn- ings to the targeted audiences, the impact of the El Niño-induced drought could have been less severe in the ESA region. A growing body of evidence 6 CHAPTER 1—The Imperative of Early Warning FIGURE 1.5 Known and Suspected Distribution of FAW in Africa (April 2017) Source: Abrahams et al. 2017. TABLE 1.1 Estimated Economic Impact of the FAW FAW affected crops Total production (million tons) Yield loss Estimated/predicted loss in all countries assuming no FAW (million tons) (US$, million) Maize 67.0 13.5 3,058 Sorghum 25.5 1.9 827 Rice, paddy 17.1 9.6 6,699 Sugarcane 90.1 46.0 2,799 Total 13,383 Source: Abrahams et al. 2017. CHAPTER 1—The Imperative of Early Warning 7 shows that effective EWSs not only save lives but also help protect livelihoods and enhance national development gains (United Nations 2006). Only a few cost-benefit analyses11 of EWSs have been carried out in devel- oping countries. However, those so far completed provide strong evidence on the benefits of investing in establishing and improving EWSs. For instance, Hallegatte’s 2012 study, although focused on the hydrometeorological infor- mation production element of EWSs, without the social vulnerability informa- tion, postulates that if hydrometeorological and early warning (EW) capacities in all developing countries were upgraded to the level of developed country standards, such investments would result in at least three potential benefits: • Avoidable asset losses of between US$300 million and US$2 billion per year because of natural hazards; • Avoidable human losses of about 23,000 per year, estimated between US$700 million and US$3.5 billion per year; and • Economic benefits between US$3 billion and US$30 billion per year with benefit-cost ratios (BCRs) between 4 and 35 with cobenefits Although the ESA countries and their regional bodies have made efforts in developing EWSs,12 there is still a gap in investing in EWSs to enhance pre- ventive, anticipatory, and absorptive capacity to food insecurity. Investing in EWSs, including cost-benefit analysis, is explored in further detail in chapter 4. Similarly, a number of studies have been conducted on the status of food security EWSs, for example, Tefft, McGuire, and Maunder (2006) and the United Nations Economic Commission for Africa (UNECA) (2011). Their assessment of food security EWSs in Sub-Saharan Africa took note of the well- known challenges of effective food security EWSs that were also highlighted by numerous studies related to an overly technical approach, externally imposed methods and institutional models, and short-term project horizons. They, however, attribute the primary cause of lack of progress in EWSs to insuf- ficient attention to, and inadequate investment in, developing institutional mechanisms and capacity for an effective, sustainable, and demand-driven EWS. The UNECA study that focused on enhancing the effectiveness of food security information systems in the SADC region recommended the need for strengthening as well as integrating food security information systems, harmo- nizing methodologies, strengthening partnerships, building capacity, creating 11 Benefit-cost analysis or cost-benefit analysis is the quantification of the total social costs and social benefits of a policy or a project, usually in monetary terms. The costs and benefits concerned include not only direct pecuniary costs and benefits, but also externalities, meaning external effects not traded in markets. These include external costs, for example, pollution, noise, and disturbance to wildlife, and external benefits such as reductions in traveling time or traffic accidents. Benefit-cost analysis is often used to compare alterna- tive proposals. If the total social benefits of an activity exceed total social costs, this can justify subsidizing projects that are not privately profitable. If the total social costs exceed total social benefits, this can justify preventing projects even when these would be privately profitable (Black, Hashimzade, and Myles 2017). 12 This includes the Regional Agriculture Trade Intelligence Network (RATIN) in east Africa and Regional Vulnerability Assessment and Analysis (RVAA). 8 CHAPTER 1—The Imperative of Early Warning financial and institutional sustainability, and strengthening communication and dissemination, all of which are critical to an effective EWS. The findings of these studies are by no means less important. With increasing exposure and susceptibility of the agriculture systems to climate change and variabil- ity, it is critical to regularly update the EWS knowledge base to appropriately respond to the changing nature of risk. Besides, the 2014 Malabo Declaration of the African Union (AU), which recommitted the member states to adopt the Comprehensive Africa Agriculture Development Program process and to ending hunger and take action of improving nutrition by 2025, recognizes the importance of EWSs. The Malabo Declaration encourages the member states to commit budget lines within their national budgets for strengthening EWSs to facilitate advanced and proactive responses to disasters and emergencies with food and nutrition security implications.13 More important, an increase in the availability of and access to multihazard EWSs and disaster risk infor- mation and assessments to the communities by 2030 is one of the key inter- ventions needed for achieving the goals of the Sendai Framework for Disaster Risk Reduction (SFDRR) and Sustainable Development Goals (SDGs). This recommendation has been further supported by the Yaoundé Declaration on the Implementation of the Sendai Framework in Africa of 2015 and the Mauritius Declaration on the Implementation of the Sendai Framework in Africa of 2016. Four Elements of an Effective People-Centered Early Warning To comprehensively analyze the data from the stakeholders’ consultations, the four commonly cited elements of EWSs (Basher 2006; IFRC 2009) were adopted (figure 1.6). United Nations International Strategy for Disaster Reduction (UNISDR) (2006) argues that a complete and effective EWS comprises four interrelated elements, spanning knowledge of hazards and vulnerabilities to preparedness and capacity to respond: risk knowledge, a monitoring and warning service, dissemination and communication, and response capability. Risk Knowledge Understanding risk and the risk drivers, as emphasized by the SFDRR, is the first step toward building effective EWSs. Risks arise from the combi- nation of hazards, exposure, and vulnerabilities at a location. Assessments of risk require systematic, standardized collection and analysis of data and should consider the dynamic nature of hazards and vulnerabilities that arise from processes such as urbanization, rural land-use change, environmental 13More details on the Malabo Declaration can be found at https://www.au.int/web/sites/default/files /documents/31247-doc-malabo_declaration_2014_11_26.pdf. CHAPTER 1—The Imperative of Early Warning 9 FIGURE 1.6 Elements of an EWS MONITORING & RISK KNOWLEDGE WARNING SERVICE Systematically collect data and Develop hazard monitoring and undertake risk assessments early warning services Are the hazards and the Are the right parameters vulnerabilities well known? being monitored? What are the patterns and Is there a sound scientific trends in these factors? basis for making forecasts? Are risk maps and data Can accurate and timely widely available? warnings be generated? DISSEMINATION & RESPONSE CAPABILITY COMMUNICATION Communicate risk information Build national and community and early warnings response capabilities Do warnings reach all Are response plans up of those at risk? to date and tested? Are the risks and Are local capacities and warnings understood? knowledge made use of? Is the warning information Are people prepared and clear and useable? ready to react to warnings? Source: UNISDR 2006. degradation, and climate change. Risk assessments and maps help motivate people, prioritize EWS needs, and guide preparations for disaster prevention and responses. Monitoring and Warning Service Warning services lie at the core of the system. There must be a sound scien- tific basis for predicting and forecasting hazards and a reliable forecasting and warning system that operates 24 hours a day. Continuous monitoring of hazard parameters and precursors is essential to generate accurate warnings on time. Warning services for different hazards should be coordinated where possible to gain the benefit of shared institutional, procedural, and commu- nications networks. DisseminaƟon and CommunicaƟon Warnings must reach those at risk. Clear messages containing simple, useful information are critical to enable proper responses that will help safeguard lives and livelihoods. Regional-, national-, and community-level communi- cation systems must be identified beforehand and appropriate authoritative voices established. The use of multiple communication channels is necessary to ensure as many people as possible are warned, to avoid failure of any one channel, and to reinforce the warning message. 10 CHAPTER 1—The Imperative of Early Warning Response Capability It is essential that communities understand their risks, respect the warning ser- vice, and know how to react. Education and preparedness programs play a key role. It is also essential that disaster management plans are in place, well practiced, and tested. The community should be well informed on options for safe behavior, available escape routes, and how best to avoid damage and loss to property. Best practice EWSs also have strong interlinks and effective communica- tion channels between all the elements. The four elements of EWSs need to be coordinated across many agencies at national to local levels for the system to work. Failure in one element or lack of coordination across them could lead to the failure of the entire system. The issuance of warnings is a national responsibility; thus, roles and responsibilities of various public and private sector stakeholders for implementation of an EWS should be clarified and reflected in the national to local regulatory frameworks, planning, budgetary, coordination, and operational mechanisms. Purpose and Scope of the Report This report presents the findings and recommendations of the regional assessment of the EWSs for enhancing food security in ESA. The objective of the assignment was to assess “bottlenecks” and opportunities for improv- ing food security EWSs to enhance resilience in ESA, which comprises 25 countries and 5 African Union Regional Economic Communities (RECs).14 EWS producers and users, drawn from governmental and nongovernmental agencies as well as local communities, were consulted in 7 out of 25 countries in eastern (includes members of the East African Community [EAC] and Intergovernmental Authority for Development [IGAD]) and southern Africa (includes member states from the SADC and Indian Ocean Commission [IOC]). All members of these RECs are also members of the Common Market for East and Southern Africa (COMESA). The country selection criterion was based on the level of disaster risk, according to the Index for Risk Management (INFORM) classification.15 The INFORM index is a measure based on an assessment of the countries’ hazard exposure, vulnerability, lack of coping capacity, and the level of development using the Human Development Index (HDI) (table 1.2). This assessment also drew on lessons learned from initia- tives across the Association of Southeast Asian Nations (ASEAN) region, par- ticularly in relation to EWS policies, investments, and technical assistance. For this purpose, the following activities were undertaken: • Map the EWS methods and assess their suitability and the extent to which the technical skills and capacity and operating models of regional and 14 There are seven RECs in ESA. Of these, five are recognized by the AU: EAC, SADC, IGAD, COMESA, and Economic Community of Central African States (ECCAS). In addition, there are two RECs that are not recognized by the AU: Southern African Customs Union (SACU) and IOC. 15 Index for Risk Management, INFORM, available at http://www.inform-index.org/. CHAPTER 1—The Imperative of Early Warning 11 TABLE 1.2 Participating Countries in the Assessment Reason for selection Regional membership Country HDI Risk class EAC COMESA IGAD IOC SADC Ethiopia Low High X X X Kenya Medium High X X X Malawi Low Medium X X Madagascar Low Very high X X X Mozambique Low High X X Zambia Medium Medium X X Zimbabwe Low Medium X X national organizations address user needs and adverse weather (El Niño) preparedness. • Evaluate the efficiency and effectiveness of existing EWSs in terms of grain markets, cross-border trade, commodity price monitoring, and seasonal-scale climate events. • Evaluate EWS policies and their alignment to meeting the needs of various users. • Evaluate the cost-effectiveness and sustainability of EWSs in the two subregions. The remainder of this report is structured into four major thematic chapters: • Chapter 2 maps out the EWS methods, the technical skills and capacity for weather and seasonal forecasting, weather observation, crop forecast- ing, vulnerability assessments, and grain, market, price, and commodity monitoring at the national and regional levels. The chapter assesses the suitability, effectiveness, and efficiency, as well as the strengths and weak- nesses and sustainability of these methods in addressing the user needs and their importance in informing preparedness and response. • Chapter 3 evaluates the performance of EWSs in meeting users’ needs using the elements of the people-centered framework (figure 1.6), namely, the production of risk information, risk monitoring and warning service, early warning (EW) communication and dissemination, response capa- bility, and EW governance at both the national and regional levels. This includes the assessment of the ability and capacity of the national meteo- rological and hydrological services (NMHSs), IGAD Climate Prediction and Applications Centre (ICPAC), the SADC Climate Services Centre (CSC), the involvement of the private sector, and resource mobilization to ensure the sustainability of EW programs. • Chapter 4 discusses the rationale for investing in food security EWSs and draws some examples of such investment from World Bank projects. Cost-effectiveness approaches and operating models and public-private partnerships (PPPs) are explored for future guidance that informs the food security EWSs in ESA. 12 CHAPTER 1—The Imperative of Early Warning • Chapter 5 provides selected best practices drawn from within and outside the region. The highlights include the potential role of the private sec- tor in providing climate information services; the ASEAN Food Security Information System (AFSIS); the application of climate and vulnerabil- ity information in local planning, which also provides opportunities to integrate indigenous EW information; and the use of climate information forums to help improve the quality of EW information and services. The chapter concludes by providing recommendations to guide future policy approaches and investments in food security EWS at both the national and regional levels in ESA. CHAPTER 1—The Imperative of Early Warning 13 CHAPTER 2 Early Warning Methods, Technical Skills, and Capacity Introduction Regional and national food security EWS in ESA are supported by a diversity of internationally recognized methodologies (table 2.1). These methodolo- gies and approaches are generally consistent with the Global Information and Early Warning System on Food and Agriculture (GIEWS) of the Food and Agriculture Organization of the United Nations (FAO) and Humanitarian Early Warning Service (HEWS) developed by the World Food Programme (WFP). The FAO-GIEWS, developed in the 1970s following the world food crisis, provides information on countries facing food insecurity through monthly briefing reports on crop and food prospects, including drought information, together with an interactive map of countries in crisis. The ques- tions addressed include the following: • What EW methods are used to assess the risk to food insecurity? • Are technical skills and capacities adequate and suitable to address user needs as well as the potential to reduce the adverse impact of weather- related hazards? • To what extent are the EW methods for grain markets, commodity price, and cross-border trade effective and efficient? • Are the methods cost effective and sustainable? In response to these questions, the common EW methods in the ESA countries can be divided into two broad categories. First are those related to understanding the nature of the hazard, such as an in situ observation system for weather and climate data, hydrological data (runoff data), and topological data (for example, elevation database to link runoff forecast with flood exten- sion). These data are obtained from two main sources: weather stations and weather satellites. Usually owned and run by the NMHSs, weather stations record a range of physical measurements of the environment and produce accurate, fine-scale data that are useful for informing and calibrating weather prediction models. Most datasets are generated by weather satellites, pro- viding nowcasting16 information on rain, lightning, temperature, and wind and are freely available to the NMHSs on online platforms such as SAT2420. 16 Nowcasting is the detailed description of the current weather along with forecasts obtained by extrapola- tion for 0 to 6 hours in the future (WMO 1992). CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity 15 TABLE 2.1 Common Methods for Generating Risk Information for an EWS 16 Theme Purpose Methods Indicators Coverage/users Impact Challenges Meteorology (drought, Weather and climate Weather prediction Average rainfall, National and Inform planting Resource constraints, floods, thunderstorms, forecasting models, satellite data, and average temperature, regional levels; and harvest dates, weak observation network, and so on) global circulation and climate forecasts commercial and off-farm activities, downscaling forecasts to subsistence and disaster high resolution, and producers (scenario) dissemination language preparedness not tailored to users plans Hydrology (water) Hydrological Collection of river flow and Annual rainfall pattern, National and Inform warning on Resource constraints, information and dam levels dam monthly capacity, regional levels; energy production weak observation network, information on dam and water supply commercial and and water and weak transboundary capacities subsistence management agreements producers Pest infestations (for Information on pest Crop monitoring and Climate forecasts and National level; Informs Limited information on the example, locusts and infestations surveillance assessment and surveys leaf tissue commercial and contingency plans FAW, weak institutional the FAW) subsistence and pesticide capacity, resources producers stockpiles constraints, and weak transboundary agreements Agriculture production Provides information on Crop forecasts (planting Crop/livestock National and Update food Human resource and (cereal and livestock) crop and livestock and preharvest surveys) production, yields regional levels; balance sheets financial constraints, estimates and cereal and postharvest surveys estimates, grain prices, commercial and livestock particularly at regional balance sheets import and export farmers, grain levels and trigger level and focus biased parity prices silo owners, imports and toward crops and less on millers and exports livestock processors, and exporters/ importers Markets (cereal and To regularly collect Market surveys and Commodity/product National and Triggers imports No clear communication livestock) market information for analysis prices, agriculture regional levels and exports and between farmers’ unions, monitoring and decision input costs, and food destocking marketing authorities, CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity making price/trade monitoring millers, and traders Vulnerability (social, To provide information Vulnerability assessment Agricultural production, VAA mostly Triggers Harmonization of VAA economic, health, and on vulnerability to food and analysis and IPC climate/weather used in southern interventions to methodology, training on nutrition; food insecurity and a information, Africa while the prevent food the IPC, and reliance on availability, access, and broader range of socioeconomic, IPC is common insecurity and external resources utilization) vulnerabilities anthropometry, and in east Africa; famine (for market prices data users—national example, safety and regional net programs and agencies humanitarian aid) FIGURE 2.1 Characteristic Spatial Scales of Weather Phenomena PAST CLIMATE WEATHER FUTURE CLIMATE Global scale Medium-range weather forecasting Short-range weather forecasting Long-range weather forecasting 10 000 km Decadal climate prediction Climate change crojection Continental/ Seasonal to interannual regional climate prediction “Normal” climate Current weather Recent weather Recent climate scale Nowcasting 1 000 km Synoptic scale 100 km Mesoscale 10 km Local scale 1 km Last Last Last Last Next Next Next Next Next Next Yesterday Now Tomorrow decade year month week week month year decade century millennium Other data sources include methods focused on vulnerabilities that incorpo- rate socioeconomic information such as population density, poverty maps, access to markets, and commodity prices. Relevant datasets are usually obtained using generally accepted methodol- ogies from the social sciences for assessing targeted populations’ vulnerability and resilience to natural hazards. Altogether, these hazard and vulnerability assessment methods constitute the first step in the design of an effective food security information and EW system. Seasonal Climate and Weather Forecasting17 Methods Providing information and advice on the past, present, and future state of the atmosphere is a central role of the NMHSs, supported by global and regional climate forecasting and prediction centers. This includes information on temperature, rainfall, wind, cloudiness, and other atmospheric variables and their influence on weather- and climate-sensitive activities and communities. The physical phenomena responsible for the weather and climate conditions are manifested at particular spatial and temporal scales (figure 2.1), which have important implications on observability, predictability, and service design. The Global Data Processing and Forecasting System (GDPFS) of the World Meteorological Organization (WMO) supports the NMHSs and the regional climate centers mainly through global numerical weather prediction (box 2.1). The GDPFS prepares and makes available meteorological analyses 17 A forecast is a statement of expected meteorological (or hydrological) conditions for a specific period and for a specific area or portion of air space (WMO 1992). CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity 17 BOX 2.1 Technical Insight: Functions of the GDPFS The real-time functions of the GDPFS include: • Preprocessing of data (such as retrieval, quality control, decoding, and sorting of data stored in a database for use in preparing output products); • Preparing analyses of the three-dimensional structure of the atmosphere with up-to-global coverage; • Preparing forecast products (fields of basic and derived atmospheric parameters) with up-to- global coverage; • Preparing ensemble prediction system products; • Preparing specialized products (such as limited-area, very fine mesh short-, medium-, extended-, and long-range forecasts and tailored products for marine, aviation, environmen- tal quality monitoring, and other purposes); • Monitoring observational data quality; and • Postprocessing numerical weather prediction data using workstation and personal com- puter–based systems to produce tailored value-added products and to generate weather and climate forecasts directly from model output. The non-real-time functions of the GDPFS include: • Preparing special products for climate-related diagnosis (that is, 10-day or 30-day means, summaries, frequencies, and anomalies) on a global or regional scale; • Comparing analysis and forecast products; monitoring observational data quality; and ver- ifying the accuracy of prepared forecast fields, diagnostic studies, and numerical weather prediction model development; • Storing long-term Global Observing System data and GDPFS products, as well as verifying results for operational and research use; • Maintaining a continuously updated catalogue of data and products stored in the system; • Exchanging ad hoc information through distributed databases between GDPFS centers; and • Conducting workshops and seminars on the preparation and use of GDPFS output products. Source: Rogers and Tsirkunov 2013, 38. and forecasting products, which are generated at a few specialized centers (for example, Pretoria, Melbourne, Moscow, and Washington, DC). With the computing power and technical staff to run these models, in many centers, the models now run at such high spatial resolution (better than 15 km horizontal resolution) that they can be used directly, or downscaled, by the NMHSs in their own forecast production systems. Seasonal forecasting is well developed in ESA. The Greater Horn of Africa Climate Outlook Forum (GHACOF) in eastern Africa and the Southern African Regional Climate Outlook Forum (SARCOF) in southern Africa are among the renowned global institutions that provide seasonal climate ser- vices and EW information (see appendix A). GHACOF and SARCOF bring together representatives of national and regional meteorological services and 18 CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity users to construct a consensus forecast for the region each year. This involves expressing rainfall occurrence for the coming season as probabilities of below normal, normal, and above normal. The positive impact of both GHACOF and SARCOF on food security is recognized, particularly in informing agri- culture production, processing, and marketing (box 2.2). Through the sea- sonal forecasts, subsistence farmers in Zimbabwe are able to make decisions on the type of crop to plant, put in place some measures in dam manage- ment, and also plan for the malaria season. In Mozambique, the national con- tingency plan for floods, drought, cyclones, and thunderstorms is based on SARCOF, which is downscaled to the national context. Some of the challenges GHACOF and SARCOF face include: • Limited capacity to downscale seasonal forecasts to high resolution; • Seasonal climate forecasts information not packaged according to user needs; and • Resource constraints. First, providing forecast information that is specific to particular users’ needs, as shown in box 2.1, helps users to make appropriate decisions and take appropriate actions. Improvements are still required in the way the cli- mate information is packaged and delivered to rural people or nontechnical people, such as pastoral communities, who tend to rely on their traditional knowledge and systems of weather forecasting rather than that of the govern- ment (see box 2.2). Thus, there is still a need across most ESA countries to transform the technical information into actionable recommendations that are provided in a timely and a culturally sensitive manner to the targeted audi- ences. Reaching the targeted communities along the “last mile” of an end-to- end EWS shall be the goal of all relevant stakeholders. Effective “end-to-end” early warning systems include four interrelated key elements: (1) disaster risk knowledge based on the systematic collection of data and disaster risk assess- ments; (2) detection, monitoring, analysis, and forecasting of the hazards and possible consequences; (3) dissemination and communication, by an official source, of authoritative, timely, accurate, and actionable warnings and asso- ciated information on likelihood and impact; and (4) preparedness at all lev- els to respond to the warnings received. These four interrelated components need to be coordinated within and across sectors and multiple levels for the system to work effectively and to include a feedback mechanism for continu- ous improvement. Failure in one component or a lack of coordination across them could lead to the failure of the whole system. Second, while replicating GHACOF and SARCOF seasonal forecasts to national contexts has become institutionalized, there is limited technical capacity to downscaling the seasonal forecasts to local levels to support local decision making. However, even if the climate forecasts are downscaled to a high resolution there is no guarantee that the predictions will be more precise, because the climate forecasts are based on probabilities. Transforming techni- cal information, including making sense of what the probability of the materi- alization of an event really means shall be factored in in the communications CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity 19 BOX 2.2 ICPAC Recommends Early Action Following GHACOF’s Rainfall Outlook for March–May 2017 According to the March, April, and May (MAM) 2017 rainfall outlook, the average onset of rains over the equatorial sector was predicted for the second week of March and April 2017 while the southern sector was already in its seasonal rains. Several dry spells were to be experienced in the season. The mean cessation dates was forecast to be in the third and fourth week of May 2017. These could, however, change if tropical cyclones developed along the western Indian Ocean. The MAM 2017 rainfall forecast had implication for droughts, floods, and other associated haz- ards such as natural resource–based conflicts and disease outbreak for both human and livestock populations over the Greater Horn of Africa. The MAM 2017 seasonal forecast reinforced the existing risk and vulnerability that would lead to serious negative impacts if not attended to. The following mitigation measures were, therefore, recommended. Mitigation measures for Disaster Risk Management (DRM): • Strengthen regional and national coordination mechanisms. • Strengthen mitigation and response interventions. • Strengthen communication and advocacy campaign. Agriculture and Food Security Sector: • Ensure diversification of livelihoods. • Plant early maturing and drought-tolerant crops. • Maximize crop production during good rains to boost production volumes. • Increase agronomy and establish plantation farms. • Avoid planting crops in flood- and landslide-prone zones. Water Sector: • Ministries of water or water agencies to carry out work related to closing open river banks/ dykes and strengthening weak ones. • Reservoir management authorities encouraged to carry out effective reservoir management suitable for above-normal inflows. • Ministries of water or water agencies to intensify rainwater harvesting. For a zone that was forecast to receive near-normal to below-normal rainfall, it was advisable that • Ministries of water or water agencies intensify rainwater harvesting and identify and main- tain strategic borehole for pastoralists; • Reservoir management authorities carry out effective reservoir management suitable for below-normal inflows; • Municipal water management authorities take water conservation and demand management actions; and • DRM institutions and nongovernmental organizations (NGOs) carry out conflict manage- ment in known hot spots in this zone. Livestock Sector should focus on the following: • Desilting of water pans • Production and storage of fodder • Livestock vaccination • Restoration and reseeding of degraded lands • Awareness campaign • Livestock insurance Source: ICPAC 2017. 20 strategy when interacting with particular audiences, including government nontechnical authorities, and the members of local communities, who rely on their traditional knowledge to understand and adapt to their environment. Third, understaffing needs to be addressed, especially in the CSC so that it can sustain its mandate, including SARCOF. The stakeholders’ consulta- tion reveals that the CSC needs about 10 staff in administration, computing, data management, climate science and specialized applications, research and development, and generation and dissemination of products. Although the Monitoring for Environment and Security in Africa (MESA) project funded by the European Union (EU) has enhanced access to reliable, timely, and accurate land, marine, and climate data and information in ESA by providing high-speed computers, these would need to be complemented by data centers, with an appropriate data storage capacity. Besides challenges in downscaling climate forecasts to high resolution, most ESA countries also have limited capacity to merge and formulate weather data into final products, weak observation network density (table 2.2), lack of equipment to observe convective weather, such as Doppler radars, and weak human resource capac- ity to support operations. Although adequate instrumentation is one important component for effec- tive monitoring and forecasting, the technical and financial resources needed for operations and maintenance are critical for ensuring the sustainability of the investments in hydromet modernization. The majority of countries in ESA are operating far below the WMO’s rec- ommended minimum density of weather and hydro-observation stations shown in table 2.3. The need to expand the observational networks has increased the demand and preference for AWSs and telemetric gauging stations in the ESA coun- tries. However, most of the NMHS staff consulted as part of this study reiter- ated their concerns regarding existing technical and human capacity within their organizations, particularly the need for additional training, as well as taking into consideration the advantages and disadvantages of using AWSs in low capacity environments (see table 2.4). The Trans-African Hydro-Meteorological Observatory (TAHMO) proj- ect provides an opportunity to improve Sub-Saharan Africa’s capacity for hydrometeorological monitoring. In particular, the TAHMO project aims to build a dense network of hydrometeorological monitoring stations—one every 30 km—in Sub-Saharan Africa by installing 20,000 stations on school grounds. TAHMO is currently collaborating with the Kenya Meteorological Service (KMS) to develop a network of AWSs in schools in Kenya. To pro- mote the sustainability of the project, the use and maintenance of the weather stations will be integrated into the educational curriculum. These objectives align with the mission of KMS to • Facilitate accessible meteorological information and services; and • Infuse scientific knowledge to foster socioeconomic growth and development. CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity 21 TABLE 2.2 Hydrometeorological Observation Network Density 22 Country Technical capacity Enabling policy environment Financial capacity Ethiopia 1,200 meteorological stations; 2,000–5,000 Limited Limited government funding to cover costs; weather stations; 25 automatic weather stations charging for services is not permitted by law. (AWSs); upper air observation unit; and AWSs at airports Kenya 39 synoptic stations, 14 agrometeorological Mandate of Kenya Meteorological Department There is limited funding to cover costs; charging stations, 72 AWSs; 3 airport weather observation revised 2007 and enabled for services is required by law. systems; 17 hydrometeorological AWSs; about 1,000 rainfall stations; 1 upper air station; and 1 global atmospheric watch Madagascar 19 active stations operated by the Directorate- Decree No. 2002-803 of August 7, 2002, provides There is limited government funding to meet General for Meteorology (DGM) and 4 stations for the organization of the Ministry of Transport human resources, equipment, and transport operated by the Agency for Arial Navigation and Meteorology. expenses. Safety in Africa and Madagascar (ASECNA);a 250 more technicians required. Malawi 12 out of 28 districts need weather stations; 21 The enabling environment has limited integration There is limited government funding. main meteorological stations are spread across of meteorology in national development plans. the remaining 16 districts; 63 AWSs Mozambique 90 stations and needs 150 more stations; has 2 Regulation No. 6 of 2010 enables the Institute of There is limited government funding; charging for radars but not functioning effectively; needs 7 Meteorological Services to implement cost services, for example, aviation and private radars; 300 hydrometeorological networks need recovery measures. companies; lack of enforcement on timely upgrading to telemetric; lack of payments. agrometeorological professionals Tanzania 26 operational surface synoptic stations; 5 AWSs; There is a 5-year plan for the enhancement of Funding is primarily provided by the government, 16 operational agrometeorological stations; 60 meteorological services for sustainable, but the NMHS does derive nominal income from operational meteorological stations; 500 socioeconomic development in Tanzania. the aviation sector. operational weather stations; 1 upper air station; limited professional staff Zambia Zambia has a network of 108 weather stations (40 National Meteorology Policy (2013) and the The total budget of Zambia Meteorological CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity manual and 68 automatic). These weather stations Meteorological Bill (2015) allow sustainability of Department (ZMD) is ~US$5,000, with less than are concentrated in 45 out of 107 districts; lack of the NMHS through cost recovery on some 10 percent of this funding derived from NGOs. professional staff selected products and services. Zimbabwe 65 meteorological stations out of 150 required; 2 The Met Services Act 8 of 2003 enables the The Met Services Department has limited radars out of 4; >2,300 rain gauges needed provision of public services and cost recovery on government funding. some products. Note: a. ASECNA is an air traffic control agency based in Dakar, Senegal. It operates in the following 17 African countries: Benin, Burkina Faso, Cameroon, the Central African Republic, Chad, Comoros, Côte d’Ivoire, Equatorial Guinea, Gabon, Guinea-Bissau, Madagascar, Mali, Mauritania, Niger, the Republic of Congo, Senegal, and Togo. TABLE 2.3 WMO-Recommended Minimum Densities of Stations (Area in km2 per Station) Precipitation Water Physiographic unit Non-lending Recording Evaporation Streamflow Sediments quality Coastal 900 9,000 50,000 2,750 18,300 55,000 Mountains 250 2,500 50,000 1,000 6,700 20,000 Interior plains 575 5,750 5,000 1,875 12,500 37,500 Hilly/undulating 575 5,750 50,000 1,875 12,500 47,500 Small islands 25 250 50,000 300 2,000 6,000 Urban areas - 10–20 - - - - Polar/arid 10,000 100,000 100,000 20,000 200,000 200,000 Source: WMO 2012b. TABLE 2.4 Advantages and Disadvantages of AWSs Advantages Disadvantages Standardization of observations, both in time and There is a high initial cost of instrumentation and quality associated equipment and then ongoing costs of operation, such as for maintenance, electrical power, Greater reliability: real-time continuous measuring of communications, and security. parameters on a 24/7 basis It is not possible to observe all desirable parameters Improved accuracy (eliminates reading errors and automatically; at key locations, it may be necessary to subjectivity) augment automatic observations with a human Collection of data in a greater volume, for example, observer to obtain information such as cloud coverage 1-minute or 5-minute data as opposed to hourly or and cloud types. once per day If solar panels are used to power a station, this may Automatic adjustment of sampling intervals of different limit the amount and type of instrumentation, local parameters in response to changing weather events computing, and telecommunication equipment that Automatic quality control/quality assurance during can be used. collection and reporting stages, including automatic Final quality control is best carried out by a staff of alerts to users and maintenance personnel when trained operators working on a 24/7 basis. errors are detected The high volume of data generated requires the Automatic message generation and transmission, development of a data archival system that can be including alerts when critical thresholds are crossed costly and will require periodic forward migration as Automatic data archiving software changes. Access to data, both real-time and archived, locally or Routine preventive and as-required corrective remotely maintenance, together with periodic sensor calibrations, requires a staff of trained maintenance Collection of data in remote, harsh, or dangerous technicians. climates Source: World Bank 2015, 20–21. For these climate and weather forecasting methods to deliver timely, accu- rate, and reliable warnings, there is a need to invest in • Strengthening national capacity to downscale seasonal forecasts beyond national levels to the local levels because of limited equipment and tech- nical capacity; • Developing capacity to ensure the EWS products are packaged according to user needs; • Strengthening data collection systems and network coverage of the hydrometeorological observation stations; and • Strengthening data sharing arrangements because sectors still work in silos. CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity 23 It is worth noticing that although TAHMO has a great potential for making a substantial contribution to ongoing efforts for improving hydromet modern- ization across regions that are confronted with inadequate coverage of monitor- ing stations, there are also potential risks to the entire hydromet modernization enterprise at the national level. Particularly, there is the risk that, in the mid- and long terms, TAHMO and other similarly well-intentioned hydromet instrumentation efforts, financed by nongovernmental sources, could become another disincentive to the national governments for committing to provide adequate support to their own NMHS. Thus, an unintended consequence of such initiatives could potentially result in less government support, reflected in inadequate budget allocation, to their own specialized agencies. This situa- tion would, in turn, exacerbate the vicious cycle of underperforming NMHSs because of lack of adequate funding that leads to even less government sup- port, compelling governments to rely more on nongovernmental sources for hydromet information needs to support decision-making processes. In addi- tion, the proliferation of monitoring weather stations does not necessarily contribute toward ensuring their compatibility with national and international monitoring efforts (that is, it may not be compatible with WMO’s standards and procedures), which could undermine ongoing efforts for improving regional weather and climate monitoring and forecasting capabilities. Crop Forecasting and Monitoring Methods In most ESA countries, crop production estimates are well established in agriculture ministries and supported by the FAO’s GIEWS, where countries can access various tools and support (box 2.3). However, delays in regular crop production forecast18 updates render them less useful as an EWS tool. Ideally, observations and measurements of crops (on parameters including area planted, percentage of damage from pests, and weed infestations) should be made throughout the crop-growing season on a monthly basis. Some stakeholders who were interviewed attributed the weaknesses in producing timely updates of crop production forecasts to at least three chal- lenges. First, the paper-based data collection systems were deemed inefficient in providing real-time or near-real-time data for effective decision making. Although migrating to mobile technology-based systems is gaining momen- tum across the ESA countries, Internet costs to upload the data to the system are still prohibitive. Second, there are delays in conducting postharvest sur- veys; for example, in southern Africa, these are conducted in about June or July with the report available usually in August, which in most cases is late for decision making. Third, limited funding for field logistics means there is inadequate follow-up for data quality assurance. With respect to pest infestations and the FAW, there are gaps in surveillance systems, contingency plans, and management approaches (box 2.4). 18 Crop production forecasts fall into three categories: soon after planting (early forecasts), during the growing season (mid-season forecasts), and some weeks before harvesting (final forecasts). 24 CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity BOX 2.3 Technical Insights of GIEWS The FAO’s GIEWS provides information on countries facing food insecurity through various tools, including the following: Crop Prospects and Food Situation. This is published four times a year by the Trade and Markets Division of the FAO under GIEWS and focuses on developments affecting the food situation of developing countries and in particular the Low-Income Food-Deficit Countries. The report provides a review of the food situation by geographic region, a section dedicated to the Low- Income Food-Deficit Countries, and a list of countries requiring external assistance for food. It also includes a global cereal supply and demand overview to complement the biannual analysis in the Food Outlook publication. Food Outlook. This is a biannual publication (May/June and November/December) focusing on developments affecting global food and feed markets. It provides comprehensive assessments and forecasts on a commodity-by-commodity basis and  maintains a close synergy with Crop Prospects and Food Situation. The Food Price Monitoring and Analysis (FPMA) tool. Since 2009 when the FPMA tool went online, the database includes over 1,400 monthly domestic retail and/or wholesale price series of major foods consumed in 94 countries and weekly/monthly prices for 85 internationally traded foods. The FPMA tool provides easy access to the data, allowing users to do the following: quickly browse and analyze trends of single price series; create comparisons between countries/markets/ commodities; and download charts, data, and basic statistics such as maximum and minimum levels, averages, percentage changes, and standard deviations over different periods. Special Alerts. These short reports describe an alarming food security situation that is developing in countries or subregions. They also alert the international community on measures to be taken. Special Reports. These short reports describe the food supply and agricultural situation in coun- tries or subregions experiencing particular food supply difficulties. They also alert the interna- tional community on measures to be taken. Special Reports are often the result of Crop and Food Security Assessment Missions (CFSAMs) or rapid evaluation missions. Country Briefs. These provide up-to-date information on the food security situation of mon- itored countries, including information on the current agricultural season and the harvest prospects for the main staple food crops and livestock situation. In addition, the briefs provide estimates and forecasts of cereal production and imports together with food price and policy developments. Other topical information may be included when relevant. The briefs are updated no less than four times per year. Country Cereal Balance Sheet (CCBS). The CCBS system is a unique database created and con- tinuously kept up to date by GIEWS with data since 1980. It contains annual supply and utiliza- tion balances for the main cereals produced and consumed in all countries of the world. Earth Observation for Crop Monitoring. To support its analysis and supplement ground-based information, GIEWS uses remote sensing data that can provide a valuable insight on water avail- ability and vegetation health during cropping seasons. In addition to rainfall estimates and the Normalized Difference Vegetation Index (NDVI), GIEWS and FAO Climate and Environment Division have developed the Agricultural Stress Index, a quick-look indicator for early identifica- tion of agricultural areas probably affected by dry spells or drought in extreme cases. Source: FAO 2017. 25 BOX 2.4 FAW Monitoring, Impact Assessment, and the EWS FAW monitoring in some countries in Africa appears to be effective, especially where systematic field surveys have been happening. Specific gaps include the following: • Some countries do not yet have a monitoring system in place. • Systematic assessment of economic impacts (present and potential) of the FAW has yet to be undertaken, although many countries are planning to do so. • Almost all countries are currently responding to the outbreak, rather than pursuing EW. Contingency Planning and Awareness Creation about the FAW between Farming Communities Most of the countries that were affected by or at risk of the FAW have put in place contingency plans through their respective ministries of agriculture/national plant protection organizations, although some of these efforts are more advanced than others. Challenges include the following: • There is lack of financial and human resources to effectively implement the plans. • There is limited knowledge about the FAW control options to guide the farming communities. • There is lack of adequate and effective control options in the local markets. • Contingency plans are not always anchored in the national laws and regulations. • Awareness about contingency plans appears limited to a few institutions in some countries. • There is lack of regional and continental contingency plans to effectively counter transbound- ary/invasive insect-pests/pathogens. Development and Dissemination of the FAW Management Options The FAW management approaches in several countries affected by the pest were weak because of the following: • The FAW management approaches were limited to synthetic pesticides (especially organo- phosphates, synthetic pyrethroids, a few neonicotinoids, and in some cases cocktails of pes- ticides). • In most countries, the pesticide applications were mainly emergency responses, but not based on any efficacy evaluation (except in a few countries such as South Africa and Uganda). • In some countries, botanical pesticides such as Neem and Tephrosia were tested and were reported to be effective. In most countries, the use of pesticides in terms of the FAW control was not so effective. • As of April 2017, South Africa has fast-tracked registration of 15 synthetic pesticides and 4 biopesticides for FAW control. Some information/observations on the efficacy of cultural control options, such as handpicking (for example, Rwanda), early planting (in many coun- tries), and management of crop residues were also presented. Very few countries have iden- tified indigenous natural enemies against the FAW and have used biopesticides against the FAW. Except South Africa, no other country in Africa has the option of testing Bt maize (MON89034), which was reported as effective against the FAW in the United States. Source: Stakeholders Consultation Meeting on “Fall Armyworm in Africa: Status and Strategy for Effective Management,” April 27–28, 2017, Nairobi, Kenya 26 CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity Looking ahead, the following actions need consideration: • An effective EWS should be in place for a smooth and timely flow of infor- mation on the FAW outbreak, coupled with predictive modeling based on relevant factors. A scale of early warning (for example, alarm levels) needs to be designed (backed by quality monitoring data), informed with lessons derived from the U.S., Brazilian, and African armyworm and locust control. • Early warning/alerts to countries outside Africa: There is an immi- nent danger of the FAW moving to North Africa, and outside, including the Middle East, Europe, and Asia, either through the pest migration or trade. It is very important for these high-risk countries/continents to be alerted in time to put in place proper contingency plans. • Impact assessments: There needs to be an array of cost-effective, clear, and harmonized impact assessment protocols for the FAW outbreak, building on the existing pre- and post-survey and systems. The impact assessments must take into account physical and socioeconomic/liveli- hood factors, covering the broad spectrum of agro-ecologies and crop- ping systems Africa embeds. • However, the FAW EWS information should be conceptualized within the broad food security information system to leverage on the institu- tional, technical, and financial resources and preclude any risk of frag- mentation of efforts. Vulnerability and Capacity Assessments As the conceptualization of food security evolves, new metrics for measuring food security have also evolved. The conceptualization of food security has gone beyond “food availability at all times of adequate world food supplies” and “physical and economic access to basic food” (Sen 1981) to include “utili- zation” as well as the ability to acquire socially and culturally acceptable foods (Jones et al. 2013, 443). Emerging out of the shifts in food security concep- tualization are composite indicators, which are underpinned by sustainable approaches and participatory approaches (Chambers 1996; Scoones 1998). These are approaches that have emerged under vulnerability and capacity assessments and have increasingly gained currency in the ESA countries. The major actors and some of the common tools and approaches used for moni- toring vulnerability, including price and commodity, and cross-border trade monitoring are summarized in table 2.5. Evidence from the Regional Vulnerability Assessment and Analysis (RVAA) reports 2012–16 illustrates that most of the countries in southern Africa conduct National Vulnerability Assessment and Analysis (NVAA) annually (table 2.6). Informed by qualitative information derived from participatory tech- niques such as interviews, focus group discussions, and mapping, the VAAs allow for a contextual understanding of food security and livelihood CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity 27 TABLE 2.5 Actors in ESA Food Security EWS Name Organization setup Information format and frequency FEWS NET FEWS NET and network members Special reports and food security alerts (depending on severity), Food Assistance Outlook (monthly), price watch (monthly) in ESA Food Security Group of NGOs, United Nations FSNWG update on central and east Africa (approximately and Nutrition (UN) agencies, Red Cross/Red monthly) Working Crescent Movement, food security Group information systems; technically (FSNWG) supported by FAO IPC National and regional (at each Maps with current situation and trends (biannual) for level multiagency and national Kenya, Somalia, Uganda, and east Africa and special governments and forums). In east/ briefs/updates on Uganda (irregular), Somalia (quarterly); central Africa region, the IPC is for information on east Africa as a region, it supports the supported by the IPC steering FSNWG. committee of the FSNWG. GIEWS FAO Crop Prospects and Food Situation (trimestral), special reports (dependent on emerging crisis, often after CFSAMs), Food Outlook (biannual), and Global Food Price Monitor (monthly) HEWS Inter-Agency Standing Committee Maps on website, specific tools/reports and links to and WFP (WFP is responsible for organizations dealing with specific hazards; various the coordination/management of website tools related to hazards such as locust content.) infestation, flooding, weather, and storms; seasonal and hazards calendar including seasonal calendar with marked food security-related hazards VAA Led by the SADC and the SADC The VAA database provides a regional snapshot of the member states supported by the SADC humanitarian situation, food insecure population by Regional Inter-Agency Standing country, regional food balance sheet (RFBS), malnutrition Committee and NGOs rates, and socioeconomic context and recommends actions. East African The EAGC is mandated by the Through the Regional Agriculture Trade Intelligence Grain Council EAC to coordinate the private Network (RATIN), the EAGC provides updates on daily (EAGC) sector and looks at two systems on price information, monthly cross-border trade, real-time Market Information System (MIS). warehouse volumes, food balance sheets, trends and projections of trade, and bids and offers for market access. Source: Ververs 2012, 132 and authors. vulnerability in specific settings. In particular, the VAAs delineate geo- graphic patterns of shared livelihoods, groups of households based on wealth and assets, and categorize livelihood strategies, including coping strategies. The NVAA data are consolidated into a regional VAA database usually in June each year to provide a regional overview, and a bulletin (figure 2.2) is issued, which provides a snapshot of the SADC humanitarian situation, food insecure population by country, RFBS, malnutrition rates, socioeconomic context, and recommendations. The success of VAAs hinges on the following: • The VAAs are multisectoral surveys conducted annually, usually in May. • There are guidelines for the VAA processes. • The multisectoral approach helps build trust across both data producers and users. 28 CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity TABLE 2.6 NVAA Reports Consolidated into RVAA (2012–16) Year Country 2012 2013 2014 2015 2016 Angola — X — — X Botswana X X X X X Congo, Dem. Rep. — X X X X Lesotho X X X X X Madagascar — — — — X Malawi X X — X X Mauritius — — — — — Mozambique X X X X X Namibia X X X X X Seychelles — — — — — South Africa X — — — X Swaziland X X X X X Tanzania X X X X X Zambia X X X X X Zimbabwe X X X X X Total 10 11 9 10 13 FIGURE 2.2 Good Practice from the SADC RVAA System Source: SADC 2016b, 40. CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity 29 BOX 2.5 VAA Challenges The VAA reports are usually produced in June, which is late if the tool is to be effective for the EWS. Although the information is made available to users at the regional and national levels, the data are not made available to stakeholders at the subnational levels and it is not in the language that is understood by communities. Local people believe that VAAs are not a useful tool because they block them from accessing food aid. There is a need to harmonize the VAA methodologies or at least to agree on the minimum indicators. In some countries, the VAA reports do not recommend actions to be considered, suggesting that the report does not trigger action. In the majority of countries, VAAs rely on funding from partners. Source: Authors. TABLE 2.7 Progress in Adopting the IPC Tool in East Africa Year Country introduced Location of the IPC Burundi 2008 Ministry of Agriculture and Livestock Ethiopia — — Eritrea — — Djibouti 2011 Ministry of Agriculture Kenya 2007 National Drought Management Authority, supported by Kenya Food Security Steering Group Somalia 2004 Food Security and Nutrition Analysis Unit and governments of Somaliland and Puntland South 2007 National Bureau of Statistics chaired by the Ministry of Agriculture, Forestry, Sudan Cooperatives, and Rural Development Sudan 2007 Ministry of Agriculture and Irrigation through its Food Security Technical Secretariat Tanzania 2008 Ministry of Agriculture, Food Security, and Cooperatives Uganda 2007 Ministry of Agriculture, Animal Industry, and Fisheries Source: IPC 2017. • The governments in the SADC own VAA results and officials access them for public use.19 The VAA challenges are summarized in box 2.5. Developed by FAO in Somalia in 2004, and conceptually complementary to the VAAs, the IPC is used by almost all IGAD countries, except Ethiopia and Eritrea (table 2.7), to classify food security outcomes (table 2.8 and figure 2.3). The advantage of the IPC is that it provides guidelines for data collection and 19 In some countries, during the early ages of NVAAs, the results were often delayed, mainly caused by political sensitivity concerning food security. To address the political challenges, the VAC reports present plain facts, with analysis and judgments left to the users in a way that is neutral to avoid contentious issues. These descriptive reports are readily accepted by the governments. However, in terms of the EWS, such VAC reports provide important inputs to the risk knowledge component of such EWSs. However, VAC reports are not intended to be vehicles for providing recommendations on the specific actions to be taken. 30 CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity TABLE 2.8 IPC Phase Descriptions Priority response Phase Name Phase description objective Household (HH) group is able to meet essential Resilience building and disaster risk Phase 1 food and non-food needs without engaging in reduction. atypical, unsustainable strategies to access food Minimal and income, including any reliance on humanitarian assistance. Even with humanitarian assistance, HH group Disaster risk reduction, protection of livelihoods. Phase 2 has minimally adequate food consumption but Stressed is able to afford some essential non-food expenditures without engaging in irreversible coping strategies. Even with humanitarian assistance, HH group has food consumption gaps Protect livelihoods, reduce food consumption gaps and reduce Phase 3 with high or higher than usual acute Crisis malnutrition; OR HH group is marginally acute malnutrition. Food-insecure people (Phase 3 or higher) able to meet minimum food needs only with accelerated depletion of assets that will lead to food consumption gaps. Even with humanitarian assistance, Urgent action required HH group has large food consumption Save lives and livelihoods. gaps resulting in very high levels of acute Phase 4 Emergency malnutrition and excess mortality OR HH group has extreme loss of livelihood assets that will lead to large food consumption gaps in the short term. Even with humanitarian assistance, HH group has an extreme lack of food Prevent widespread death and Phase 5 Famine/ and/or basic needs even with full total collapse of livelihoods. Catastrophe employment of coping strategies. Starvation, death and destitution are evident. Source: FAO 2017. dissemination of information, and its results can be compared with other neigh- boring countries or regions and can identify action triggers. There are, however, some important considerations regarding the way in which the IPC protocols are implemented in different countries, attributable in part to differences in the manner in which some countries use specific food security outcome indicators. Other countries may make inferences based on proxies to compensate for data gaps. Another important consideration is the timing of the IPC analyses. Seasonal factors can affect the outcomes of the IPC analysis. Although the adoption of the IPC protocols continues to grow, several countries as well as regional organizations have yet to endorse them. Methods for Grain, Market, Cross-Border, Price, and Commodity Monitoring Grain, market, cross-border, price, and commodity monitoring are among the key components of an effective food security EWS. Monitoring these compo- nents helps to track food security conditions and trends at the regional, national, CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity 31 FIGURE 2.3 IPC Predicts Famine in Somalia (IPC Phase 5) Caseloads Trends (IPC Phase 3 & Above/Require Humanitarian Assistance) GHA food insecurity trends: Main source & validity of current Country IPC 3-5 (require humanitarian assistance, in millions) caseloads Sept 2016 Dec 2016 Feb 2017 May 2017 Djibouti 0.3 0.2 0.2 0.2 IPC November 2016 Ethiopia 9.7 5.6 5.6 7.8 DRMTWG, April 2017 Kenya 1.3 1.3 2.2 2.2 IPC/NDMA, Feb 2017 Somalia 1.1 1.3 2.9 3.2 Joint FSNAU-FEWSNET Somalia Food Security Alert, May 2017 S. Sudan 4.8 4.6 4.9 5.5 IPC May – June 2017 Projection Sudan 4.4 3.6 3.0 2.8 IPC April 2017 Update Uganda 0.4 0.4 1.6 1.6 IPC Jan – March 2017 Burundi 1.5 1.5 1.5 2.6 IPC April – May 2017 Tanzania - - 1.2 1.2 IPC Feb 2017 Total 23.5 18.5 23.1 27.1 Source: Compiled from FSNWG reports available at http://www.fao.org/disasterriskreduction/east-central- africa/fsnwg/en/. 32 CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity subnational, community, and household levels. If the monitoring finds unusual patterns or behaviors, this information becomes useful in forecasting food secu- rity conditions and makes recommendations on the expected type of responses. ProducƟon Monitoring: The NaƟonal Food Balance Sheet The national food balance sheet (NFBS) is a globally tried and tested method and primarily used to present a comprehensive picture of the pattern of a coun- try’s food supply during a specified reference period. In particular, the NFBS estimates the total domestic foodstuffs production added to the total quantity imported and adjusted to any change in stocks that may have occurred since the beginning of the reference period. This information is compared with the average national demand for consumption and other uses, such as feeding live- stock and seed, as well as losses during storage and transportation (FAO 2001; Jacobs and Sumner 2002). Consequently, the NFBS is one of the common tools used for food security EWSs across the ESA countries (Tefft, McGuire, and Maunder 2006). In southern Africa, the NFBSs are well established, institu- tionalized within government structures, and updated annually. Table 2.9 and figure 2.4 show the food production deficits for 13 SADC countries between 2011 and 2016, which informed the SADC’s appeal for humanitarian aid in June 2016 (SADC 2016a), following the El Niño-induced drought. Similarly, in east Africa, the EAC identifies the RFBS as a critical tool for enhancing intraregional trade because it provides policy makers with the data they need to make informed decisions on policies that affect regional food security. Supported by the East Africa Trade Hub and the five EAC countries (Burundi, Kenya, Rwanda, Tanzania, and Uganda), and through the minis- tries of agriculture, the National Food Balance Sheet Committees feed the data into the RFBS Portal. The EAGC is mandated by the EAC to coordinate the private sector and looks at two systems on MIS: • The RATIN, which monitors prices and volumes • The EAGC RFBS, which monitors actual food stock. Currently, the EAGC monitors 14 borders regionally and 41 markets focusing on six staples: wheat, maize, sorghum, beans, rice, and green grams. The Market Access Subgroup that includes FEWS NET, EAGC, WFP, and FAO contributes data from the informal arrangement and consolidates them into one quarterly report for monitoring food security. This is one of the tools used by relief agencies to apply for humanitarian assistance. Notwithstanding their invaluable contribution to the EWSs, the NFBSs face numerous challenges, including: • Delays in regular crop production forecasts (as stated) and • Lack of clear guidelines for agencies involved in developing the NFBSs. The Zambia Disaster Management and Mitigation Unit (DMMU) provides standard operating procedures (SOPs) on the steps that the responsible sec- tors need to take to produce the NFBS (table 2.10). CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity 33 34 TABLE 2.9 The SADC Regional Cereal Production (Tonnes) 2010–16 2015–16 vs 2015–16 vs 5 yr Country 2010–11 2011–12 2012–13 2013–14 2014–15 2015–16 2014–15 percent avg percent Angola 1,367,429 505,795 1,672,184 1,820,348 2,016,566 2,374,208 18 61 Botswana 61,796 52,607 33,756 260,000 90,317 5,610 −94 −94 Congo, Dem. Rep. 2,537,145 2,602,074 2,583,228 2,797,317 3,127,252 3,257,829 4 19 Lesotho 103,170 58,162 120,094 103,526 89,035 26,747 −70 −72 Madagascar 4,729,495 4,998,597 3,989,872 4,344,037 4,051,671 4,530,365 12 2 Malawi 3,895,181 3,623,924 3,639,866 3,978,123 3,001,730 2,531,703 −16 −30 Mozambique 2,934,591 3,715,000 2,371,190 2,509,788 2,845,000 2,388,806 −16 −17 Namibia 127,600 168,500 81,500 131,900 67,800 80,000 18 −31 South Africa 13,084,335 14,764,619 14,502,889 16,940,000 12,206,315 9,323,455 −24 −35 Swaziland 88,502 76,091 81,934 118,871 93,653 33,860 −64 −63 Tanzania 7,033,498 7,435,957 7,806,580 9,828,540 8,918,999 10,139,108 14 24 Zambia 3,367,182 3,195,355 2,890,045 3,643,877 2,898,054 2,943,807 2 −8 Zimbabwe 1,648,404 1,129,845 943,620 1,718,630 868,017 637,843 −27 −49 SADC 41,197,438 42,425,504 40,840,506 48,321,075 40,398,477 38,273,341 −5.3 −10.2 Source: SADC 2016b. CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity FIGURE 2.4 Cereal Deficits in the SADC, 2011–16 80% 60% 40% 20% 0% Cereal deficit a a C o r i e ia Sw ica nd ia a e C aw ca ol an qu bi bw th R D ib an ila fr ng D m SA as so al sw am bi –20% ba nz A az Za M ag A am Le ot h Ta m N ut ad B Zi oz So M M –40% –60% –80% –100% –120% 2015/16 vs 2014/15 average 2015/16 vs 5 year average Source: SADC 2016b. TABLE 2.10 Good Practice on Processes Leading to the NFBS Production in Zambia Activity Due date and responsible body Seasonal rainfall forecast By the end of September, forecasts are generated by the Zambia Meteorological Department (ZMD) and submitted to the DMMU and other relevant departments and authorities. Hydrological conditions By the end of every month, information on hydrological conditions is generated and submitted by the Department of Water Affairs and submitted to the DMMU and other relevant departments and authorities. Preliminary crop forecast By January 31 each year, the Ministry Responsible for Agriculture and Livestock (National Early Warning Unit), the agency responsible for national statistics, and the ZMD will generate the preliminary crop forecast report and submit to the DMMU and other relevant departments and authorities. Final crop forecast By April 30 each year, the National Early Warning Unit in the Ministry Responsible for Agriculture and Livestock, the agency responsible for national statistics, and the ZMD will generate the final crop forecast report and submit to the DMMU and other relevant departments and authorities. NFBS The NFBS will be published by May 15 each year. Comprehensive needs assessment By June 15 each year, the DMMU in collaboration with other multisectoral agencies will publish a report on the impact of drought and vulnerabilities. Source: DMMU 2015. CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity 35 Markets and Price Monitoring for Food Security Markets are critical for informing food security early warning analysis because they help shed light on two of the three elements of food security: availabil- ity and access. Markets will highlight signs of deterioration or improvement in food availability and access such as production shortfalls, nonseasonal increases in the prices of food and inputs, falling agricultural output prices, distress sales of livestock, and uncharacteristically early or large migration of people in search of casual employment. Most of the ESA countries have established marketing and price monitor- ing systems, most of them being housed in their Ministry of Agriculture’s economics, marketing, or agribusiness department. In addition to these departments, some countries, such as Botswana, Kenya, Malawi, Zimbabwe, and Zambia, have agriculture-marketing authorities, boards, centers, or cor- porations. Table 2.11 shows market analysis for Taita Taveta County in Kenya, which helps determine when prices are prohibitively high for some house- holds and/or when particular events or conditions prevent participants of market networks from responding by releasing stocks or moving commodi- ties from one location to another. Labor market conditions that result in low wages or insufficient labor opportunities also affect household food access (FEWS NET 2009). Although in most ESA countries commodity price monitoring is conducted on a monthly and weekly basis, in some countries there are no specific recom- mendations to help users take action. The market report in table 2.11 provides objective information that helps end users decide on the action to be taken. In contrast, while the example from Zimbabwe provides a list of prices for the week, it is not analytic and does not elicit action. However, providing infor- mation without analysis and recommendation is a deliberate process because this allows users to make their own analysis and decisions. In this way, the reports are neutral to avoid contentious political issues. There are also institutional challenges in commodity price monitoring that, according to study participants, tend to be more problematic than resource constraints. The liberalized nature of cereals marketing has brought a pau- city of players in market and commodity price monitoring. Although in the sampled ESA countries there are strategic grain reserves (SGRs), with some level of government control over its use, there is little information exchange between the players. In Zambia, there are at least five sets of players in com- modity and price monitoring data. The government collects data through the Agricultural Marketing Information Centre and shares with the Food Reserve Agency and the National Early Warning Unit, whereas the Zambia National Farmers’ Union and FEWS NET each collect their own data. The same applies in Zimbabwe where the Agriculture Marketing Authority shares data with the Grain Marketing Board (GMB), whereas the Zimbabwe Farmers’ Union (ZFU) and FEWS NET each produce their own data. Although the data are collected for different purposes, for example, the government and FEWS NET use the data for food security monitoring whereas the farmers tend to 36 CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity TABLE 2.11 Market Performance for February 2017 for Taita Taveta County, Kenya Cattle prices Goat prices The average price of a 3-year-old bull from 210 The average price of a 3-year-old goat from 210 sampled households decreased to K Sh 16,800 from K sampled households remained at K Sh 3,800, as in Sh 17,000 in the previous month. the previous month. The drop in price could be attributed to deteriorating The low price could be attributed to fair goat body cattle body condition. condition. Compared with the long-term mean, the average price Compared with long-term mean, the average price is is the same. higher by 6.1 percent. Livelihood variations were notable; in the horticulture/ No significant variations in prices of goats were dairy livelihood zone, cattle prices range from K Sh observed across livelihood zones. 25,000 to K Sh 40,000, while in hotspot areas, mainly in food crops/livestock livelihood zone, prices ranged from K Sh 8,000 to K Sh 10,000. Maize Beans Compared with the previous month, average price of Compared with the previous month, the average price maize per kg at household level remained at K Sh of beans per kg at the household level remained at K 41.80. Sh 91.80. Lowest prices ranged from K Sh 30 to K Sh 35 in the Lowest prices were recorded in mixed farming: mixed farming: irrigated cropping/livestock livelihood irrigated cropping/livestock/food crops livelihood zone, areas of Challa and Eldoro in Taveta subcounty. zone, Challa at K Sh 60 to K Sh 70 because of incoming beans from Tanzania, while high prices were Highest price was recorded in mixed farming: food recorded in mixed farming: food crops/livestock crops/livestock livelihood zone, Mwakajo, Rukanga, livelihood zone, Mwakajo, Mwachawaza, and and Mwachawaza at K Sh 40.00 to K Sh 45.00 in Voi Rukanga at K Sh 90 to K Sh 110. and Mwatate subcounties. Compared with the long-term mean, bean prices are Compared with the long-term mean, the price is higher almost equal. by 4.2 percent. Income Terms of trade (TOT) Analyzed income from 210 sampled households show TOT remained favorable from sampled households in that sale of charcoal, casual labor, and remittances the month under review. rose by 3 percent, 3 percent, and 1 percent The sale of 1 goat at K Sh 3,800 resulted in the respectively. purchase of 90.9 kg of maize at K Sh 41.80 per kg. Petty trading dropped by 4 percent while formal This is a drop compared with 95.12 kg posted in the employment, sale of livestock products, and sale of previous month. crops decreased by 1 percent each compared with the Though TOT remained the same as in the previous previous month. month, the trend is decreasing because of increasing Most households are now depending on charcoal prices of maize as a result of poor harvest and burning and remittances. decreasing prices of goat emanating from deteriorating body condition. Source: Compiled from Kenya National Drought Management Authority. focus on marketing their commodities, there is no real information exchange between the government departments and the stakeholders. In relation to the SGR, its primary function is to help a country cope with food emergencies and stabilize grain prices (table 2.12). The reserves can be in physical grain or financial reserves. For example, in addition to the buffer stock of 500,000 tonnes in Zimbabwe’s GMB, there is a cash reserve equiva- lent of 436,000 tonnes. The physical stock aims at meeting Zimbabwe’s food shortfalls for 3 months, and assuming the financial reserve equivalent is avail- able to import the grain, then grain will be available for another 3 months. Oftentimes, countries do not have adequate financial reserves to import food. Properly managed SGRs contribute to price stabilization. SGRs are replen- ished during times of normal or above-normal production to be used when CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity 37 TABLE 2.12 Role of the SGR in Grain and Price Monitoring Zimbabwe Zambia Kenya Malawi Agency GMB Food Reserve Agency National Cereal Agricultural (FRA) and Produce Development and Board Marketing Corporation (ADMAC) Functions Operates as parastatal Operates as a Parastatal with a Commercial and with a commercial and semiautonomous commercial and social roles social role body on commercial social role and social lines Strategic reserve 936,000 tonnes 500,000 million tons Stock of up to 8 100,000 million tons (500,000 physical to give at least three million bags for stock and 436,000 months of buffer stock food security and tonnes, backed by facilitates cash reserve of logistics for equivalent) famine relief food Pricing Premium prices for Premium price for Premium price for Premium price for maize maize maize—among maize, suggested by the highest in government Africa Price monitoring Weekly by ZFU, FEWS National Farmers’ Kenya National ADMAC, Ministry of NET, and Agriculture Union, Food Reserve Farmers’ Union Agriculture, FEWS Marketing Authority Agency, National and FEWS NET NET, and National Early Warning Unit, Small Holder Farmers and FEWS NET Association of Malawi Challenges Getting information Farmers expect FRA Information not Prices offered below from private sector as at premium price but easily accessible production costs; the GMB is a stocked maize from private challenges in competitor; it is difficult offloaded at sector obtaining real-time to get information from subsidized prices data because of the millers during times of food Internet costs; weak insecurity coordination markets are not able to meet the demand, increasing the risk of vulnerable households and communities to fall into a food insecurity situation. Under such conditions, tapping into the SGRs to meet the demand of struggling populations makes a lot of sense, and it is part of the government’s tools for addressing its contingent liabilities. Governments have the authority and the responsibility for intervening, even by distorting market prices, when there is an urgent need for protecting vulnerable populations. Governments, communities, and/or farmers are the main targeted audi- ence in the development of EWS information. To meet the expectations of each user group, it is best practice to have analysts with in-depth knowledge analyze the data and present the minimum of information needed, concisely, with narrative and in context. Additionally, the level of confidence in the pro- jections being made must be made clear. Collecting, compiling, and presenting data in tables and graphs isn’t enough. Data need to be analyzed, interpreted, and placed in context, with appropriate perspective provided. It is essential to know who the audiences are and what their perspectives and needs are to design information products that will be useful and used. Users should not be overwhelmed with data. The 38 CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity analyst should select the minimum set of data required to explain and sub- stantiate conclusions. Perspective should be provided that helps the user to understand whether the situation and trends observed are usual or represent anomalies. In this case, if the target audience is composed of farmers, they need weather forecasts that are precise but contain no technical language, are interpreted in local languages, and are presented in an easy-to-understand form, for example, scripts to be streamed as videos on television and broad- cast over FM radio. Policy and decision makers (government) need to understand why specific pieces of data are being presented to them, and why they are important. This information should be disaggregated geographically and illustrate condi- tions for populations typically vulnerable to a given disaster risk (FEWS NET 2009). Communities and farmers need to understand the extent of the anom- alies arising from the situation (in form of warnings or alerts) and the cor- responding sufficient coping capacities (responses) in relation to the locally affordable coping options—what ought to be done within their capacities to mitigate the disaster. Concisely, although it may seem self-evident, an EWS that monitors with- out forecasting and provides warnings or alerts without being linked to a well-thought-through plan and range of response options will not meet its objectives. Therefore, it is imperative for the analysts to link early warning to response and link information flows to decision-making processes. CHAPTER 2—Early Warning Methods, Technical Skills, and Capacity 39 CHAPTER 3 Performance of EWSs in East and Southern Africa Introduction The effectiveness of food security EWSs depends on the level of capabilities of national and regional bodies in terms of risk knowledge, monitoring and warning service capacity, dissemination and communication, and response capability. For these elements to produce a positive impact on food secu- rity EWSs, coordination of national and regional agencies, appropriate legal frameworks, clear roles and responsibilities, involvement of the private sector, and predictable funding mechanisms are critical. This chapter is guided by five key questions (themes): • Are systems that are regularly updated and accessible in place at national and regional levels for key hazard and vulnerability identification and analysis, supported by integrated maps for areas and communities that could be affected by natural hazards? • Does an effective national and regional hazard monitoring and warning service with a sound scientific and technological basis, language, and communication protocols exist? • Is there an established system for organizational and decision-making processes with a strong warning dissemination and communication chain, with the messages tailored to the specific needs of those at risk? • To what extent does the EW information inform preparedness and response plans, and what measures are in place for public awareness and education? • Are roles and responsibilities of various public and private sector EW stakeholders clarified and reflected in the national and regional regu- latory frameworks, planning, budgetary, coordination, and operational mechanisms? In response to these questions, this chapter is based on the questionnaire survey that was administered to 46 producers and users of food security EW information in 15 countries.20 Responses to thematic subjects under each theme were scored based on a hierarchy of perception on a 5-point Likert scale with 1 point awarded to Strongly Disagree and 5 points awarded to Strongly 20Responses to the questionnaires were further supplemented by consultation with at least 50 stakeholders in seven countries. CHAPTER 3—Performance of EWSs in East and Southern Africa 41 TABLE 3.1 Effectiveness of Risk Assessments at the National Level Frequency, percent Strongly Somewhat Strongly Subjects agree Agree disagree Disagree disagree Agencies have clear roles and responsibility in 32 53 5 5 5 risk assessments. National standards for risk assessments exist. 15 50 20 5 10 Hazards are regularly analyzed and evaluated. 24 47 14 10 5 Integrated risk and hazard maps are regularly 14 42 10 24 10 developed. Community vulnerability assessments are 10 24 28 28 10 conducted annually with disaggregated results. Affected communities and industry are consulted 10 29 19 32 10 in risk assessments. Risk assessment results are integrated in local 0 50 19 19 12 risk management plans. A central database for risk assessments exists. 0 33 9 29 29 Average 13 41 16 19 11 Source: Authors. Agree. The frequencies of perceptions of the thematic subjects21 on the 5-point scale were calculated by dividing the number of respondents with a score by the total number of respondents available and expressing them as a percentage. Performance of EWSs at the National Level EWSs and Risk Knowledge at the NaƟonal Level National risk assessments form the backbone of an effective EWS. Risk assessments require systematic, standardized collection and analysis of data, supported by legal and institutional frameworks that clarify roles and respon- sibilities for agencies involved in risk assessments, and the affected com- munities. Table 3.1 shows that risk assessments are regularly conducted in most ESA countries. About 85 percent of the respondents agree that agencies involved in risk assessments are identified and their roles and responsibilities clarified. About 65 percent felt the national standards for risk assessments exist, whereas about 71 percent opined that hazards were regularly evalu- ated. However, 66 percent disagreed that risk assessment results were disag- gregated to the social demographics, whereas the lack of involvement of the private sector and local communities in risk assessment received a score of 61 percent. Most participants (67 percent) also felt that a central database did not exist, yet such an information system is one of the critical elements of an effective EWS. 21 Thematic subjects refer to statements used to measure the five key questions. 42 CHAPTER 3—Performance of EWSs in East and Southern Africa Across all studied countries, respondents agreed that responsible agencies have clear roles and responsibilities in risk assessments; however, it was only in Rwanda, Djibouti, Sudan, and Swaziland where respondents were fully confident that national standards for risk assessment exist. With the excep- tion of Ethiopia (0 percent), Somalia (0 percent), and Zimbabwe (28.5 per- cent), other respondents confirmed that hazards are regularly analyzed and evaluated, with integrated risk and hazard maps developed as appropriate. Only respondents from Djibouti (100 percent) and Rwanda (100 percent) noted that vulnerability assessment maps were prepared annually with disag- gregated results. Consultation of affected communities and industries while conducting risk assessment is done sufficiently only in Djibouti (100 percent) and Swaziland (100 percent); however, it is only in Kenya (33.3 percent) and Sudan (100 percent) that risk assessment results are integrated in local risk management plans. With the exception of Somalia (0 percent), respondents in all other countries were confident of the existence of a national central risk assessment database. EW Monitoring and Warning Services at the NaƟonal Level Credible regional hazard and vulnerability monitoring, in addition to being supported by science and technology, also needs strong regional coordination in order to leverage limited resources and interagency capabilities. Table 3.2 summarizes the responses to the subject on the effectiveness of hazard mon- itoring and warning services. Hazard monitoring and warning systems at the national level are well established. At least 75 percent of the respondents felt that agencies with a hazard monitoring and warning remit were clearly iden- tified and their roles were clarified in legal and policy frameworks. However, although 75 percent of the responses indicate food security information sys- tems are in place in most countries, the field consultations established that these information systems tend to be fragmented along sector lines, which poses difficulties for users to access information. Even though an information system was established, it was noted that government departments were not accustomed to sharing their data with other departments. To address data-sharing challenges, respondents suggested the conve- nience of developing memoranda of understanding (MoUs) between the relevant agencies. Through establishing MoUs, it is possible to specify types of data, their format, and the frequency in which datasets should be shared. Key informants’ (KIs) opinions were split on their perception of topics regarding hazard monitoring and warning services provided at the national level. Agency roles and responsibilities in monitoring and issuing warnings are clear and well elaborated in Djibouti (100 percent), Sudan (100 percent), and Swaziland (100 percent), supported with binding agreements for con- sistency in warning language and communication channels. However, only respondents from Rwanda, Sudan, and Swaziland were fully aware of the existence of a national food security system. In Djibouti, Ethiopia, Kenya, CHAPTER 3—Performance of EWSs in East and Southern Africa 43 TABLE 3.2 Effectiveness of Monitoring and Warning Services at the National Level Frequency, percent Strongly Somewhat Strongly Subjects agree Agree disagree Disagree disagree Agencies have clear roles and responsibilities 20 55 10 10 5 in monitoring and issuing warnings. Agreements are in place for consistency in 15 15 35 15 20 warning language and communication channels. Food security information system is in place. 40 35 0 20 5 EWSs are subjected to regular tests. 15 25 20 30 10 Warnings are verified to check they have 5 30 20 30 15 reached recipients. Warning centers are staffed 24 hours. 0 25 10 50 15 Hazard measurement parameters and 10 30 20 35 5 specifications are regularly documented. Equipment for hazards and vulnerability 5 55 20 15 5 monitoring is suitable, and personnel trained in operation and maintenance (O&M). Hazards and food security data from regional 16 42 16 5 21 networks, adjacent territories, and international sources are accessible. Data are received and processed and 15 25 15 35 10 warnings disseminated timely, in meaningful formats, and in real or near-real time. Data analysis, prediction, and warning are 10 65 15 5 5 based on accepted scientific and technical methodologies. Average 14 37 16 23 11 Source: Authors. Rwanda, and Sudan, 100, 50, 60, 100, and 100 percent of the respondents, respectively, noted that EW systems are subjected to regular tests but only in Kenya (80 percent) and Sudan (100 percent) was verification done to ascer- tain whether warnings had reached intended recipients. In Kenya, Sudan, Swaziland, and Zimbabwe, 100, 50, 100, and 83.3 percent of the respon- dents, respectively, agreed that equipment for hazard and vulnerability monitoring is suitable, as is personnel trained in O&M. With the exception of Somalia and Rwanda, respondents from other countries noted that haz- ards and food security data from regional networks, adjacent territories, and international sources are accessible. It is only in Djibouti (100 percent), Kenya (80 percent), and Zimbabwe (50 percent) that respondents had con- fidence that data are received and processed and warnings are disseminated in timely, meaningful formats, in real or near-real time. With the exception of Djibouti and Somalia, respondents in other countries noted that data analysis, prediction, and warning are based on accepted scientific and tech- nical methodologies. Of concern is the absence of 24/7 warning services in most countries, with 75 percent of the respondents stating that warning centers were not 44 CHAPTER 3—Performance of EWSs in East and Southern Africa staffed for 24 hours; however, this is one of the important requirements of an effective EWS. With the exception of Rwanda and Swaziland where a 24/7 EWS service exists, in other countries, warning centers operate for 12 hours mainly because of limited staff and funds to pay for overtime. This is prob- ably because disaster profiles for Rwanda (MIDIMAR 2013) and Swaziland (Swaziland 2011) are dominated by many short lead time events such as fire, floods, earthquakes, landslides, hail storms with strong winds, lightning and thunderstorms, traffic accidents, diseases, and epidemics that disrupt peo- ple’s lives and livelihoods, destroy the infrastructure, and interrupt economic activities and retard development. Therefore, it is worth noting that food security EWS 24/7 services are most important for short lead time events such as lightning, storms, earthquakes, tropical cyclones, floods, pests, and diseases but not as critical for long lead time events such as droughts. For example, drought is a slow onset disaster, thus allow- ing time for food security reduction and mitigation measures to be undertaken. In some countries (for example, Malawi), the weather services operate only during regular working hours because of staff shortages. Although some coun- tries have emergency operation centers (EOCs), these do not operate 24/7 throughout the week and do not operate outside working hours and during holidays, unless there is an emergency. Nonetheless, several countries surveyed aspired to develop their EOCs along the lines of Mozambique’s national EOC, which received extremely favorable reviews during the stakeholders’ consulta- tion. Although these are likely to develop in slightly different ways depending on the context, there was a strong sense among the stakeholders interviewed that EOCs would have many functions beyond the activation of emergency response and they would serve as the EWS information hubs. If this suggestion is to come to fruition, then there will be a need to mobilize resources for the establishment of EOC infrastructure and technical expertise to support the EOC functions. EW InformaƟon DisseminaƟon and CommunicaƟon at the NaƟonal Level Effective EWS communication rests on the credibility and trustworthiness of the sender, format and wording of the warnings, dissemination methods, edu- cation and preparedness of stakeholders, and their understanding of the risks they face (WMO 2010). Almost all EW information providers will have, as a basic public task, the provision of forecast and warning services to the general public. EDS developers should ask themselves the following questions: • Is the EW information tailored to the needs of users? • How is this information communicated? • Is it communicated directly to the public by the EW information service provider through its own staff or through partner organizations such as emergency management agencies and the media? • Are online channels of communication (ranging from the traditional websites to social media) being used? CHAPTER 3—Performance of EWSs in East and Southern Africa 45 Except in Somalia, respondents from all other studied countries revealed that communication and dissemination of warnings are tailored to the needs of individual communities through multiple communication media. Interestingly, only in Kenya and Swaziland are warning alerts and messages tailored to the specific needs of those at risk. The understanding of the values, concerns, and interests of those who will need to take action is incorporated into warning messages in Sudan, Swaziland, and, to some extent, Kenya and Zimbabwe. Additionally, only in Djibouti, Kenya, Rwanda, Swaziland, and Zimbabwe are private sector resources used in dissemination of warnings. Only in Kenya and Ethiopia, 50 and 100 percent of the respondents, respec- tively, noted that consistent warning dissemination and communication systems are used for all hazards. In Ethiopia, Kenya, Sudan, and Zimbabwe, 100, 75, 100, and 33.3 percent of the respondents, respectively, noted that equipment maintenance is implemented with backup systems fully installed and  operational in the event of a failure. Specificity about the nature of threat and impacts is well elaborated in warnings issued in Kenya, Rwanda, Sudan, and Zimbabwe. Table 3.3 shows that in most countries in the ESA region, warnings are generally specific to the hazard, such as drought and floods. Most of the respondents (76 percent) were of the impression that warnings were specific to the hazard, whereas 64 percent believed multiple communication channels were used to disseminate the EW information. KIs confirmed that many users regard warnings to be credible and trustworthy because they are enforced through government channels and passed to affected communities. The major criticisms were the following: • Some 71 percent of participants felt the warning messages and alerts are still generic and not clearly tailored to the needs of the users to trigger action. • Key informant interviews (KIIs) revealed that the traditional forms of dis- seminating the EWS information (radio, telephone, sirens, visual warn- ings, drums, and messenger runners for some remote locations) are still dominant. The uneven coverage of these traditional forms of communi- cation, especially in remote areas, means that warnings are not received on time. However, the coverage in some countries has improved because of the RANET radios, for example, in Kenya and some parts of Zambia. • The use of social media outlets such as Facebook, Twitter, and WhatsApp is still in its infancy because of poor mobile network coverage as well as prohibitive Internet costs. Interestingly, these are the areas where the EWS information is needed most (see box 3.1). Part of the solution to increasing the use of social technologies in remote areas is to strengthen collaboration and partnerships between the public and private sectors. In many countries, there is increasing collaboration between telecommunications regulators, mobile phone providers, and the national disaster management authorities in the dissemination and communication 46 CHAPTER 3—Performance of EWSs in East and Southern Africa TABLE 3.3 Dissemination and Communication of EWS Information at the National Level Frequency, percent Strongly Somewhat Strongly Subjects agree Agree disagree Disagree disagree Communication and dissemination of warnings 6 41 23 12 18 are tailored to the needs of individual communities. Multiple communication media are used for 23 41 12 18 6 warning dissemination. Private sector resources are used in 6 18 29 41 6 disseminating warnings. Consistent warning dissemination and 6 35 35 18 6 communication systems are used for all hazards. Equipment maintenance is implemented and 0 17 41 18 24 backup systems are in place in the event of a failure. Warning alerts and messages are tailored to the 0 29 24 35 12 specific needs of those at risk. Messages incorporate the understanding of the 6 41 23 24 6 values, concerns, and interests of those who will need to take action. Warnings are specific about the nature of threat 6 70 12 12 0 and impacts. Average 7 37 25 22 10 of the EWS information. In Zimbabwe, for instance, the Department of Civil Protection and ECONET Wireless, one of the mobile phone providers, have developed partnerships in EWS information dissemination. Before the message is sent out to subscribers, the Department of Civil Protection and ECONET Wireless agree on the message to be disseminated. The role of the Department of Civil Protection is to craft the messages in three main lan- guages (English, Shona, and Ndebele), whereas the role of ECONET Wireless is to disseminate the message to the subscribers. Many stakeholders, however, strongly feel these public-private collaborations and partnerships are less developed in the food security EWS. Some EWS users in Mbeta Island would not use social media because of illiteracy issues. If EWS information is about generating action of users, then the EWS stakeholders should find appropriate means of communicat- ing, including the use of the language understood by users. Along these lines, many EWS users, ranging from those in Mbeta Island to top officials, crit- icized the EWS producers for communicating their products in a language that is not clearly and widely understood (box 3.2). EWS Response Planning at the NaƟonal Level Although the effectiveness of dissemination and communication of EWS information plays a critical role for users to take action, it is important to translate this information to mitigation and preparedness and response plans CHAPTER 3—Performance of EWSs in East and Southern Africa 47 BOX 3.1 “We’ve No Options but to Rely on Indigenous Knowledge” In Mbeta Island, Namakusi Village, people have observed climate changes over the years. The rains used to start in October, following which planting would commence. This is no longer the case. In 2016–17 rainy season, the rains commenced on December 15 instead of October. The farming season has shortened, and yields have declined as a result. The floods and water have become less, and fish stock has dwindled to unprecedented levels. Droughts have increased over the years. Animals are affected by lack of grass, and some natural vegetable species have disap- peared. Pests have increased because of high temperatures. Based on their indigenous knowledge, inhabitants can tell when the wet season is approach- ing. This includes abundant clouds, no winds in September or winds that are south-easterly, and whether there are tumbimbi birds, usually in October. They can also tell when an outbreak of pests is imminent. Although they recognize that indigenous knowledge can sometimes be misleading, they have no other options. Mbeta Island has poor TV and radio signals. They cannot watch TV or listen to the radio. However, they were not using social media (Facebook and WhatsApp) because they do not have the financial resources to buy data bundles. In any case, social media are for children who are literate and not for adults (who may be illiterate). When days are good and people receive radio signals, they access weather information. However, this information is not always correct; it is sometimes misleading. One time the radio informed that the wind would be blowing from north to south and the opposite happened, affecting the fishermen. Inhabitants also felt excluded from development programs compared with people on the main- land. Of primary importance for them were seeds and farming tools, which need to be delivered no later than August in preparation for the rainy season. Source: Focus group discussion. BOX 3.2 Seasonal Forecasts Not Packaged according to Language of Users Although the Meteorological Service Department (MSD) information is scientifically sound, it is not packaged in the language that is useful to farmers. When they say, “below normal,” “normal,” or “above normal,” what does this mean? These are vague phrases and so are not useful for farm- ers. The MSD should disseminate information in a language that is understandable to ordinary farmers. They should clarify climate change words such as El Niño, La Niña, and variability and relate these to the local setting. Using technical jargon that is not understood by farmers and drawing examples from China are not useful to Zimbabwean farmers. Information should be properly packaged to suit its users. Sending graphs or maps with colors from the MSD is not suit- able for local people. Something easy to understand such as dates when people will receive or not receive rain and how much rain will be received will be more useful. The information needs to be translated into the local language: Shona to Shona farmers, Ndebele to Ndebele farmers, and so on to cover all the 16 languages in Zimbabwe. If the MSD does not have resources to do so, farmers could translate the materials. In any case, the informa- tion disseminated by the MSD should be quality checked by farmers themselves. Source: Member, ZFU. 48 CHAPTER 3—Performance of EWSs in East and Southern Africa TABLE 3.4 Applying EWS Information in Response at the National Level Frequency, percent Strongly Somewhat Strongly Subjects agree Agree disagree Disagree disagree Hazard and vulnerability maps are used to 29 47 6 18 0 develop national emergency preparedness and response plans. Target communities respond effectively to 12 18 40 24 6 early warnings. EWS information is built in school/university 0 24 29 12 35 curricula. Regular public awareness/education 18 47 0 29 6 campaigns are conducted. Regular simulation exercises are undertaken 0 26 26 16 32 to test the effectiveness of the EWS systems. Average 12 32 20 20 16 because these strongly affect the livelihood options, safe behavior, and how best to avoid damage and loss to property. This requires four critical inputs: 1. Preparedness and response planning 2. Public education and awareness 3. Simulation exercises 4. EOC. In terms of preparedness and response planning, table 3.4 shows 76 percent of the respondents believe that the majority of the ESA countries have some kind of system for translating the EW information into preparedness and response plans, with contingency plans being the most popular. KI interview- ees, however, criticized the quality of contingency plans, stating that some of the contingency plans are not based on plausible scenarios that precisely identify vulnerable communities, livelihoods, critical infrastructure, and gaps in resources, capacities, and roles and responsibilities. Most worrying was that the national contingency plans were rarely informed by sector and sub- national plans, because the systems for these plans were either less developed or nonexistent. These criticisms call the attention to the following: • The importance of building sector and subnational capacity in developing contingency plans • The need to develop a generic contingency plan template to guide the ESA countries in developing multihazard and multisectoral contingency plans. This might require the ESA countries to consider their participa- tion in the African Risk Capacity (ARC),22 which has the technical exper- tise in contingency planning. 22The African Risk Capacity (ARC) was established as a Specialized Agency of the African Union (AU) to help member states improve their capacities to better plan, prepare, and respond to extreme weather events and natural disasters, therefore protecting the food security of their vulnerable populations. The ARC pro- vides member states access to disaster risk finance that can be deployed in times of natural disasters and CHAPTER 3—Performance of EWSs in East and Southern Africa 49 • The need to make contingency plans appealing and engaging reading, not voluminous sets of directives that are often barely understood by busy decision makers and therefore remain as “decorations” on shelves, often considered useless when the time comes to address evolving adverse situations. One way of realizing the value of these as a tool of contin- gency planning is through regular simulation exercises, involving all key stakeholders. With respect to public education and awareness, 65 percent of the par- ticipants felt that the public education programs were appropriate. KIs also stated that interest has grown in the education sector concerning the intro- duction of disaster education, particularly following the Hyogo Framework for Action 2005–15. At the higher education level, a substantial number of universities in the ESA countries have introduced disaster studies. This progress is partly attributed to the advocacy in building local disaster risk–related capacity by the Periperi U, a partnership of African universi- ties established in 2006. At least 12 universities in Ethiopia, Ghana, Kenya, Madagascar, Mozambique, South Africa, Tanzania, Uganda, Zambia, and Zimbabwe offer disaster studies. In contrast, the progress of introducing disaster education at primary and secondary education levels has been slower, although doing so has the potential of wider and timely dissemina- tion of EWS products, particularly considering children’s large appetite for the use of social technologies. With the exception of Djibouti, Ethiopia, and Somalia, in all other studied countries, development of national emergency preparedness and response plans is informed by hazard and vulnerability maps; across the board, target communities appropriately respond to issued early warnings. Surprisingly, EW information is built into school/university curricula only in Djibouti (100 percent), Kenya (50 percent), and Rwanda (100 percent), with regular pub- lic awareness/education campaigns being conducted only in Rwanda, Sudan, and Zimbabwe. Regular simulation exercises are undertaken to test the effec- tiveness of the EW systems in all studied countries except Somalia. In terms of public awareness, this is well established in most ESA coun- tries, although these tend to be biased toward rapid onset events such as floods and cyclones. Strengthening public awareness requires the develop- ment of broad-based public education strategies, involving multiple stake- holders across the public sector, private sector, NGOs, civil society, and cooperating partners. Some 74 percent of the respondents indicated that simulation exercises were regularly conducted to test the EWS. The added value of simulation exercises is their role in testing and validating the planning assumptions. Exercising extreme weather events. This financing, coupled with predefined contingency plans, enables governments to respond to affected households on time thereby preventing household loss of livelihoods and building resilience. The ARC provides a link between early warnings through its advanced satellite-based software, Africa Risk View, and contingency planning for early action with objective and predictable financing through its insurance payouts. 50 CHAPTER 3—Performance of EWSs in East and Southern Africa enhances the awareness of the roles and responsibilities of responders and tests SOPs, and action triggers and builds morale in responders. The recent armyworm outbreak in southern Africa has underscored the importance of simulation exercises to develop coherence in responders. In Zambia, the armyworm outbreak exposed weakness in preparedness and contingency planning at the national, subnational, and sector levels. The “bottlenecks” in information sharing, coordination, communication, and logistics could have been identified and possibly addressed during simulation exercises. The contribution of simulation exercises is their potential to creating demand for EWS information to meet the needs of the users. The simulation exercises are likely to influence the way information is packaged from the producer to the end user. If simulation exercises should become a constitutive element of the EWS in ESA, then the capacity for developing simulation exercises requires development. This will require developing and utilizing skills in universities, NGOs, the private sector, and cooperating partners. Although simulation exercises are fairly well developed in some ESA countries, these tend to focus on rapid onset events such as floods and cyclones (for example, in Mozambique) and dam failure (for example, Kariba Dam in the Kariba Gorge of the Zambezi river basin between Zambia and Zimbabwe). Simulation exercises are rare for slow onset events such as drought and insect infestations. For drought, for example, a simu- lation exercise involving responders, transporters, warehouse authorities, and distribution centers would reveal the gaps in the efficient distribution of humanitarian aid. EWS Governance Mechanisms at the NaƟonal Level Although the ESA countries have witnessed remarkable progress in EWSs since the adoption of the Hyogo Framework for Action in 2005, there are also glaring gaps and shortcomings that should be addressed. On the positive side, table 3.5 shows that 55 percent of the participants indicated that a legal or pol- icy framework that had an EWS element existed in their countries. Along these lines, 50 percent agreed that a national committee with an EWS remit existed, suggesting that in some countries, the institutional arrangements are fairly developed from the production of EWS information to activation of response plans. This was corroborated by data from sampled countries (Ethiopia, Kenya, Madagascar, Malawi, Mozambique, Zambia, and Zimbabwe), where key agencies involved in the EWS are clearly identified and their roles clarified in legal and policy frameworks. The most relevant sectors and agencies include meteorological services, hydrological services (water), agriculture, health, and food and nutrition. Most of these sectors have legal frameworks that outline their mandates on the EWS. However, the absence of an overarching legal or policy framework that brings together the EWS stakeholders means the EWSs in some countries are fragmented along sectoral lines. Only Ethiopia, Kenya, Rwanda, and Swaziland have a national all-hazards EW committee with a remit on food security that highlights economic CHAPTER 3—Performance of EWSs in East and Southern Africa 51 TABLE 3.5 Performance of EWS Governance and Investment at the National Level Frequency, percent Strongly Somewhat Strongly Subjects agree Agree disagree Disagree disagree National all-hazards EWS 20 30 30 5 15 committee is in place with a remit on food security. Economic benefits of EW are 24 18 18 34 6 highlighted to senior government and political leaders. EW is integrated into national 12 29 24 29 6 economic planning. Local decision making and 6 29 29 12 24 implementation of EWS is in place and resourced. Regional and cross-border 0 35 24 12 29 agreements are established to ensure EWS are integrated. Capacities of agencies are 6 29 18 29 18 assessed and capacity-building plans developed and resourced. Government funding 0 46 24 24 6 mechanisms for EWS are developed and institutionalized. PPPs are utilized for EWS system 6 29 24 29 12 development. Strategy is in place for obtaining, 5 25 35 30 5 reviewing, and disseminating data on vulnerabilities. EWS legal or policy framework 10 45 15 25 5 exists. Average 9 32 24 23 13 benefits of early warning to senior government and political leaders. However, it is only in Ethiopia, Kenya, and Rwanda that capacities of agencies are assessed with capacity-building plans developed and resourced. In addition to Ethiopia, Kenya, Rwanda, and Swaziland, Sudan has also integrated early warning information into national economic planning, and has supported and financed local decision making and implementation of early warning systems. In Somalia and Swaziland, all respondents noted that government funding mechanisms for EW systems are neither developed nor institution- alized. The majority of respondents in Kenya (75 percent), Rwanda (100 percent), and Sudan (100 percent) reported that regional and cross-border agreements are established to ensure early warning systems are integrated. Only Ethiopia, Rwanda, and Sudan have embraced public-private partner- ships in EW system development, with a strategy for obtaining, reviewing, and disseminating data on vulnerabilities only operational in Djibouti. With the exception of Rwanda, Somalia, and Zimbabwe, respondents from other studied countries were fully aware of the existence of the EW legal or policy framework. 52 CHAPTER 3—Performance of EWSs in East and Southern Africa FIGURE 3.1 Organizational Structure of the EWS in Zambia National disaster management council National disaster technical committee Health and Water and Relief and Environment, Training and Early Finance and Infrastructure Agriculture Security nutrition sanitation logistics DRR and public warning tender sub- sub- sub- sub- sub- sub- sub- climate education sub- systems sub- committee committee committee committee committee committee committee change committee committee Disaster management and mitigation unit Provincial Provincial Parliamentary disaster disaster committee for management management DDR & committees climate coordinators change Private sector, District disaster District NGOs, management disaster UN system management House of coordinators media and chiefs on committees DDR & religious climate Satellite organizations change disaster management committees Source: ZMD. Even without overarching frameworks, many ESA countries have coor- dination systems for information sharing. These are mainly part of the national DRM platforms. In some countries, the national platform meet- ings are normally chaired by senior government officials, and in other cases these are cochaired by the senior government officials and the UN Resident Coordinator. Further, some countries have established national EW commit- tees or working groups, which also report at the national platform meetings (figure 3.1). However, some countries, such as Malawi and Zimbabwe, do not have a dedicated EWS working group or committee similar to the Zambian structure. This study found that strong subnational structures are one of the prereq- uisites of an effective EWS. This recognition has found currency in most ESA countries. Many countries are strengthening the subnational structures at the provincial, district, and community levels. Depending on the legal frame- works, these structures take various forms—some of the EWS functions are assigned to disaster risk reduction (DRR) committees and ward development committees, whereas others are sector specific such as food and security committees, water users committees, river catchment councils, and environ- mental committees. In Ethiopia, there is a clear EWS structure from the fed- eral level to the kebele (ward) under the National Disaster Risk Management Commission. The kebeles, for example, are actively involved in the National Productive Safety Net program where they use the products from the EWS, CHAPTER 3—Performance of EWSs in East and Southern Africa 53 for example, in targeting members of their communities eligible to participate in public works programs. Although many countries recognize the role of subnational structures in the EWS, because these are outlined in their pol- icy frameworks, these structures do not exist in practice, mainly because of resources constraints. Uganda has the National Disaster Preparedness and Management Policy, 2011, which talks about collection of EW information by the mandated institutions and dissemination by the National Platform for Disaster Risk Reduction and the media. The policy mentions other activities such as con- ducting risk mapping and vulnerability assessments, research, and documen- tation as a mode of obtaining EW information and developing EWS. However, there is no clear, established channel through which the collected informa- tion should continue to flow to the coordination center (National Emergency Coordination and Operations Centre [NECOC]), in the Department of Relief, Disaster Preparedness and Management. The policy also stipulates that the Uganda Broadcasting Council and Uganda Telecommunications Commission will establish memoranda of understanding (MoUs) with FM radio stations and mobile phone telecommunications companies to enable the use of their facilities to send out early warning messages whenever the need arises. Currently, the electronic media and FM radio system offer free airtime in the form of talk shows to disseminate EW information upon release of weather forecasts, for example, El Niño and La Niña events, and any other events that the department finds necessary to communicate to the public ahead of time. The Department of Relief, Disaster Preparedness and Management has started the process of drafting the National Disaster Preparedness and Management Bill; therefore, the legal framework should clearly stipulate how the responsible institutions shall ensure the flow of EW information on a reg- ular basis to NECOC (figure 3.2). In addition, the legal framework needs to task the telecommunication com- panies with sending out early warning messages as mentioned in the NDPM policy. The bill should assign these companies roles as private sector contrib- utors to disaster preparedness and management. Involving the private sector in disaster risk reduction is emphasized by the Sendai Framework for Disaster Risk Reduction 2015–30. The policy assigns NECOC a coordination role and technical (modeling) in addition to the responsible institutions. However, capacity building of NECOC staff to scientifically develop and analyze the provided EW informa- tion is still low. Some 65 percent of respondents felt that the EWS activities were not ade- quately resourced. Evidence from consultation with stakeholders and from the literature (for example, Tefft, McGuire, and Maunder 2006) shows that efforts to develop capacities at the subnational level have generally been externally driven, with tremendous challenges of sustainability. The 1990s Regional Early Warning System in the SADC countries is a case in point, where the benefits of the program were not financially sustainable when the support from the Danish government and FAO ceased. Several agencies, NGOs, International 54 CHAPTER 3—Performance of EWSs in East and Southern Africa FIGURE 3.2 Uganda National Integrated Early Warning System Information Flow Source: Authors. Federation of Red Cross and Red Crescent Societies (IFRC), UN agencies, and the World Bank have been involved in the EWS capacity-building programs in different forms. In Somalia, understandably because of the civil conflict there, food security assessments are mainly conducted by international agencies, led by FAO with minimal participation from the government (box 3.3). However, even in stable environments, the external agencies still dominate the EWS activities. In the rural district of Binga and Nyaminyami in Zimbabwe, a huge amount of donor resources toward strengthening decentralized systems has been provided since Zimbabwe’s independence from Britain in 1980; today, these districts are still dependent on humanitarian assistance. Similarly, in CHAPTER 3—Performance of EWSs in East and Southern Africa 55 BOX 3.3 Risk Assessments Led by International Agencies Currently, NGOs that have the means, knowledge, and resources are very much involved in assessing hazards and vulnerability for food security, and the findings obtained from these assess- ments are set only to benefit the NGOs that created this system and not the communities and the  government. On the other hand, the  government  agencies’ staff cannot assume their roles even though their roles on the [EWS] subject are clearly identified, and the reason is because they do not have the means, knowledge, and the resources to achieve these objectives. There are some hazard and vulnerability assessments data collected by government agencies but they are collected in an unconventional way. Source: Survey questionnaire. Kenya’s arid districts, such as Turkana, the food security challenges remain after protracted donor support. Indeed, there are 18 pilot projects that are being conducted by the World Bank’s Pilot Program for Climate Resilience, including in Zambia and Mozambique. In this project, the staff costs are borne by the government whereas the operating costs are borne by the World Bank. When the fund- ing ends, it is conceivable that the benefits accruing from these interventions are unlikely to be sustained. The reasons for this, as this study found, are the following: • About 60 percent of stakeholders felt that senior governmental and polit- ical leaders are not aware of the economic benefits of the EWS because cost-benefit analyses of previous disasters are rarely conducted; therefore, there is limited “buy-in” from the decision makers. Not only is better evi- dence of the cost-benefit analysis of the EWS needed in supporting deci- sion making but also the method of delivering these benefits should be communicated in the language these stakeholders will understand. • The EWS is viewed as an emergency and seasonal activity, with about 59 percent of the participants agreeing that the EWSs were rarely inte- grated into national economic planning. This makes the EWS an ad hoc activity, relying on emergency budgets, which often creates competition between the EWS and the regular development agenda. Some people interviewed in this study strongly advocated for a shift from viewing the EWS as an ex post event. Rather, the EWS should be viewed as an ex ante process that is mainstreamed in development frameworks. • There are limited alternative sources of funding. Some 65 percent of the participants felt that the capacities of the agencies involved in the EWS were not well resourced. At the same time, 69 percent of the participants felt that the PPPs were underutilized in EWS development. As the results of this study attest, there is also limited engagement in PPPs to help mobi- lize resources internally. This means the legal frameworks should be clear on how these partnerships could yield maximum EWS benefits. 56 CHAPTER 3—Performance of EWSs in East and Southern Africa Across Uganda, significant levels of investment into systems that collect and share early warning information on a range of hazards including floods, droughts, and diseases exist. The state of early warning systems in the country was analyzed in terms of organization, kind of early warning information delivered, dissemination channels, and technical and financial sustainability. According to Lumbrosso (2016), EWSs in Uganda can be categorized as the following: 1. Drought and Food security related early warning systems • Karamoja Drought Early Warning System (DEWS). Established in 2008, this system collects and analyzes data on 31 indicators to monitor hazards and vulnerability that fall into five main categories, namely, crops, livestock, human health, water, and livelihood. Every month, data are taken from a representative sample of 10 households per parish (75 parishes in total are sampled), recorded, analyzed by district departmental heads, and communicated as a monthly drought bulletin through a variety of means, including e-mails to decision makers and to communities through radio programs, dra- mas through role plays, and mobile phones. The bulletins provide four drought risk classifications, namely, normal, alert, alarm, and emergency. The system is largely implemented by the district local government with oversight from line ministries and financially by a nongovernmental organization ACTED, supported by FAO. DEWS appears technically sustainable in that it has a track record of using district-level and community-based actors to collect information. Although an exit strategy was developed in 2012 aimed at a phased handover to the government both in terms of the financing of DEWS and independent technical management of the activities, this has not been implemented and donor funding has come to an end. • Famine Early Warning Systems Network (FEWS NET). Established by the U.S. Agency for International Development (USAID) in 1985, FEWS NET helps decision makers plan for humanitarian crises and provides evidence-based analysis through monitoring of hazards that may have an impact on food security (for example, droughts, floods, price shocks, livestock epidemics). This is integrated with informa- tion and data on markets and trade, nutrition, livestock and crop pro- duction, and livelihoods to evaluate current and future food security conditions. The indicators used include TOT, satellite rainfall esti- mates, Normalized Difference Vegetation Index (NDVI), price data, nutrition indexes, monthly price data of staple foods, livestock, and livelihood commodities such as firewood, charcoal, and wage labor. FEWS NET uses the Integrated Food Security Phase Classification (IPC) scale to determine food insecurity. FEWS NET relies on a network of partners to access the data needed including market prices of staples in Karamoja from the World Food Programme (WFP) and Farmgain Africa for other CHAPTER 3—Performance of EWSs in East and Southern Africa 57 major markets in Uganda. FEWS NET uses seasonal rainfall fore- casts provided by the Uganda National Meteorological Authority (UNMA), as well as forecasts by the European Centre for Medium Range Weather Forecasts (ECMWF) and the National Oceanic and Atmospheric Administration. UNMA forecasts are accessed after the meetings of the Greater Horn of Africa Climate Outlook Forum (GHACOF). Sometimes there are updates on the rainfall forecast between GHACOF meetings. Based on the seasonal rainfall forecasts and by asking what people have planted and how it has germinated, FEWS NET develops general predictions regarding crop perfor- mance. These are validated in the field. A bulletin on food security and vulnerability in Uganda is issued quarterly and is shared with policy makers. However, the results produced by FEWS NET do not appear to be widely used by the Ugandan government. FEWS NET has a commitment to deliver monthly reports to USAID and hence cannot always wait for the offi- cial approval of the Ugandan government before issuing bulletins. FEWS NET is wholly funded by USAID based on a 5-year funding cycle. Its reliance on donor funding could affect its long-term sustain- ability because any change in donor funding priorities would result in its downscaling or in termination of the early warning activities. • World Food Programme (WFP) Vulnerability Analysis and Monitoring (VAM). The WFP on a regular basis, either as stand- alone WFP activities or in partnership with other agencies such as FEWS NET, collects and analyzes food security, market, and nutrition data. Every 6 months in partnership with UNICEF, WFP conducts a food security and nutrition assessment for Karamoja that collects data on several regional indicators. On a monthly basis, WFP collects food price data on various commodities in major markets across the country, such as maize, beans, rice, goats, and sorghum, and issues a monthly bulletin showing price trends. These bulletins are distributed by e-mail to policy makers, central and district-level governments, and food security and agricultural livelihoods cluster members and donors, and is downloadable from the WFP’s Vulnerability Assessment and Monitoring (VAM) website. The WFP’s VAM uses staff at a dis- trict level. From a technical point of view, the methods used to collect the data are technically sustainable. However, without support from the WFP, these data would not have been collected. • Disaster Risk Financing (DRF). This is established under the sup- port of Third Northern Uganda Social Action Fund Project (NUSAF III) with a loan from the World Bank. The Department of Relief, Disaster Preparedness and Management through NECOC imple- ments a disaster risk financing subcomponent in Karamoja by col- lecting, analyzing, and storing risk-related information. The primary indicator for whether DRF should be triggered within Karamoja is based on NDVI, which is freely available from various remote sensing 58 CHAPTER 3—Performance of EWSs in East and Southern Africa sources online. A monthly NDVI analysis is performed during the crop growing season and findings shared with the members of the DRF subcommittee. The secondary indicator is Integrated Food Security Phase Classification (IPC), which uses information collected by other EWSs such as DEWS, FEWS NET, and WFP (for example, food con- sumption–related indexes, coping strategies, the number of children admitted to feeding centers, staple food prices, terms of trade [TOT]). A DRF subcommittee was formed in December 2015, and com- prises organizations whose activities contribute to the country’s food security. The organizations include Office of the Prime Minister (OPM); FAO; WFP; FEWS NET; Ministry of Health; Ministry of Water and the Environment; ACTED; United Nations Development Programme (UNDP); UNMA; Makerere University; Ministry of Agriculture, Animal Industry and Fisheries (MAAIF); and the Uganda Bureau of Statistics. The major function of the DRF subcom- mittee is to approve the findings in the DRF report upon compilation of trigger indicators, and it is also another platform for information dissemination on Karamoja food security status. However, currently there is no established information flow from the local level to the national level. The EWS is still reliant on under- taking fieldwork by the national technical team. In addition, DRF does not encompass forecasts. Drought forecasts provide a longer lead time in which to implement public works programs in areas that are predicted to be affected by droughts. It is worth noting that the above-mentioned drought EWSs are implemented in one region of the country (Karamoja region), yet with climate change the risk has changed both in occurrence and magnitude in the cattle corridor that extends from the eastern and central to the southwestern areas of the country. 2. Flood early warning systems At the national scale, no flood forecasting and warning system exits, and there is no flood forecasting and warning system specifically for the majority of areas at risk of flooding. The Department of Relief, Disaster Preparedness and Management concluded hazard, risk and vulnerability profiling at the district level and these indicated areas at risk of flood- ing. Subsequently, studies (Goretti 2013; Lumbroso 2016) have been con- ducted, with limited mapping and identification of the areas at flood risk. The following includes some of the localized initiatives that have been implemented: • Butaleja District Flood Early Warning System, established in 2014 on River Manafwa as a joint venture between the government of Uganda and UN International Telecommunication Union to pro- vide communities with sufficient time to move to areas outside the flood plains when high water levels are detected upstream. It should be noted that the system does not provide forecasts for flood levels, CHAPTER 3—Performance of EWSs in East and Southern Africa 59 meaning that it does not provide a long lead time to allow for for- mulation of appropriate decisions with a bearing on improving food security. Once the water reaches a certain level on the sensors, a sig- nal is sent to the central command center to activate a siren that can be heard in a 5 km radius. The siren gives a message in both English and the local language to warn communities about a possible flood and whether they need to evacuate to higher grounds. The EWS is relatively simple and would appear to be technically sustainable. However, no budgetary provisions have been made at a district level to cover any maintenance issues that may arise. • Pilot forecast-based financing scheme for floods for northeast Uganda. This has been piloted by the IFRC Climate Center, the German Red Cross and the Uganda Red Cross Society focusing on 16 villages in Abim, Katakwi, Kotido, and Soroti (Jongman et al. 2016). The pilot relies on the Global Flood Awareness System (GLOFAS) run by the European Centre for Medium Range Weather Forecasts (ECMWF), which uses probabilistic forecasts of relevant variables (rainfall, temperature) with a simple hydrological and hydraulic model to produce probabilistic forecasts for 40 days in advance. The GLOFAS is set to produce warnings based on the probability of exceeding certain probabilities of flow-return periods of 1 in 2, 1 in 5, and 1 in 20 years (Jongman 2016). To improve its accuracy, the model is validated with phone Short Message Service (SMS) text messages and this informed the decision by the German Red Cross to enter into partnership with Ureport—an SMS-based communication plat- form launched by UNICEF in 2010. Because the GLOFAS model uses a forecast of the flood hazard, it could be a useful tool to implement low-regret actions in flood-risk areas. The GLOFAS website is freely accessible; however, from a technical sustainability point of view, its accuracy is low. It would also require the Ministry of Water and Environment to set aside sufficient bud- get to allow its members of staff to assist the ECMWF with improv- ing its forecasting accuracy for large rivers in Uganda. Although the IGAD/ICPAC recently rolled out flood forecasting models such as GeoCLIM to member states, these are still based at NECOC await- ing training of technical experts from line institutions and districts. Thus, this will take some time because it requires financing of short trainings. 3. Weather forecasts and warnings for farmers Approximately 80 percent of Ugandans depend on agriculture for income and food security; hence, any threat to agricultural production degrades Uganda’s socioeconomic status and puts a huge population at risk of poverty and hunger (Lumbroso 2016). In many low-income countries effectively disseminating weather forecasts and warnings to farmers is challenging and meteorological authorities often generate warnings solely in English using technical language. 60 CHAPTER 3—Performance of EWSs in East and Southern Africa The following details two weather forecasting and warning sys- tems aimed at farmers in Uganda: • Africa Climate Change Resilience Alliance (ACCRA) weather fore- cast and warning system. Under this ACCRA system run by UNMA, at the beginning of each season, a half-day meeting composed of key sector actors is held, the output of which is a simplified forecast for each Uganda district (ACCRA 2014). The forecast from GHACOF is translated into local languages (so far, 22 out of 54 locally spoken lan- guages in Uganda) and the translations are produced on prerecorded audio CDs, which are then sent to radio stations and posted on the UNMA website as audio broadcasts (Lumbroso 2016). It is worth noting the system is one of the few in Uganda that uses forecasts and generates warnings aimed at a specific user group. Because it uses regional forecasts from GHACOF, it should be technically sustain- able. However, because ACCRA is an NGO, it is not clear whether budgetary provisions for funding have been made within the relevant government department. • Mobile 3-2-1 service. This is being developed based on lessons learned in Malawi and Madagascar by Human Network International (HNI), an international development organization specializing in technology dissemination in partnership with Airtel Uganda, a tele- communication company. HNI is currently working with Airtel on a free SMS-based information and warning system covering only Airtel subscribers. The system will be owned and operated by Airtel in the districts bordering Lake Victoria and along the cattle corridor. The service provides users with 10 free SMS messages per month con- taining information on microfinancing, health, and water; however, in the future, it will provide information on weather warnings; there- fore, it has the potential to raise sizable awareness on risks posed by weather, including risks to agriculture. The system is one of the few primarily reliant on a private sector organization. However, pilot pro- grams in Malawi and Madagascar have shown that it has the potential to be both financially and technically sustainable. 4. Disease warning tools There are a number of tools in Uganda that are used or are being piloted for the surveillance of livestock diseases. It worth noting that these tools are not strictly EWS. These are briefly described in the following: • Pictorial Evaluation Tool (PET). This is a pilot phase in Karamoja as an objective, ground-based system to provide objective means of rap- idly of assessing crop yields and livestock conditions by comparing observations in the field with photo-indicators of actual crops and animals. Training for district and MAAIF officers by FAO on use of the tools is currently under way. • EMPRES-i event mobile application. Commissioned by the Irish government and FAO in 2013, it was piloted in 10 districts as a smartphone-based tool to report animal disease outbreaks. It allows CHAPTER 3—Performance of EWSs in East and Southern Africa 61 for workers to use smartphones to communicate disease outbreaks to the Uganda National Animal Disease Diagnostic and Epidemiology Centre. Although it was found to be useful, it is unclear whether it continues to be used and how sustainable it is from a financial and technical point of view. 5. MAAIF Early Warning System The Planning Unit at the Ministry of Agriculture, Animal Industries and Fisheries (MAAIF) issues warnings after analyzing implications of UNMA season forecasts on agriculture. The warnings are accompanied by advisory messages and mitigation measures for crops, livestock, and fisheries. Within each growing season crop, livestock and fish monitoring takes place. A postharvest assessment is also carried out. The IPC working group composed of MAAIF, FAO, WFP, Makerere University, and other key actors handles data analysis at the national level. It is worth noting that access to adequate data, especially for the Karamoja, remains a challenge. Additionally, schedules for regular mon- itoring of livestock, fish, and crops are nonexistent. Notably, triggers for responses have never been set. From the reports developed, there is one specifically focusing on Karamoja because it is the most vulnerable region in the country. The information generated is disseminated as advisories twice a year through newspapers and e-mails to local government, as well as via local radios. Although the MAAIF EWS uses seasonal climate forecasts from UNMA just like most other EWSs in Uganda, it does not forecast the risk. Additionally, it appears underresourced and lacks a feedback mechanism, which makes it difficult to ascertain how effective it is in reaching vulner- able communities and engendering a response. 6. NECOC Disaster Monitoring System NECOC Disaster Monitoring System, developed with support from UNICEF, is a community-led alert system managed by the Department of Disaster Preparedness and Office of Prime Minister (OPM). It is a coun- trywide web-based application system that receives and sends SMS mes- sages with the use of a short code on mobile-based communication tools. In 2016, OPM applied to the Uganda Communication Commission for a toll-free number so that users can send messages at no cost (Lumbroso 2016). At the district level, data are collected by Chief Administrative Officers using the Open Data Kit (ODK) to collect for detailed information to inform response to a disaster. The ODK has a number of forms that cover issues related to a hazard, such as: • Number of households affected • Origins of the data • Coordinates of the hazard • Number of mothers affected • Number of children affected 62 CHAPTER 3—Performance of EWSs in East and Southern Africa The ODK uses the DesInventar platform for collecting and generating data- sets. NECOC has an ICT infrastructure to operate the system and has a ded- icated technical team, so the system is technically sustainable. The District Disaster Management Committee has the responsibility to profile, validate the nature of the disaster, and thereafter decide on the appropriate response. The long-term goal is to get one trusted member of the community at a sub- county level operating the system (that is, one person per 60,000). However, the system does not forecast hazards and relies on an individual to feed information into the ODK. The system is currently funded by UNDP and it is unclear if OPM will provide funding to NECOC for future operations. Challenges Faced by EWSs in Uganda Most of the challenges in Uganda’s EWSs relate to governance and finance gaps. Good governance is facilitated by robust legal and regulatory frame- works and effective institutional arrangements. Critical challenges include: • Technological development, especially the use of mobile phone by most CSOs to collect EW information. This has led to “individualization” of EWS, citing failure of responsible agencies to act in real time. Thus, NECOC with the relevant line ministry take responsibility for any false warnings, accuracy, and authenticity of the information, as well as related costs. • Poor coordination of response and accountability. Early warning infor- mation is transmitted to relevant ministries, CSOs, and other agencies that quite often use it in bits to develop plans. This makes it impossible to arrive at a coordinated and timely action. • Lack of constant, guaranteed, and adequate funding, which results in unsatisfactory equipment and staffing levels. This therefore threatens the sustainability of early warning systems. • Limited technical capacity owing to the fact most of the country’s EWSs do not have a forecasting element to them, making it hard to engender early action. Therefore, inadequate capacities in terms of knowledge and skills continue to be a challenge to supporting the proper functioning of different early warning systems. Additionally, EW methodologies, tools, and techniques are often inadequate or poorly integrated. This, therefore, threatens the reliability and timeliness of EW information. RecommendaƟons: 1. Support incorporation of climate forecasts into most nationally available EWSs and tools to foster formulation and operationalization of early (timely) action. This will necessitate facilitating specialized training on use of forecast models and tools. 2. Produce and disseminate simplified EWS messages that include risk information designed to link threat levels to contingency preparedness and response actions; understood by authorities and end users; and issued from a single (or unified), recognized and authoritative source. CHAPTER 3—Performance of EWSs in East and Southern Africa 63 3. Support alignment of technical capacities with financial across national and local governments to facilitate development and out-scaling of rele- vant EWSs and tools and corresponding response mechanisms. 4. Establish a national early warning committee or secretariat to facilitate coordination with respect to existing EWSs and relevant tools. 5. Strengthen Uganda’s institutional framework for disaster management by supporting development of a comprehensive law on disaster risk manage- ment and food security related emergencies. 6. Build capacity to downscale global and regional climate change informa- tion to national and subnational level to support decision making. 7. Develop tools to support vulnerable households and communities to establish household community systems that can respond to emergencies. Performance of Food Security EWS at the Regional Level The EWS and Risk Knowledge at the Regional Level Conducting regular food security risk assessments is one of the key roles of the regional institutions. Understanding regional hazards and key vulnerabil- ities requires regular systematic collection and analysis of data to help priori- tize EWS needs and guide preparations for disaster prevention and responses. Table 3.6 shows the proportion of responses that indicate weaknesses in the risk assessment thematic subjects, scoring low on the 5-point scale, and the thematic subjects under risk assessment that are ineffective, showing low frequencies of the categories of Agree and Strongly Agree. Most participants (69  percent) agreed that the agencies involved in risk assessments at the regional level have clear roles and responsibilities. This was substantiated by two main reasons, as outlined in the following discussion. First, the main agencies involved in the collection of hydrometeorological hazard data are ICPAC and the SADC’s CSC. Institutionalized in IGAD and SADC, respectively, both ICPAC and CSC organize annual seasonal climate forecasts, through GHACOF and SARCOF, where they bring together mem- ber states and development agencies to report on weather forecasts for the next rainy season, often 6 months ahead of time. The impacts of GHACOF and SARCOF manifest on the extent to which they inform the National Climate Outlook Forums and help users to decide on their farming activities and dam management options and guide disease preventions such as malaria prevention and control. What is clearly a good practice is the extent to which the GHACOF report outlines actions that should be considered by the users (see box 2.2), which is not clearly stated in the SARCOF reports. Some of the challenges ICPAC and CSC face are summarized in box 3.3. Similarly, the roles and responsibilities of agencies involved in vulnera- bility assessments are outlined across the RECs. In IGAD the FSNWG is a forum for building consensus on food and nutrition policy and interventions, 64 CHAPTER 3—Performance of EWSs in East and Southern Africa TABLE 3.6 Effectiveness of Risk Assessments at the Regional Level Frequency, percent Strongly Somewhat Strongly Subjects agree Agree disagree Disagree disagree Agencies involved in hazard and 25 44 6 25 0 vulnerability assessments are identified and roles clarified. Hazards are regularly analyzed 13 31 19 24 13 and evaluated. Integrated risk and hazard maps 19 25 6 50 0 are regularly developed and assessed. Regional vulnerability 0 37 13 37 13 assessments are conducted annually and with disaggregated results. Regional vulnerability 19 31 31 19 0 assessments are conducted annually on the impact of seasonal climatic hazards. Industry is consulted during risk 0 7 53 27 13 assessments. Results of risks assessment are 19 37 25 13 6 integrated into regional risk management plans and warning messages. Central database is established 0 13 25 43 19 to store risk information and is available to users and is regularly updated. Average 12 28 22 30 8 and updates the food security and nutrition situation analysis (EW). With a membership of approximately 80 organizations (IGAD, UN agencies, NGOs, donors, and research institutions), the FSNWG has been serving regional governments, donors, and nongovernment agencies since 2006. The key out- puts of the FSNWG are monthly regional food security outlooks that provide key messages and actions required by member states and agencies working in the region. The major challenge for the FSNWG is that it is not institution- alized as an IGAD coordination body. Currently, there is no regional policy mandating the preparation of hazard and vulnerability plans for the regional food and livelihood security because such a policy has not been passed by the IGAD member states. Consistent with the vulnerability context of the Horn of Africa, the regional livestock and pastoralism coordination working group has also been estab- lished as a subgroup of the FSNWG. It aims at providing a platform for technical discussion and coordinated planning by professionals and experts dedicated to livestock and pastoralism programming in the Horn of Africa (covering seven countries: Djibouti, Ethiopia, Kenya, Somalia, South Sudan, Sudan, and Uganda). CHAPTER 3—Performance of EWSs in East and Southern Africa 65 However, although IGAD has established these working groups, the absence of a working group with a clear EWS mandate is one of the crit- ical gaps. The role of the EWS working group would include coordination of initiatives by various organizations such as the International Maize and Wheat Improvement Center–led Integrated Agricultural Production and Food Security Forecasting System for East Africa project that was launched in April 2015. Funded by the Consortium of International Agricultural Research Centers Research Program on Climate Change, Agriculture and Food Security under its Flagship 2 initiative, the project aims to develop a robust, scientif- ically sound, and user-friendly forecasting system that integrates improved seasonal climate, production, and food security forecasts for east Africa. It will also provide accurate and spatially disaggregated early warnings to local and national governments and relief agencies, enabling them to respond to climate crises in a timely and efficient manner. In the SADC, vulnerability assessments are guided by the Regional Vulnerability Assessment Committee (RVAC) that was established in 1999. This is a multiagency structure tasked with strengthening national and regional vulnerability analysis systems and supported by multiagency work- ing groups, with the following: • Nutrition Information Management Urban Assessments IPC Centre of excellence (of five universities) and capacity-building markets assessments. • The RVAC is supported by the National Vulnerability Assessment Committees (NVACs). Although since its inception in 1999, the Vulnerability Assessment Committee (VAC) system has relied on donors and was carried out as a project, the structure reviewed in 2017 shows that the VAC system is now institutionalized in the SADC structure. Although most of the countries are at different levels in terms of capacity, there is a shift toward domestic funding of the VAC process, for example, Botswana, South Africa, and Zambia generate their own funds to conduct the vulnerability assessments. • However, about two-thirds of the participants generally disagreed on the effectiveness of the risk assessments processes and outcomes. A closer look at the thematic subjects shows that 88 percent of the participants indicated that an integrated regional food security database did not exist in each of the regions. This was confirmed through regional interviews and during the validation workshop where participants indicated that although directorates at the regional level had risk databases, they were still working in silos because there was no central portal for sharing the risk information. Some 62 percent of the participants disagreed that the risk assessment results were disaggregated according to age, gender, or disability, yet this information is critical when targeting the at-risk popu- lation. Of all the thematic subjects, the lack of involvement of the private sector in risk assessments came under the spotlight with 93 percent of the participants indicating that the private sector was not involved in risk assessments. 66 CHAPTER 3—Performance of EWSs in East and Southern Africa TABLE 3.7 Effectiveness of Monitoring and Warning Service at the Regional Level Frequency, percent Strongly Somewhat Strongly Subjects agree Agree disagree Disagree disagree Roles and responsibilities of 7 13 7 66 7 agencies in generating and issuing EWs are established. Agreements are in place for 0 7 40 46 7 consistency in warning language and communication channels. Regional EWS systems are 0 20 27 33 20 subjected to systemwide tests and exercises annually. Regional EWS systems 13 7 47 33 0 committee is in place. Warnings are verified to check 0 20 40 33 7 they have reached the intended member states. A 24/7 regional warning center is 0 13 13 61 13 in place and staffed. Equipment for hazards and 0 36 57 7 0 vulnerability monitoring is suitable, and personnel trained in O&M. Hazards and food security data 7 33 27 33 0 from regional networks and international sources are accessible (shared across boundaries). Data are received and 0 42 29 29 0 processed and warnings disseminated on time, in meaningful formats, and in real or near-real time. Data analysis, prediction, and 7 13 7 66 7 warning are based on accepted scientific and technical methodologies. Average 3 20 29 41 6 EWS Monitoring and Warning Service at the Regional Level A credible EWS service is supported by a sound scientific basis for predicting and forecasting hazards to generate accurate and timely warnings. The warn- ing should be coordinated to gain the benefit of shared institutional, proce- dural, and communication networks. Table 3.7 shows the proportion of responses that indicate the effectiveness of the monitoring and warning service thematic subjects on a 5-point scale. Generally, more than two-thirds (about 70 percent) of the respondents were of the impression that the monitoring and warning service was ineffective at the regional level. A closer look at the thematic subjects reveals that 80 per- cent felt the roles and responsibilities for agencies mandated to issue warning were unclear, because there were no regional agreements or protocols that CHAPTER 3—Performance of EWSs in East and Southern Africa 67 clarify such roles. As well, 80 percent of the participants were not aware that a regional EW committee existed. The issuance of warnings tends to be the responsibility of directorates that have mandates in those areas. For exam- ple, the Food, Agriculture, and Natural Resources directorate takes respon- sibility of the issuance of warnings. Although at the national level there are established procedures, at the regional level there are no clearly defined pro- cedures. Developing such standards will not require considerable technical investment. Because the GHACOF and SARCOF models, for example, have a well-established system of issuing the EWs, it might be worthwhile building the standards using these models. Another concern expressed by about two-thirds (60 percent) of the respon- dents was that the EW information was rarely shared across boundaries. For example, whereas it is critical to access data from the Democratic Republic of Congo for flood warning for the Zambezi river, this information is not avail- able for countries in the Zambezi basin. Sharing information across boundar- ies is one of the requirements for effective monitoring of threats. The EWS for each country gives preference to the safety of its citizens. From this perspec- tive, cross-boundary risk can receive less attention. Although some data can be accessed from satellite-based datasets, there are considerable challenges in sharing EWS information between member states. The absence of MoUs or a regional policy or protocol compounds the challenge. For example, Somalia heavily relies on the rivers flowing from Ethiopia. Information on river levels in Ethiopia is rarely shared with food security analysts in Somalia; however, this information is critical for the Somalis whose livelihoods are dependent on those rivers to act if the information is shared on time. In some cases, the information is posted online but the problem is that rural communities may not be able to access such information because of lack of Internet access and resources to purchase Internet bundles. Systemwide testing and exercises for EWSs for hazards such as drought, floods, structural failures (for example, dam failures), fire, lightning, cyclones, thunderstorms, and pest infestations should be regularly conducted. About 80 percent of the participants felt that EWSs were rarely tested at the regional level. Although some countries conduct simulation exercises, for exam- ple, Madagascar and Mozambique on cyclones and floods and Zambia and Zimbabwe on the Kariba Dam wall failure, the tests for fire and insect infesta- tions such as the FAW are rare. A structured regional approach is required to enhance the capacity testing and simulation in the region. This will help expose the information and resources gaps. These weaknesses were compounded by the absence of a 24/7 regional warning center, or EOC, with about 87 percent of the participants correctly disagreeing that such a facility existed. Effective EWSs have a feedback system to verify whether warnings have been received and acted upon. Although the risk information on health, nutrition, food availability, migration, and cross-border traders, for example, is shared through the FSNWG and RVAC bulletins, 80 percent of the par- ticipants stated that there were no regional verification systems in place to ascertain the uptake and utilization of this information. Establishing such a 68 CHAPTER 3—Performance of EWSs in East and Southern Africa TABLE 3.8 Effectiveness of EWS Dissemination and Communication at the Regional Level Frequency, percent Strongly Somewhat Strongly Subjects agree Agree disagree Disagree disagree EW communication and 29 7 43 14 7 dissemination are tailored to the needs of the member states. Private sector resources are 7 7 50 29 7 available where appropriate. Warning alerts and messages 0 36 14 50 0 are tailored to the specific needs of those at risk. Average 12 17 36 31 5 system might not require substantial investment for users who have access to information technology because it might involve creating an interactive platform on the website. However, for the target group, and in this case the affected communities, it might be possible to use the existing focal persons network in member states who will be requested to provide feedback to verify whether the information has been acted upon. EW InformaƟon DisseminaƟon and CommunicaƟon at the Regional Level An effective EWS depends on the robustness of the dissemination and com- munication system that is tailored to the needs of the users. Table 3.8 sum- marizes the responses on the status of dissemination and communication of regional EW information. Although about 36 percent of the participants agree that a regional system for disseminating EWS information through, for example, SARCOF, GHACOF, FSNWG, and RVAC is generally well estab- lished, about 64 percent of the participants felt that warning messages were not tailored to the needs of at-risk communities, suggesting the EW informa- tion is of little use to the communities to help them make decisions. Although the involvement of the private sector in the dissemination and communication of EWS information has come under the spotlight over the years, about 86 percent of the participants stated that the private sec- tor resources were rarely used in the dissemination and communication of warnings at the regional level. KIs revealed that where the private sector was involved, these tended to be ad hoc, with no agreements or MoUs governing such involvement. UƟlizing EW InformaƟon for Response Planning The capacity to apply the EW information to preparedness and response plan- ning is one of the indicators of an effective EWS. Using a 5-point scale, table 3.9 presents respondents’ views on the extent to which the EW information trig- gered regional response mechanisms. The results reveal that most (64 percent) CHAPTER 3—Performance of EWSs in East and Southern Africa 69 TABLE 3.9 Effectiveness of Response to Regional Warnings at the Regional Level Frequency, percent Strongly Somewhat Strongly Subjects agree Agree disagree Disagree disagree Hazard and vulnerability maps 21 43 21 15 0 inform regional emergency preparedness and response plans. Regular tests and drills are 7 14 43 29 7 conducted to test the effectiveness of EWS dissemination and responses. Member states respond 0 21 36 29 14 effectively to EWs associated with all seasonal climatic hazards. Member states respond 7 14 43 29 7 effectively to all EWs associated with socioeconomic hazards. Average 9 23 36 26 7 of the participants agreed that EW data informed the regional preparedness and response plans. In the SADC region, the preseason workshops that are held before the onset of the rainy season use the data from SARCOF and RVAC to assist member states to develop contingency plans for the approaching rainy season. Similarly, in IGAD, the FSNWG issues bulletins, which assist member states to review and develop their contingency plans. Although the study found that risk data were used to develop contingency plans across the regions, the feeling by some 79 percent of the respondents is that there are no systems for simulation exercises. Particularly considering the threat of the FAW, the benefits of simulation exercises could be immense because they help in identifying gaps in preparedness and response plans, including information deficiencies, unrealistic planning assumptions and scenarios, and resource gaps. To develop such a system does not require a substantial amount of investment, because the major input is to develop the exercising materials, which can be developed with the support of cooperating partners. However, to be effective, what might be required is to develop an exercising group that will provide guidelines and schedules for such exercises. Timely response to warnings is of essence to save lives and livelihoods. If the affected communities responded on time, it is possible to reduce the dev- astating impact of a materialized hazard. However, 79 percent of the partic- ipants in this study felt that the regional response was often delayed until the situation turned into an emergency. It is clear from the way the SADC responded to the 2016 El Niño-induced drought that the SADC EWS faces challenges, because this system was unable to adequately inform the SADC of the impending food insecurity. At a meeting in Johannesburg in February 2016 that was intended to produce a road map for the humanitarian appeal, most member states did not have up-to-date information. It was not until 70 CHAPTER 3—Performance of EWSs in East and Southern Africa June 2017 that member states could produce the assessments results to inform the appeal that was launched in July 2016. This suggests the system is “not well oiled” and, therefore, not as effective in providing EWS information on time. One way of prompting early action is to engage in advocacy with policy makers and senior government officials on the benefits of taking early action. This requires evidence on the costs and benefits of taking early action, which can be obtained post disaster events. EWS Regional Governance Mechanisms Well-developed governance and effective institutional arrangements support the successful development and sustainability of sound EWS systems. These should be supported by good governance, robust legal and regulatory frame- works, long-term political commitment, broader participation, and adminis- trative and resource capabilities at the national and subnational levels. Vertical and horizontal communication and coordination between EW stakeholders should also be established (UNISDR 2006). Table 3.10 summarizes the regional EWS institutional arrangements. Some 92 percent of study participants noted a lack of awareness of regional lead- ers on the benefits of investing in EWSs, whereas 77 percent of the partici- pants felt that EWSs are not integrated into the regional planning process. This suggests that EWSs are perceived as a response activity rather than part of regular development processes. There was concern over unreliable and unpredictable resource mobilization from the RECs, with all the participants, both questionnaire and interviews, stating that EWSs were primarily funded by external sources, which raises questions on the sustainability of programs. Interestingly, all participants also felt that the PPPs were underused to help generate alternative sources of funding. Besides the resource challenges, the ESA region faces coordination chal- lenges. The following five RECs in ESA are recognized by the AU: • EAC • SADC • IGAD • COMESA • ECCAS In addition, there are two RECs that are not recognized by the AU: • Southern African Customs Union (SACU) • IOC All the RECs have a focus on regional economic cooperation and inte- gration, and considering the ESA region being agro-based, all the RECs have a remit on food security. Table 3.11 shows that most countries in the ESA region have multiple memberships to the RECs, with 11 countries being members of three RECs. This creates a situation of overlapping and in some cases competing commitments across member states, which poses CHAPTER 3—Performance of EWSs in East and Southern Africa 71 TABLE 3.10 Institutional Arrangements and Investment at the Regional Level Frequency, percent Strongly Somewhat Strongly Subjects agree Agree disagree Disagree disagree Benefits of EWS are highlighted 0 8 38 46 8 to regional leaders. EWS is integrated into regional 0 23 62 15 0 planning. Regional and cross-border 0 8 58 34 0 agencies are established for EWS integration. Capacities of agencies are 0 0 38 54 8 assessed and capacity-building plans developed and resourced. Regional funding mechanisms 0 0 46 39 15 for EW are developed and institutionalized. PPPs are utilized to support the 0 0 54 23 23 EWS system. Regional EWS policy is in place. 13 19 56 6 6 Regional standards for EWS are 0 38 31 13 18 in place. Agreements are in place for 0 7 40 47 8 consistency in warning language and communication channels. Regional EWSs committee is in 13 7 47 33 0 place. Warning dissemination chain is 0 21 36 36 7 enforced through regional policy. Agreements are developed to 0 15 54 23 8 utilize private sector resources in the dissemination warning. Average 2 12 47 31 8 challenges for members to discharge their obligations, including maintain- ing their membership subscriptions. A closer look at table 3.12 reveals some duplication within the RECs on food security, but they also tend to heavily rely on external resources. Of the countries studied, Swaziland and Zimbabwe belong to the SADC, whereas Djibouti, Ethiopia, Kenya, Somalia, and Sudan are IGAD members. Synthesis of data from IGAD and SADC member states reveals minimal differences. Across the board, the following are salient issues that deserve pol- icy action at a REC level: • Nonexistence of national standards for risk assessment • Limited reliance on hazard and vulnerability maps in the development of national emergency preparedness and response plans • Nonexistence of regular public awareness/education campaigns concern- ing disaster risks • Limited consultation of affected communities and industries while con- ducting risk assessment 72 CHAPTER 3—Performance of EWSs in East and Southern Africa TABLE 3.11 ESA RECs and Their Member States Overlaps Country COMESA EAC ECCAS IGAD IOC SACU SADC Angola X X X Botswana X X Burundi X X X Comoros X X Congo, Dem. Rep. X X X Djibouti X X Egypt X Ethiopia X Eretria X X Kenya X X X Lesotho X X Libya X Madagascar X X X Malawi X Mauritius X X X Mozambique X X Namibia X X X Rwanda X X X Seychelles X X X Somalia X South Africa X X South Sudan X X X Sudan X X Swaziland X X X Tanzania X X Uganda X X X Zambia X X Zimbabwe X X • Absence of binding agreements for consistency in warning language and communication channels • Failure to tailor warning alerts and messages to the specific needs of those at risk and incorporate the understanding of the values, concerns and interests of those who will need to take action This means challenges to EWSs across the African continent are similar, with scattered best practice information. Thus, the need exists to encourage and support knowledge and experience sharing between countries, regardless of REC membership. CHAPTER 3—Performance of EWSs in East and Southern Africa 73 TABLE 3.12 Regional EWS Institutional Summary 74 Region Agency/system Roles and responsibilities Products and services Weaknesses SADC CSC Provides operational services for monitoring Weather/climate forecasts, including SARCOF and Technical capacity and resources and predicting climate extremes capacity building Food Security Regional Early Warning System provides Food security bulletins, seasonal outlooks, and Technical capacity, external Information System advance information on food crop yields and monthly regional balance sheet resources, and diverse food supplies and requirements. methodologies RVAA Humanitarian situation, RFBS, and food insecure Technical capacity, resources, population and multiple methodologies DRR Unit Develops regional DRR strategies and DRR strategies, including regional preparedness Limited technical capacity and facilitates preseason preparedness plans and response strategies reliance on external resources and postseason lessons learning Regional Early Warning Strengthen the SADC mechanisms for Security information and inputs to RVAA program No clear integration to food Centre conflict prevention, management, and security resolution IGAD ICPAC Climate monitoring, prediction, applications, Dekadal (10 day) and monthly bulletins, seasonal Reliance on external funding capacity building, environmental monitoring, outlooks including GHACOF, and climate watch DRR, dissemination, and research Conflict Early Warning Assess situations that could potentially lead Linking conflict and food security and EWSs by Weak link with food security EWS and Response to violence or conflicts and prevent considering ICPAC climatic forecast and reliance on external Mechanism escalation resources Center for Pastoral Regional livestock and dry-lands policy, Transboundary animal disease control, livestock Reliance on external funding Areas and Livestock research and development, coordination, products trade, and mapping cross-border Development and capacity building resources FSNWG Up-to-date food security and nutrition Monthly regional food security and nutrition outlook External resources and not analysis and building of consensus on through e-mail and posted online institutionalized in IGAD structure critical issues facing policy and intervention EAC Agriculture and Food Improve food security, strengthen the EWS, RFBS, cross-border trade monitoring, and EAGC Policy, technical capacity, and Security and increase inter/intra-regional trade bulletin multiple membership CHAPTER 3—Performance of EWSs in East and Southern Africa COMESA Climate Change Unit Address the impacts of climate change in the Updates on the COMESA-EAC-SADC tripartite Reliance on external funding, COMESA-EAC-SADC region and promote arrangements coordination between the RECs, climate-smart agriculture and multiple membership Food and Agricultural Improve agricultural marketing through the Information portal and cellular-based information Reliance on external funding, Marketing Information dissemination of market information, policy platform to inform conservation agriculture, weather, coordination between the RECs, System changes, and impacts and agro-dealers and multiple membership CHAPTER 4 Strategies for Long-Term Sustainability of Investments on Early Warning Systems Introduction Documenting the value of hydrometeorological services helps to justify investment in NMHSs as well as those agencies tasked with vulnerability assessments and monitoring. It is particularly important to consider that more than 100 NMHSs in developing countries need substantial investment to bring their services to a level at which they can provide timely, reliable, and accurate forecasts of high-impact weather to the public and to national eco- nomic sectors (Rogers and Tsirkunov 2013). This chapter highlights some of the investment required to strengthen an EWS. The chapter begins by a reflec- tion on the cost-benefit analysis of investing in EWSs. The costs are estimates based on other studies and field consultations. It is important to highlight that only a few comprehensive studies on the benefit-cost ratio have been carried out in developing countries. Some of the costing presented in this chapter has not been subjected to rigorous costs-benefit analyses to determine the BCR and the economic rate of return to such investments. The Costs and Benefits of Investing in an EWS Although there is a widespread recognition of the need to improve countries’ capacity to prepare and respond to evolving and future food insecurity crises, as the data suggest in chapter 3, there is still a challenge investing in food security EWSs in the ESA region. Part of the problem is lack of studies that demonstrate the benefits of such investments. Most of the studies that have been conducted have tended to be focused toward assessing the benefits of investing in agrometeorological services. There is still a high level of com- partmentalization in the analyses that hinder the ability to integrate the entire food security and EW information system, considering all the components of effective end-to-end EWS, that include physical climate hazards and social vulnerability dimensions. However, even without a comprehensive analysis of costs and benefits of the entire information system, the studies on weather and climate services are overwhelmingly positive and suggest the triple bot- tom line benefits of investing in NMHSs (box 4.1). The BCRs as illustrated in table 4.1 range from 2:1 to 36:1, and in one study, in which the value of lives was quantified, a BCR of 2,000:1 was estimated. CHAPTER 4—Strategies for Long-Term Sustainability of Investments 75 BOX 4.1 Benefits of Early Action A study by the U.K. Department for International Development that examined the relative costs of early and late action for drought in Kenya and Ethiopia suggests that the economic case for early action is compelling. Over a 20-year period of droughts on a 5-year cycle, it was estimated that early procurement by agencies and early interventions to support pastoralist livelihoods such as commercial destocking, supplementary animal feeding, and veterinary services could generate significant savings relative to an emergency response. In the Wajir district of Kenya these were estimated to be US$392 million, and in the more populous region of southern Ethiopia it was US$3,066 million: more than US$1,000 per beneficiary in each case (table B4.1.1). TABLE B4.1.1 Cost Estimates for Drought Responses in Horn of Africa, Discounted over 20 Years Saving Beneficiaries Emergency Early action Saving per capita Wajir, Kenya 367,000 US$606 million US$214 million US$392 million US$1,068 Southern 2,800,000 US$3,800 US$734 million US$3,066 US$1,095 Ethiopia million million Source: Bailey 2012, 6. In developing countries, which include the ESA countries, the benefits, on average, range from 4:1 to 36:1, for example, in Nepal (10:1) and Ethiopia (3:1 to 6:1). There is, however, a need for caution when analyzing those World Bank projects in Sub-Saharan Africa in table 4.2 that have a strong element of food security and EW information systems. Although the range of BCRs appears to be at the lower end of the average of developing countries, if the wider social benefits were included in the cost-benefit analyses, in addition to the direct quantifiable benefits, the benefits are much higher. Similarly, the BCR of improving national hydrometeorological services in developing countries ranges from 4:1 to 36:1. In Ethiopia, a drought EWS can potentially reduce livelihood losses and dependence on assistance, with a BCR ranging from 3:1 to 6:1 (WMO 2015). There are several examples of the benefits of investing in EWSs. Some EWS projects are currently being implemented across Africa (table 4.3). In the Democratic Republic of Congo, the World Bank is supporting a project called “Strengthening Hydro-Meteorological and Climate Services.” The aim of the project is to improve the quality of the government of the Democratic Republic of Congo’s targeted hydrometeorological and climate service by doing the following: • Improving hydrometeorological and EW capacity and strengthening net- works through open data and information sharing • Leveraging partnerships and fostering interagency coordination to maximize economies of scale and regional integration and promote south-south coop- eration to ensure transformational change and longer-term sustainability 76 CHAPTER 4—Strategies for Long-Term Sustainability of Investments TABLE 4.1 Examples of Triple Bottom Benefits of Investing in Hydrometeorological Services Theme Benefits Social Avoidance of loss of life and/or injuries/illnesses from natural disasters Safety and security of the traveling public; improved information and data for the scientific community Contribution to the day-to-day safety, comfort, enjoyment, and general convenience of citizens, including: • Recreation • Travel and commuting • Preparation for severe weather • Home improvement decisions • Other direct and indirect forms of societal benefits • Event management Avoided climate-related illnesses (for example, heat-related illnesses, vector-borne diseases such as malaria that are worsened by climate) Environmental Long-term monitoring of basic indicators of the state of the environment Minimization of release of toxic substances and other pollutants Management of local environmental quality Support for addressing major global environmental issues Water savings Reduced runoff from fertilizer application, resulting in improved water quality Economic Avoidance of crop losses from frost, hail, or drought Increased farm production and sales More efficient scheduling of the use of agricultural machinery Reduced transportation fuel consumption through route planning Improved scheduling of flight arrivals and departures Minimization of airline costs from aircraft diversions Minimization of search and rescue costs Minimization of drought-relief costs Efficient scheduling of ship loading facilities Avoidance of unnecessary shutdown of offshore oil and gas operations Avoidance of weather damage to personal property More efficient planning of energy production and delivery Source: Lazo et al. 2009, cited in WMO 2015, 60. • Aligning with the principles of the Global Framework for Climate Services (GFCS) and identifying the requirements of users as a starting point for generation of services, products, and data The costs associated with the project amount to a total of about US$8 million (table 4.4) with an estimated NPV23 of US$117.27 million and with a BCR of 7.36:1 at 3 percent baseline discount rate.24 Based on the “Triple 23 Net present value is the present value of a security or an investment project, found by discounting all pres- ent and future receipts and outgoings at an appropriate rate of discount (see note 24). If the NPV calculated is positive, it is worthwhile investing in a project (Black, Hashimzade, and Myles 2017). 24 Discount rate is the interest rate at which future benefits or costs are discounted to find their present value (Black, Hashimzade, and Myles 2017). CHAPTER 4—Strategies for Long-Term Sustainability of Investments 77 TABLE 4.2 Illustrative Economic Assessments of Meteorological/Hydrometeorological Services 78 Geographic Social economic benefits study location Sectors Benefits methods/measures BCR Contingent valuation study of the public weather service in the Sydney Sydney, Australia Households Willingness to pay survey of households 4:1 metropolitan area (Anaman and Lellyett 1996). Benefits of Ethiopia’s Livelihoods, Early Assessment, and Protection Ethiopia Households Quantification of avoided livelihood losses and 3:1 to 6:1 drought early warning and response system (Law 2012) decreased assistance costs Success of the U.S. National Weather Service Heat Watch/Warning Philadelphia, Households/ Regression analysis to determine lives saved, 2,000:1+ System in Philadelphia (Ebi et al. 2004) Pennsylvania, elderly application of the U.S. Environmental Protection United States Agency value of a statistical life estimate The benefits to Mexican agriculture of an El Niño-Southern Oscillation Five-state region Agriculture Change in social welfare based on increased 2:1 to 9:1 (ENSO) early warning system (Adams et al. 2004) in Mexico crop production with use of improved information Economic efficiency of NMHS modernization in Europe and Central Asia 11 European and Weather- Sector-specific and benchmarking approaches 2:1 to 14:1 (World Bank 2008) Central Asian dependent to evaluate avoided losses countries sectors Benefits and costs of improving met/hydro services in developing Developing National level Benefits-transfer approach to quantify avoided 4:1 to 36:1 countries (Hallegatte 2012) countries and weather- asset losses, lives saved, and total value added sensitive in weather-sensitive sectors sectors Social economic benefits of enhanced weather services in Nepal—part Nepal Agriculture, Statistical inference and expert judgment 10:1 of the Finnish-Nepalese project (Makela et al. 2011) transport, and hydropower Economic and social benefits of meteorology and climatology (Frei 2010) Switzerland Transport, Benefit transfer, expert elicitation, decision 5:1 to 10:1 energy, aviation, modeling agriculture, and households Socioeconomic evaluation of improved met/hydro services in Bhutan Bhutan National level Benefit transfer, expert elicitation, cardinal rating 3:1 (Pilli-Sihvola, Namgyal, and Dorji 2014) method CHAPTER 4—Strategies for Long-Term Sustainability of Investments Source: Adapted from WMO 2015, 8–9. TABLE 4.3 Benefits of Investing in Climate Information and EWSs Net present value (NPV) BCR Investment US$, Discount rate Country Name of project Project duration (US$, millions) millions (percent) Kenyaa Kenya Climate-Smart Agriculture Project with a 2017–22 279 304 6 1.40–5.56 component on Supporting Agro-weather, Market, Climate and Advisory Services Mozambiqueb Mozambique Climate Resilience: Transforming 2013–18 22.50 391.17 3 6.56 Hydrological and Meteorological Services Congo, Dem. Rep.c Strengthening Hydro-Meteorological and Climate 2017–22 8.029 117.27 3 2.39–7.36 Services Malid Strengthening Climate Resilience in Sub-Saharan Feasibility study 22.75 124.4 5 5 Africa: Phase 1 Mali Country Madagascar National Action Plan for Improvement of Feasibility study 58.48 44.9 3 1.67–7.60 Hydrometeorological Services Nigeria Transforming Irrigation Management in Nigeria 2014–22 560.3 12 13.5 Not stated Zambia Zambia Strengthening Climate Resilience (PPCR 2013–19 36 18.3 3 Not stated Phase II) Tanzania Second Water Sector Support 2017–21 230 180 10 Not stated Uganda Water Management and Development 2012–18 135 24.6 15 Not stated Note: CHAPTER 4—Strategies for Long-Term Sustainability of Investments a This includes US$32.9 for the development of agro-weather forecasting and marketing information system and their dissemination tools through improving agrometeorological forecasting and monitoring institutional and technical capacity; using big data to develop a climate-smart, agro-weather MIS and advisories. The Project Appraisal Documents (PADs) can be accessed at http://documents. worldbank.org/curated/en/440241486868444705/pdf/Kenya-PAD-01182017.pdf. b The PAD can be accessed at http://documents.worldbank.org/curated/en/506641468321881521/pdf/759390PAD0P1310040901300SIMULT0DISC.pdf. c The analysis follows the overall structure of the “Triple Dividend of Resilience” framework, which includes (i) avoided damage and losses, (ii) unlocked economic potential, and (iii) development cobenefits. The PADs can be accessed at http://documents.worldbank.org/curated/en/289901488209393335/pdf/DRC-GEF-PAD-02222017.pdf. d See http://www.greenclimate.fund/documents/20182/226888/GCF_B.13_16_Add.04_-_Funding_proposal_package_for_FP012.pdf/2af42103-a150-4e0b-a3f3-29d8b433e377. 79 TABLE 4.4 Financing for Strengthening Hydrometeorological and Climate Services in the Democratic Republic of Congo Component, subcomponent, and indicative budget (US$) Total (US$) Component A: Institutional and regulatory strengthening, capacity building, and 1,578,000 implementation support A (i) Reinforce the legal and regulatory framework of MettelSat to develop partnerships and SOPs for delivery of service A (ii) Strengthen the quality management systems to raise standards and quality control/verification procedures across the institutions A (iii) Implement a long-term and on-demand capacity development and training program for staff Component B: Modernization of equipment, facilities, and infrastructure for basic 4,462,000 observation and forecasting B (i) Hydrological and meteorological monitoring networks (small-scale rehabilitation of priority stations and installation of new sensors) B (ii) Transmission, data management, and data dissemination hardware B (iii) Refurbishment of facilities needed to support the services B (iv) Technical systems and software for performing meteorological, hydrological, and climate modeling, and forecasting Component C: Improvement of hydromet information service delivery 1,256,000 C (i) Define requirements, delivery, and feedback mechanisms with different user groups (in line with the National Framework for Climate Services) C (ii) Develop customized products and services made available to user groups through dedicated interfaces Component D: Project management 733,452 D (i) Coordination and technical implementation support D (ii) Fiduciary and safeguard aspects and audit Total 8,029,452 Source: PAD for Financing for Strengthening Hydro-Meteorological and Climate Services in the Democratic Republic of Congo. Report No: PAD1864 Dividend of Resilience” framework, the benefits will include (a) avoided damage and losses, (b) unlocked economic potential, and (c) development cobenefits. Although the Democratic Republic of Congo project identifies the impor- tance of legal and regulatory frameworks and gender awareness, it does not have a strong institutional capacity-building component at the subnational level, which participants of this study called “the more encompassing and proactive food information approach which addresses all components of food security in line with its definition. However, it is important to emphasize and recognize that the project’s main objective was not necessarily on address- ing food security risks. The objective is directly aimed at hydromet modern- ization, which has direct, measurable impacts on improving the country’s capacity to provide enhanced agro-met services, and in turn, improving agri- cultural extension services. “It is widely recognized that enhancing the knowledge on hydromet haz- ards is key for improving the effectiveness of multi-hazard EWS. However, while the significance of meteorological data/information in food security information cannot be underestimated, it remains one of many other data/ 80 CHAPTER 4—Strategies for Long-Term Sustainability of Investments TABLE 4.5 Estimated Costs for National Action Plan for Improvement of Hydrometeorological Services (NAPIHMS) Project in Madagascar (US$) Phase 1 Phase 2 Phase 3 O&M (years (years 1–2) (years 3–5) (years 6–10) 11–15) Totals Monitoring network equipment 25,785,402 Hydrological observation network 531,250 1,593,750 3,187,500 5,312,500 Groundwater observation network 403,200 604,800 1,008,000 Meteorological and climatological 263,135 726,405 1,011,810 2,001,350 observation network Radar 6,000,000 6,000,000 12,000,000 Volunteer rainfall network 3,305 9,915 19,830 33,050 Vehicles 96,000 288,000 576,000 960,000 Water quality lab equipment 26,600 79,801 159,601 266,002 Discharge measurement equipment 63,000 189,000 378,000 630,000 Field equipment 215,625 646,875 1,293,750 2,156,250 Data management 2,715 8,145 16,290 27,150 Rehabilitation of the DGM hydrometeorological 150,000 150,000 instrumentation laboratory. Installation of new hydrometeorological 111,510 384,930 744,660 1,241,100 stations Forecasting and service delivery 5,065,285 Headquarters and regional hydrological 72,000 781,714 1,508,571 1,560,000 3,922,285 forecasting center O&M Headquarters and regional hydrological 207,000 346,000 185,000 155,000 893,000 forecasting center capital investment Weather forecasting equipment and 150,000 50,00 50,000 250,000 software O&M 18,335,633 Monitoring network 582,369 2,362,314 7,494,225 7,494,225 17,933,133 Stations rehabilitation works 11,250 26,250 37,500 Stations rehabilitation equipment 109,500 255,500 365,000 Research and development 992,008 1,488,012 2,480,020 Training and capacity development 222,300 666,900 1,333,800 2,223,000 NMHSUG training 100,000 100,000 200,000 Consulting 1,760,000 1,320,000 1,320,000 4,400,000 Total by phase: 4,677,559 17,180,707 27,371,849 9,209,225 Grand 58,489,340 Total: Note: The costings are based on the NAPIHMS World Bank project in Madagascar. information types of food security. Hence, there is a need to pay attention to the other data types and sources.”25 The examples from Madagascar (table 4.5) and Zambia (table 4.6) are more reflective of the comprehensive investment needs of most ESA countries as they address both the technical and the insti- tutional needs at both national and subnational levels. 25 One of the participants consulted as part of this study. CHAPTER 4—Strategies for Long-Term Sustainability of Investments 81 TABLE 4.6 Estimated Costs of the Food Security EWS in Zambiaa Sector Capacity needs Cost (US$) ZMD Equipment (AWSs, radars, computers, software, and so on) 13,080,000 Technical capacity (training for staff) 2,250,000 Institutional capacity (training, AWSs/radars, O&M, and so on) 35,000 Subtotal 15,365,000 Hydrological Services Equipment (telemetric stations, computers, and so on) 7,553,150 Technical capacity (human resources capacity building) 935,000 Institutional capacity (training catchment councils, and so on) 3,340,000 Subtotal 11,828,150 Agriculture Equipment (increasing rain gauge network) 180,000 Technical capacity (capacity for data analysts) 20,000 Institutional capacity (capacity for extension services) 50,000 Subtotal 250,000 DMMU Technical capacity (national level) (information system setup) 1,089,000 Institutional capacity (national level) 1,598,050 Provincial-level capacity building 559,000 District-level capacity building 19,527,500 Satellite-level capacity building 36,380,000 Subtotal 59,153,550 Grand total 86,596,700 a The cost estimates were produced by the DMMU, with inputs from Zambia Meteorological Services, Hydrology, and Agriculture. However, the estimated investments are beyond the capacity of most ESA countries. A phased approach might be required. A study conducted by the World Bank (2015) illustrates three investment options on a continuum, from traditional to advanced weather station programs (table 4.7), with each option having a high BCR. This suggests that even when a phased approach is considered, the benefits accruing from the investment are still high. The same study applies the results to Ethiopia and Kenya, which may also apply widely to most ESA countries. In Ethiopia, agriculture accounts for about 39.4 percent of the national gross domestic product (GDP) = US$41.72 billion (2012 estimate, or about US$16.44 billion). If a tradi- tional or minimal agrometeorology program is considered, it would increase the agricultural contribution to GDP by about 0.18 percent, a transitional program by about 1.5 percent, and a modern or advanced program by about 6.1 percent. Similarly, in Kenya, agriculture accounts for about 24 percent of the national GDP of US$40.70 billion (2012 esti- mate or US$10 billion). If a traditional or minimal agrometeorology pro- gram is considered, this would increase the agricultural contribution to GDP by about 0.30 percent, a transitional program by about 2.5 percent, 82 CHAPTER 4—Strategies for Long-Term Sustainability of Investments TABLE 4.7 Options for Investing in Observation Networks Option Traditional/minimal Transitional Modern/advanced Description Mostly traditional instrumentation, Still mostly traditional National network of with automated systems at only a instrumentation, but more automated systems very limited number of locations; automated systems in supplemented by limited professional regional networks; staff of traditional agrometeorological staff, very agrometeorologists and instrumentation, limited field training or extension supporting technicians still professional staff work; no real-time operations modest in size but now now able to support sufficiently large to begin full-scale real-time routine training of farmers operations and and support fieldwork for training and extension support Initial capital US$3 million US$20 million US$50 million investment in equipment, facilities, training Continuing annual US$1 million US$5 million US$10 million cost Cost-benefit ratio 1:30; US$30 million increase in approximately 1:50; 1:100; US$1 billion agricultural production US$250 million increase in increase in agricultural production agricultural production and a modern or advanced program by about 10 percent. The comments arising from this are as follows: • Clearly, as one progresses from a traditional to advanced program, the contribution to the GDP increases. In Uganda, for instance, the remark- able decrease in poverty levels is directly attributable to increases in the contribution of agriculture to the GDP. • A traditional or minimal program is consistent with the current invest- ment levels in most ESA countries and such a program does not contrib- ute much to the national GDP. The resources could be deployed elsewhere where they can yield better benefits. • Although the advanced program has the potential of delivering increased economic benefits, policy makers may not support the investment required. This will need evidence-based advocacy through contextualized EWS cost-benefit studies. According to the findings of this study, upgrading the observation networks alone is inadequate for achieving the goals of measurably reducing the levels of poverty and food insecurity across the ESA regions. However, it is a required ingredient in the establishment of effective and efficient end-to-end EWSs, including those for food security. This means a phased strategy is required to incrementally build on the successes of each of the phases. A strategy that works in partnership with the private sector and farmers’ union might be required to leverage on the resources. CHAPTER 4—Strategies for Long-Term Sustainability of Investments 83 Governments in Africa should commit to and take responsibility for the adequate funding of the NMHSs, as basic hydromet monitoring and forecast- ing services shall be provided as a public good. Strong NMHS can develop additional sources of funding through the provision of specialized services to sectors and industries, such as aviation and insurance, but the societal ben- efits of the provision of basic hydromet services at no cost to all audiences surpass the direct costs to the government. Governments should ensure that NMHSs are able to discharge their institutional responsibilities in an ade- quate and sustainable manner. As part of the strengthening of the NMHs and the services they provide, exploring win-win NMHS-led partnerships with the private sector should also be encouraged. Cost-Effective Strategies for Improving Agrometeorological Observation Systems One of the key findings of this study is that most ESA countries have some system for hydrometeorological observation services. Many countries have both the traditional observation networks, which employ observers or local volunteers, with only modest training and AWSs. There are two strategic options for improving the weather observation networks, and ultimately the quality and types of agrometeorological services that can be provided to tar- geted audiences. First is to enhance the existing manual weather stations and the second is to migrate incrementally to a modern and advanced observation network. OpƟon 1: Enhancing the Manual Weather StaƟons Most meteorological and hydrological observation networks in the ESA countries are manual weather stations. Although it is recognized that not all the instrumentation or methods used for hydromet monitoring and fore- casting are used for or are applicable to agrometeorology (for example, lake and maritime navigation), adequate coverage of the observational networks contributes to improve countries’ capacity to provide adequate, timely agro- meteorological to agriculturists, pastoralists, and other audiences involved in weather- and climate-sensitive sectors. In Ethiopia, for example, there are several hundreds of manual weather sta- tions, a similar situation across the ESA countries. In most cases, the obser- vations from manual networks are recorded by hand and often mailed into a central office. Such observations, when passed by a rigorous quality assurance process, provide invaluable data for both long-term climate studies and mon- itoring of how given growing seasons are evolving. Equipment such as rain gauges can be locally fabricated, and such weather stations can be placed in schools and rural health centers. A simple data- sheet is recorded by hand, as is the case in Zambia’s Sesheke weather station, after which the data are transmitted by mobile phone or mailed to a central 84 CHAPTER 4—Strategies for Long-Term Sustainability of Investments location. The data transmitted through the mobile phone go to an automated receiving system that records the data, does simple error checking, and pro- vides a first-order analysis. Although manual weather stations have several disadvantages, such as data inaccuracies and slowness in data collection and transmission, the manual weather stations tend to have relatively low O&M costs. It has been well documented that in the medium to long term, O&M of hydromet monitoring and forecasting represent the highest cost items of any investments in hydromet modernization, surpassing the original costs of upgrading equipment and software. Such high costs of O&M are the primary reason why so many well-intentioned equipment improvement projects failed when the funding ended. The application of mobile technologies, as is the case at Sesheke weather station in western Zambia, facilitates the transmission of near-real-time data to Lusaka three times per day. Using mobile phone technology enables fast data aggregation and calculation of temperature and precipitation data. In Sesheke, mobile phones are also used to share weather data with local farmers, for example, through WhatsApp groups and community radio stations. The estimated cost of upgrading a manual station to use mobile phone tech- nology in data collection and transmission in Zambia is about US$75,000, which is consistent with the World Bank (2015) study that estimated the costs between US$50,000 and US$75,000 to install the central data collection computer. There will be deployment and training costs to get the new equipment into the field and to train the observers or volunteers to operate it. Upgrading a manual weather station, as is the case with Sesheke, comes with continu- ing charges for the mobile phone services and maintenance of the central computer. However, if mobile phones charges are unbearable, the observers switch to pen and paper format. In any case, in Sesheke, for example, in addi- tion to the data recorded in a mobile phone, a manual copy is also maintained as a backup. The study by the World Bank (2015) reiterates the importance of training the operators at the central computer to a higher skill level in pro- gramming and system administration. Additional personnel should include a small team of three or more quality assurance meteorologists who review the raw data in parallel with the automated procedures, flagging questionable data. OpƟon 2: TransiƟon to Modern and Advanced Networks There are several different costs involved in developing a national meteoro- logical observation system. The study by the World Bank (2015) categorizes the costs that may be involved in investing in advanced observation networks. These are summarized in table 4.8. The study by the World Bank (2015) also suggests the steps to be followed for the strategy to be cost effective. CHAPTER 4—Strategies for Long-Term Sustainability of Investments 85 TABLE 4.8 Estimated Annual Costs of an Advanced Network Estimated Element Description Cost (US$) Site costs Purchase or long-term lease of the site; civil works and site preparations 55,000 as needed (access road, security fence, electrical power if available or a solar panel/battery system—the last one is a good backup against outages even if electrical power is available); tower or mast; instrumentation; data logger and software; telecom interface (could be microwave radio, cell phone connection, hardwired Internet connection, and so on; can be two-way to allow remote interrogation of the data logger); cables and related hardware; labor for site improvements, installation, and so on; miscellaneous costs such as shipping costs related to getting equipment on site. Costs can run between US$10,000 and US$100,000 per site, depending on the number and types of instruments installed. A good mean cost would be US$55,000. Central data Facility preparation (space modifications, provision of power, provision for 225,000 collection facility air conditioning, dehumidification, heating as necessary); telecom costs interface for receiving incoming data streams (and transmitting commands to data loggers in the field); computer(s) for receiving, decoding, quality checking, and basic analysis (this could be the same system that receives data from the traditional network); display system Can run between US$50,000 and US$350,000 (average = US$225,000) Central Facility preparation (space modifications, provision of power, provision for 150,000 maintenance facility air conditioning, dehumidification, heating as necessary); facility furnishings to include workspace for maintenance and calibration, and storage space for spare instruments, repair parts, and expendables; tools and calibration equipment; miscellaneous costs such as shipping costs Costs can run between US$50 and US$250,000. Staff costs Data analysts, software developers, technical maintenance/calibration 130,000 staff—about 8 to 10 professionals required at US$15,000 per year = US$120,000–US$150,000 Operating costs Telecom charges; repair parts and spare instruments; expendable 140,000 supplies for operating and maintaining the data collection system; vehicles, fuel, and the related costs (a small group of vehicles is needed for all agrometeorologists to go out in the field to study local conditions and to support technical maintenance/calibration staff making routine site visits). Costs can run between US$25,000 per year and US$250,000 per year. Total Approximately US$1 million 700,000 Step 1a. In parallel, build up and enhance the existing traditional network of rain and temperature observations taken by observers and volunteers. Where the cell network allows, provide observers with mobile phones to call in their observations once per day. Step 1b. In parallel, establish an automated dial-in data delivery sys- tem so that people can begin sending information. This computer should have sufficient capability to service both the dial-in delivery of data from network of traditional sites, data from manual weather stations received through mail, and the acquisition of data from the modern or advanced sites. This means the central data collection goes in at this early stage, so 86 CHAPTER 4—Strategies for Long-Term Sustainability of Investments that it is ready to receive data from the AWSs (as well as from the enhanced traditional network). Step 2. Select sites for the installation of the AWSs supporting agrometeo- rology. Keep in mind that this is an agrometeorology network, so the stations will be concentrated in agricultural regions with sites selected to be represen- tative of local farming operations. Ideally sites would be within, but distinct from, fields. However, this is not always practical, so sites at the edge of fields may be utilized. The program can start with a few stations and then gradually upscale to cover the whole country. Strengthening Food Security Information Systems at the Regional Level At the regional level, there are efforts in developing the EWS information systems, which include food security. Currently, the RECs have multiple information systems and networks, some of which have continental and international standing. ICPAC and Conflict Early Warning and Response Mechanism in IGAD and the VAA, CSC, and Regional Early Warning Center EW in the SADC are exemplary of successful information net- works. Most of the other systems are either in their infancy or limited to specific programs. For these systems to be effective, they require policy, legal, institutional, and operational mechanisms in place for coordinating, streamlining, and harmonizing the activities of these systems to optimize their performance efficiently and effectively. Inevitably, disparities exist in data and information exchange systems and processes between these net- works, which lead to duplication of efforts, wastage of resources, and even undesirable competition. In the SADC, the DRR unit is in the process of developing a disaster management information system portal that will be accessible to users. The need to strengthen the information system has also been expressed in the SADC’s 2017 Regional Disaster Preparedness and Response Strategy. In east Africa, the EAC is working on an initiative to develop climate information sharing, and so is the COMESA. In IGAD, there are efforts to revamp the Regional Integrated Information System (RIIS). Because the major role of the RECs is mainly to integrate national information systems into a coordinated regional information system, the investment is modest. A good example is IGAD’s proposed RIIS whose estimated costs are provided in table 4.9. Table 4.10 assumes that the member states will meet the operating costs during and after project implementation to sustain project benefits. As empha- sized by participants at the validation workshop as part of this exercise, in addition to member states mobilizing resources internally, support from coop- erating partners and PPPs would be required. CHAPTER 4—Strategies for Long-Term Sustainability of Investments 87 TABLE 4.9 IGAD RIIS: Estimated Budget for the Formulation Process of the RIIS Phase Description Cost (US$) Phase 1 Assessment of Information System in the IGAD Region Consultants Fees and local transport for seven National consultants 17,500 (1 per member state) for two weeks Total fees for two international consultants for 30 days 39,000 Travels 11,000 Workshops Seven national workshops (1 per member state) 49,000 One regional workshop to review assessment findings 100,000 Phase 2 Design/Architecture of the RIIS Consultants Fees for two international consultants 19,500 Workshops One regional workshop to review the RIIS design 100,000 Phase 3 Formulation of the RIIS Project Document Consultants Fees for two international consultants 39,000 Workshops One regional workshop to validate the RIIS project document 100,000 Total 475,000 Contingency 25,000 Total 500,000 TABLE 4.10 Roles of Regional and National Funding Mechanisms Regional mechanisms National mechanisms Member states should provide funding for EWS PPPs should be developed as one of the key programs. priorities for sustainable EWSs, for example, the PPPs can build on existing partnerships with telecom Mechanisms should be developed to address service providers to provide the EW information to transboundary issues. their subscribers for free as in Zimbabwe. Centers of leadership, for example, could downscale Government should create the EW units for global models to national levels; and capacity governance, policy, strategy, and coordinating development should be developed. agency for national EWS funding mechanisms. Funding policies for EWS should be developed, harmonized with existing policies, and effectively implemented. Develop policies/protocols for data sharing Instances where economies of scale would benefit from regional funding mechanisms should be prioritized. Source: Validation workshop. Operational Models and PPPs Generally understood as a “long-term contract between a private party and a government agency, for providing a public asset or service, in which the pri- vate party bears significant risk and management responsibility” (World Bank 2012, 11), PPPs are more than just an interesting idea; they are a necessity in the provision of climate services (Snow et al. 2016). The widely held idea that specialized government agencies (NMHSs) have the authority and responsibility for the provision of hydromet forecasting and monitoring services as a government-financed public good is exem- plified by the role played by leading meteorological agencies in the United 88 CHAPTER 4—Strategies for Long-Term Sustainability of Investments States and Europe that provide their basic services at no cost to everyone, including the private sector. This proposition is supported by the accounting of tangible benefits, including the financial returns from revenue collected by the governments from private companies’ development and marketing of derivative products and services based on “free” hydromet information. In addition, indirect financial gains for the government include the reduc- tion of its contingent liabilities from losses derived from uninsured assets, because readily available basic hydrometeorological information allows and promotes the development of new industries such as insurance markets for weather- and climate-related perils such as flood risk and crop failure, which in their absence would compel the government to intervene as the insurer of last resort to compensate uninsured victims. In this context, there are several reasons for involving the private sector in climate and weather information provision. For instance, private sector enter- prises can leverage technical and financial resources for addressing evolving information needs, such as the increasing demand for localized weather infor- mation, for example, by large agribusinesses, allowing the NMHSs to focus on improving their country-level monitoring and forecasting capabilities, as well as providing entirely new services. Thus: • Leveraging on private sector capacity, NMHSs may increase the sustain- ability of the benefits of the climate and weather information services. The increasing access to the Internet provides an opportunity for research institutions, businesses, and individuals to generate weather data using freely available satellite data and by purchasing their own weather stations. • Mutually beneficial collaborative arrangements can be established between the NMHSs and the academic and research institutions as well as with the private sector aimed at expanding the types and quality of weather and climate information and services that can be provided to general public as well as to specialized audiences. There are several PPP operational models. Although Snow et al. (2016) identify four PPP models on a continuum, the five business operating models that are relevant to the ESA countries are those identified by the World Bank (2016) (box 4.2). There is potential for PPPs in the ESA countries (table 4.11) to generate revenue from the sale of weather/climate goods and services to the private sector. This potential is a result of a positive enabling environment for PPPs, as well as effective communication channels to disseminate information. The capacities for NMHSs to engage in PPPs vary across the ESA region, with Tanzania and Zambia being at the level where they can start engaging in PPPs compared with the NMHSs in Ethiopia, Malawi, and Uganda, in which the following challenges are more pronounced: • Limited technical capacity to generate tailored products and services to meet the specific needs of private sector clients • Limited funding from their governments CHAPTER 4—Strategies for Long-Term Sustainability of Investments 89 BOX 4.2 Operating Models of Climate Information Services Providers TABLE B4.2.1 Operating Models Operating models Public Private but Private and departmental not profit profit International Characteristics unit Public body oriented oriented organization Government Directly Indirectly controlled Indirectly Indirectly No, host country control controlled controlled controlled agreements Own legal entity No Partially or fully Yes Yes Yes separate Legal basis Public law Public law Private law Private law Convention Financing State budget, State budget, grants, Grants, own Own Grants grants own revenues revenues revenues Control Direct political Statutes, laws Regulation Regulation Host country mechanism agreements Ministerial Yes Partial No No No responsibility Autonomy No Yes Yes Yes Yes Each of these models is characterized by the following: Public departmental unit. This model is characterized by lack of autonomy, finance by state budget, problem of monopoly, limited and unreliable public financing, weak technical capac- ity, low organization incentives, little freedom or incentive to compete in commercial markets, and poor service delivery. Public body. This model faces less political and hierarchical influence, has more operational and managerial freedom, supplements state budgets with grants and revenue from service delivery, and faces challenges in enforcing payments for services from sister departments. Private but not profit oriented. These are climate adaptation projects and project-based spon- sorships, targeting a limited number of beneficiaries, with limited scope of service provision, and are unsustainable. Private and profit oriented. These operate in the free market, generate their own revenue, enjoy high freedom of autonomy, and are subject to government regulations. Privatization may not offer a solution to effective climate information service delivery. International organizations. These are funded by members and grants and do not generate revenue from climate services. Source: World Bank Group 2016. • Weak network of weather observation stations to generate fine-scale weather data • Inadequate political framework for regulating PPPs During this study, KIs expressed concerns regarding the privatization of climate services, particularly on the risk of further marginalizing vulnerable communities who are unable or unwilling to pay for the climate services that 90 CHAPTER 4—Strategies for Long-Term Sustainability of Investments TABLE 4.11 Potential for NMHSs to Develop PPPs in Weather Products in Some ESA Countries Country Potential for PPPs Ethiopia Opportunities for PPPs in Ethiopia are limited, but private companies are legally required to share climate and weather data with the NMHS. Malawi The PPP Bill and PPPs Commission facilitate PPPs in Malawi. Tanzania There is a dedicated PPP Unit in the country (Tanzania Investment Centre). Uganda There is scope for PPP between the NMHS and the private sector. Legally, the NMHS is the sole provider of weather information in the country. Demand for information has been expressed by the private sector, particularly aviation, construction, and mining. Currently, there is a cost recovery agreement between the NMHS and Uganda’s Civil Aviation Authority. Zambia The Meteorological Bill promotes PPPs in the development of meteorological services, particularly for the following sectors: (a) aviation, (b) agriculture, (c) water resources management, (d) education and research, (e) health, and (f) building and civil engineering. Source: Adapted from Mills et al. 2016. are unaffordable to them. In this regard, there is a need for further exploration of communities’ willingness to pay for specialized climate services across the ESA countries. The establishment of PPPs should not be limited to NMHSs; they can also address broad food security EWS needs. However, for the PPP to be effective at both regional and national levels, these should be supported by legal frame- works, protocols, and MoUs for partnerships to yield maximum benefits. These arrangements should provide guidance and clarity on the benefits of private sector interventions, including their contributions in terms of human and technical resources as well as the products and services to be delivered. In most instances, support will still be required from international development and cooperation partners for the design and strengthening of the enabling environment for effective PPPs with, for example, mobile network compa- nies, water, and electricity utility companies as well as with private enterprises engaged in weather- and climate-sensitive sectors, such as agribusiness, fish- eries, and tourism. CHAPTER 4—Strategies for Long-Term Sustainability of Investments 91 CHAPTER 5 Best Practices and Recommendations for Institutional Strengthening Introduction This chapter presents three cases of good practice that have been drawn from within and outside ESA. The exemplars include good practices on PPPs, food security information systems, and feedback forums to help improve delivery of the EWS. First, this study found that there was limited engagement between the public and private sectors in production, monitoring, dissemination, and utilization of EW information. This is despite the increasing demand for greater involvement of the private sector in the provision of climate services to inform adaptation decisions. The Environmental Analysis and Remote Sensing (EARS) and other emerging entrepreneurs in Africa are explored to highlight the potential benefits of involving the private sector in climate services provision. Second, food security and EW information systems are generally weak at both national and regional levels. Consequently, the users do not receive timely EW information to help them make decisions that will trigger appropriate action. The ASEAN region appears to have successfully set up a regional food security information system that has responded to the needs of the member states. Finally, as the purpose of the EW information is about triggering action on the ground, setting up a framework for users to integrate science-based and traditional climate forecasts adds value to people’s lives. This also helps the producers of EW information to improve ways of packaging EW information to meet user needs. Innovations for Improving Food Security EW: Role of the Private Sector As this study attests, there is potential for the private sector to play a criti- cal role in complementing the public sector efforts to satisfy the demand for climate services in the ESA countries. EARS is one of the private initiatives that provides a sustainable business model that can complement the NMHSs capacity (box 5.1). CHAPTER 5—Best Practices and Recommendations 93 BOX 5.1 Good Practice from EARS Since 2012, EARS has been involved in piloting a project in Senegal using different climate data collection techniques, including satellites, to improve data quality for Weather Index-Based Insurance. Depending on the design parameters, climate, and drought sensitivity of the crop, the index detects when the drought level is such that yield losses are imminent. In that case, a payout for that location is triggered by the index. After the first year, the project demonstrated that there is considerable potential for improve- ment to calibrate the weather indexes by (a) incorporating additional ground data made available by the project; (b) interacting directly with local experts, including NMHS staff; and (c) integrat- ing different methodologies. Background. EARS is a remote sensing and geo-informatics company that has developed its own innovative technology, known as the Energy and Water Balance Monitoring System. This tech- nology is used to derive and map temperature, radiation, evapotranspiration, cloudiness, and rainfall data from Meteosat data. EARS has processed 33 years of Meteosat data on evaporation and precipitation data fields that cover the entire African continent at a 3-km spatial scale, with daily temporal resolution. This Meteosat reception and processing continues in real time at the head office. In addition, EARS receives data from the WMO Global Telecommunication System for calibration and validation purposes. Based on the aforementioned data, EARS has developed systems and services for river flow forecasting, drought monitoring, crop yield forecasting, and crop index insurance. The model. Since 2009, EARS has been developing a microcrop index insurance called the Food Early Solutions for Africa (FESA). This initiative aims to insure farmers against damage caused by drought, extreme precipitation, and winter or night frost. Once farmers have signed up for the insurance, EARS will use the Meteosat receiving system to (a) monitor the index during the growing season; (b) calculate payouts; and (c) report the results. The project was cofunded by the Ministry of Foreign Affairs of the Netherlands as a contribution to achieving the UN Millennium Development Goals. Benefits. FESA has been developing and providing low-cost drought and excessive precipitation insurance in Senegal, Mali, Burkina Faso, Benin, Kenya, Tanzania, Rwanda, Uganda, Malawi, Mozambique, and Botswana. However, as with many microinsurance schemes, the model was found not to be commercially viable without public sector funding. Upscaling and satellite data versus ground weather stations. Through the use of Meteosat-derived data, full spatial coverage and long-time series can be provided in Africa through EARS. These satel- lite-derived data can address the current geographic gaps that exist between meteorological ground stations in Africa. However, the investment and O&M of these satellites are expensive. To provide the long-time series that are required for a proper risk assessment and insurance design, the satellite data need to be complemented by real-time data from on-the-ground weather stations. Relationship with the NMHSs. EARS is a provider of tailored climate information systems and products that are generated from Meteosat-derived data. EARS therefore does not use data from the NMHSs. As a result, the company has only incidental relationships with NMHSs, often with the objective to obtain data that are not available through the WMO Global Telecommunication System to validate satellite weather data. Source: Mills et al. 2016, 27–28. 94 In addition to international companies such as EARS, small to medium entrepreneurs are emerging who have the capacity to develop user-specific weather products to a variety of market segments and disseminate this infor- mation, the result of which would strengthen the weather information value chain. Severe Weather Consult (SWC) in Rwanda is another company that has recently been established to provide weather information services. SWC, for example, has piloted a weather information service in Rwanda following a market survey that identified weather information needs for farmers and city dwellers. As weather data were not available promptly, SWC opted to install a few ground weather stations with which they will combine saturate data. In future, SWC will charge farmers to sustain the system, and they will con- tinuously explore other areas where data could be valuable, such as lightning information. However, involving the private sector in weather information products and services needs some considerations of the following key issues: • The legal frameworks would need to be amended to allow for the par- ticipation of nongovernmental entities in the provision of weather and climate services. In most countries across the ESA region, NMHSs are the sole authoritative providers of basic weather information products and services, and this condition continues to be considered as the most effective scenario. In Ethiopia, for example, the National Meteorological Agency is the sole authority and provider of tailored climate and weather information products. For the legal framework to be amended, it is likely to take some time to liberalize the weather and early warning information services. • Consistent with WMO’s Resolution 40 (Cg-XII), which urges members to strengthen their commitment to the free and unrestricted exchange of meteorological information, the resolution highlights the need to strengthen the role of the NMHS, particularly in regard to the potential impact of commercialization of basic services and the impact of for-profit activities by private sector enterprises. Accordingly, Resolution 40 states the following: considering . . . (4) The continuing requirement for Governments to provide for the meteorological infrastructure of their countries, (5) The continuing need for, and benefits from, strengthening the capabilities of NMSs, in particular in developing countries, to improve the provision of services, . . . (6) The dependence of the research and education communi- ties on access to meteorological and related data and products, . . . (7) The right of Governments to choose the manner by, and the extent to, which they make data and products available domestically or for international exchange, . . . Recognizing further: (1)The existence of a trend towards the commercialization of many meteorological and hydrological activities, (2) The requirement by some Members that their NMSs initiate or increase their commercial activities, (3) The risk arising from commercialization to the established system of free and unrestricted exchange of data and products, CHAPTER 5—Best Practices and Recommendations 95 which forms the basis for the WWW, and to global cooperation in meteo- rology, (4) Both positive and negative impacts on the capacities, expertise and development of NMSs, and particularly those of developing countries, from commercial operations within their territories by the commercial sec- tor including the commercial activities of other NMSs, Reminds Members of their obligations under Article 2 of the WMO Convention to facilitate worldwide cooperation in the establishment of observing networks and to promote the exchange of meteorological and related information; and of the need to ensure stable ongoing commitment of resources to meet this obliga- tion in the common interest of all nations.26 • The installation of new weather stations is unlikely to be always consistent with the WMO standards. One of the approaches to creating this con- sistency for weather stations is to ensure that all installations of weather stations are done by the NMHS. In Tanzania, the government has made it compulsory that weather stations—from both the public and private sector—be installed by the Tanzania Meteorological Agency (TMA) and TMA staff are trained to maintain these stations. However, the challenge here is that many ESA countries have limited capacity to deploy and maintain their own networks. It is therefore unlikely that the NMHSs will have the capacity to manage a commercial operation. As indicated, there is a recognition that the governments have the responsibility for the allocation of adequate resources for the NMHS, which will allow them to properly discharge their institutional responsibilities, including becoming the authoritative source of hydromet monitoring and forecast- ing information to improve the country’s decision-making and invest- ment-planning processes. Currently, most NMHSs in the ESA regions do not have the capacity to compete with the private sector in areas such as staff training, equipment maintenance, or payment of subscriptions for certain specialized services. ASEAN Food Security Information System Southeast Asia is one of the most disaster-affected regions in the world, with the region having experienced two of the world’s mega-disasters during the past decade: the Indian Ocean tsunami in 2004 and Cyclone Nargis, which hit Myanmar in 2008. More recently, the floods in Thailand in 2011 caused more than US$45 billion in damages, and the latest major disaster, super typhoon Haiyan, was the deadliest disaster in 2013, with more than 6,000 fatalities. According to the International Disaster Database, the region accounted for more than 31 percent of all global fatalities from disasters and 8.83 percent of those affected by disasters from 2003 to 2013. Losses related to natural 26World Meteorological Organization, http://www.wmo.int/pages/prog/www/ois/Operational_Information /Publications/Congress/Cg_XII/res40_en.html. 96 CHAPTER 5—Best Practices and Recommendations disasters cost the ASEAN region, on average, more than US$4.4 billion annu- ally over the past decade. ASEAN was founded August 8, 1967, by Indonesia, Malaysia, the Philippines, Singapore, and Thailand. Its goal was to accelerate economic growth, social progress, and cultural development and to promote peace and stability in the region. After the Cold War, its membership expended to 10 by admitting Brunei Darussalam, Cambodia, the Lao People’s Democratic Republic, Myanmar, and Vietnam. Located in Jakarta, Indonesia, and led by the Secretariat, ASEAN consists of four major departments: Political and Security Community, Economic Community, Sociocultural Community, and Community and Corporate Affairs. In 2012, the Secretariat had a budget of US$15.7 million and 260 staff. To fulfill its objectives, one of the initiatives of ASEAN was to create the ASEAN Food Security Reserve (ASFR) in 1979. Following the establishment of the ASFR, several enterprises were also initiated. This included an agency to manage the regional emergency rice reserve, assess in phases the level of food security in the Southeast Asian region, and prepare information about the food security development policy in member states. This initiative led to the setting up of the ASEAN Emergency Rice Reserve for use during crisis. The term emergency was used to refer to the states or situations of extreme suffering resulting from disasters of a natural or anthropogenic nature. Some disasters would overwhelm the member states’ capacity to address such cir- cumstances through national reserves and also leave them unable to cater to the supply needs through normal trade channels (Briones 2011). Two decades later in 2003, the ASEAN Food Security Information System (AFSIS) was initiated. Today, AFSIS has become a critical element of the ASEAN nations because it is the basis of EWS information. AFSIS was born out of the agreement at the first meeting of the ASEAN Ministers of Agriculture and Forestry and the Ministers of Agriculture of China, Japan, and South Korea, held in 2001. The aim of AFSIS is to provide accurate and reliable information in a timely manner, which is necessary for food security in the ASEAN region. To achieve the objectives, AFSIS is supported by two main components: Human Resource Development and Information Network Development. First, the Human Resource Development is aimed at raising the technical capacity of member states at two levels. The first level was achieved through a combination of activities such as training, workshops, and national seminars. These activities were planned to provide related personnel in member states with knowledge and skills in statistics and the development of food security information systems to ensure that they will be capable of implementing the project’s activities competently and at the same standard. At the second level, the project introduced the activity for mutual technical cooperation to replace the national seminars of the first level. The objective was to share the knowledge and views between the ASEAN member states through the organization of training courses and technical visits. Under this CHAPTER 5—Best Practices and Recommendations 97 activity, the more advanced member states were requested to help improve the capacity of the others to accomplish the implementation of the project’s activities. Third, the Information Network Development component focused on the development of a regional food security information network. This included the development of the database to enable stakeholders in member states to disseminate and acquire food security–related data required for policy planning and implementation. This component also planned to provide member states and the project management units with the sets of computer hardware including printers and necessary application software, as well as annual operating costs to develop information networks at national and regional levels. Having built the technical capacity and the appropriate physical infrastruc- ture, the project emphasized the enrichment of databases and data analysis. The development of Early Warning Information and Commodity Outlook was included to monitor and analyze food security in the region. In addition, the project planned to provide network equipment to some member states as considered necessary for project implementation. The expected outputs of the project were as follows: • Member states will be capable of providing accurate, reliable, and timely information required for the construction of regional food security infor- mation at the same standard. • The project will be able to provide complete information needed for plan- ning and implementation of food security policy in the region. • The development of EW information and commodity outlook will facili- tate the management of food security policies and programs. These activ- ities will help assess food security situations in the region and identify the areas where food insecurity is likely to occur as well as the degree of seriousness. • The responsible agencies will be better aware of the problem in food security, so that affected people will receive more robust responses and support. AFSIS InsƟtuƟonal Arrangements The key institution, which oversees the implementation of AFSIS, is a focal point agencies group from every member state. This group is responsible for managing and assisting in carrying out the project’s activities. The functions are organized through the focal point meeting (FPM). The FPM is the decision-making mechanism of the project and is held at least once a year to discuss, review, and decide on the following: • Work plans of the project • Implementation of project activities at the regional and national levels • Other important matters relevant to the implementation of the project 98 CHAPTER 5—Best Practices and Recommendations The ASEAN Food Security Information and Training Centre, in the Ministry of Agriculture and Cooperatives in Thailand, is the secretariat of the FPM and also doubles as a center for regional training and as a hub of infor- mation on regional food security. Food Security InformaƟon Database Figure 5.1 summarizes the food security database and the benefits to member states and other users. Lessons for ESA Countries from the AFSIS Project • AFSIS is a legally binding policy framework for cooperation, coordi- nation, technical assistance, and resource mobilization in food security information systems in the ASEAN member states. If the ESA countries should adopt a similar system, then they need to consider developing a legally binding agreement. • The AFSIS FPM is made up of the respective food security focal points to oversee the implementation of the program. The ESA countries could adopt the same framework perhaps through the already existing work- ing groups, for example, FSNWG in IGAD and Regional Early Warning Unit (REWU) in the SADC. However, these systems will need to be institutionalized. The resources for AFSIS are mobilized by member states, and if the ESA countries would like to sustain the benefits of such a project, then they should mobilize resources internally. FIGURE 5.1 Contents of AFSIS Database and Benefits to Users Source: AFSIS. CHAPTER 5—Best Practices and Recommendations 99 Participatory Scenario Planning for Coproducing User-Based Climate Services Application of climate and weather information at the local community level is one of the key bottlenecks that has been highlighted by participants of this study. At the local level, for example, in Mbeta Island in Zambia, it was clear that access to weather information is limited and, if available, is often not specific enough, not easily understood, or not framed as actionable decision points. In Zimbabwe, climate information from meteorological services is often viewed as of high quality and based on science and technology, but this information is of limited value because it has limited interface, if any, with local knowledge from users, particularly farmers. Participatory scenario planning (PSP) is one of the tools that provides an opportunity to integrate science-based climate information and context-spe- cific local knowledge (box 5.2). Successful adaptation and resilience to climate variability and change requires building people’s capacity to continu- ously make adaptive decisions that review, anticipate, and have the flexibility to respond to climate risks, uncertainties, and opportunities. To make good decisions, the government at different levels, organizations, and communities need context-specific information on the climate and its uncertainties, poten- tial climate impacts, and response options. Quality Assurance Measures and Service Improvements through Producer-User Interface Forums One of the criticisms leveled against producers of the food security EW information is not only limited understanding of user needs but also feed- back from users. It is then critical for the EW producers, including NMHSs, agriculture departments, UN agencies, NGOs, and private sector providers to regularly assess service quality by evaluating current products and services, as well as using surveys and focus group interviews. Such assessments are likely to increase understanding of gaps in service provision in terms of who is accessing information, how that information is being used, and the expe- rience of users in matching that information to their specific needs as well as exploring potential demand for new services. Srinivasan, Rafisura, and Subbiah (2011) outline the benefits of creating a user platform to provide a two-way dialogue between producers and users of EWSs as one of the ways of reaching climate information users. Climate forums were initiated by the Asian Disaster Preparedness Centre in several countries in Asia, including Mongolia, Myanmar, Sri Lanka, and Vietnam. The climate forums helped build institutional mechanisms for communicat- ing forecasts to users and for allowing feedback from users to providers. It is a regular two-way dialogue and multistakeholder process of understanding and applying climate information. The climate forum is usually initiated and 100 CHAPTER 5—Best Practices and Recommendations BOX 5.2 Application of Climate Information through PSP Participatory Scenario Planning (PSP) is a method used by CARE International under its Adaptation Learning Program (ALP), implemented in Africa, for the collective sharing and interpretation of climate forecasts. The ALP supports communities and local governments to use seasonal climate forecasts and information on climatic uncertainty for decision making as part of the community-based adaptation approach. FIGURE B5.2.1 Five Major Steps of the PSP Process Step 1. Designing Step 2. Preparing Step 3. Facilitating Step 4. Step 5. Feedback, the PSP process for a PSP workshop a PSP workshop Communicating monitoring and advisories from a evaluation Developing an Involve all relevant Multistakeholder PSP workshop Two-way informed, locally stakeholders, forum—access, communication & relevant, and recognizing their understanding, & Reaching all actors feedback between appropriate PSP roles, and utilizing combining who need to use producers, process, including their specific meteorological & the information, intermediaries, and deciding the level knowledge and local seasonal in understandable users of climate (national, county/ capacities to enable a forecasts; language, accessible information enabling province, district participatory process interpretation into channels, & in continuous, iterative, and so on) at which that is responsive to locally relevant good time to inform & shared learning on to conduct PSP & user needs. and actionable decisions & plans. climate information forming partnerships information for services. for sustainability. seasonal decision making & planning. PSP is an iterative learning process Source: http://careclimatechange.org/wp-content/uploads/2017/03/ALP-PSP-Brief-2017.pdf. The PSP method creates space for meteorologists, community members, local government depart- ments, and NGOs to share scientific and traditional local knowledge. It allows these stakeholders to find ways to combine and interpret these two sources of information into locally relevant and useful forms. Participants of the PSP method consider the probabilities of changes in the climate; assess their likely hazards, risks, opportunities, and impacts; and develop scenarios based on such an assessment. They discuss the potential implications of these scenarios on livelihoods, which lead to agreement on plans that respond adequately to the identified levels of risk and uncertainty. For example, in Kenya, the PSP method helped local communities make local agricultural deci- sions by prompting consideration of the different types and varieties of crops that would respond to the different levels of risk. Decisions were made about what crops to plant in the coming season and, crucially, how much of each crop type and variety to plant so as to spread the risk of total crop loss because of whatever climate that actually occurred. Plans were also made about risk reduction strategies that needed to be put in place by communities and how local government and NGOs could support these strategies through their ongoing and planned activities. Source: https://unfccc.int/files/adaptation/nairobi_work_program/application/pdf/care_psp_indigenous_knowledge.pdf (accessed June 13, 2017). CHAPTER 5—Best Practices and Recommendations 101 facilitated by the national hydrometeorological agency and includes partici- pation by different stakeholders from the public and private sectors. The cli- mate forum draws upon Indonesia and the Philippines as they have extensive experiences in institutionalizing climate forums at national and local levels. In March 2003, the Philippine Atmospheric Geophysical and Astronomical Services Administration initiated the national-level climate forum in the aftermath of two severe El Niño events (1997–98 and 2002–03) that demon- strated the need to establish a platform for discussing climate forecasts. The holding of a local-level climate forum was also initiated at the provincial level. In general, the objectives of the climate forum are • To ensure that forecast products, including their uncertainties and limita- tions, are understood by and communicated to users; • To encourage risk mitigation in various climate-sensitive sectors; and • To provide a platform for interagency coordination of policies and pro- grams for dealing with potential impacts of climate-related hazards on a season-to-season basis. Because the countries have different contexts, the mechanics and specific objectives of climate forums are in recognition of the diverse institutional char- acteristics and mandates. To ensure regularity, the climate forum is anchored on a recurring phenomenon, such as the onset of the monsoon. Meetings are convened at least twice a year, just before the onset of the monsoon and then again after the monsoon. The risks and opportunities characterizing each sea- son vary to a great extent; hence it is important that the forecast providers and users meet more frequently to make sure that the nature and timing of risks are well understood. For climate forums to be successful, it is critical that the users must be will- ing to participate and constantly engage the hydrometeorological service. Second, the national hydrometeorological agency should have a good under- standing of the user context. It should have an understanding of the following: • Who are the users and what are their key activities, for example, water resources management, agriculture, disaster management, or public health? • What kind of climate information do they need to carry out their func- tions more effectively and when is this information needed? • Do they have the capacity to receive, interpret, and translate climate fore- casts into sectoral impacts? If not, what capacity-building activities are needed? Having a communication plan for forecast providers and users is essential. There should be regularity in releasing forecast information. Users should also give regular feedback to the national hydrometeorological agencies in the form of postseason reports or other channels. Finally, users, specifically user agencies, must have a mechanism for receiving and utilizing climate informa- tion within their respective organizations. 102 CHAPTER 5—Best Practices and Recommendations The experience in Myanmar indicates that sometimes there is no need to come up with entirely new mechanisms for allowing feedback from users to NMHSs. Some user agencies have a long-standing practice of preparing post- monsoon season reports. Most of the questions relate to the usefulness of information decisions made by users, actions that were taken, and lessons learned (Srinivasan, Rafisura, and Subbiah 2011). • What were the expected and actual climates as forecasted by the hydro- meteorological agency? • What were the options considered or recommended by the department to end users? • What options were adopted as the season progressed? • What were the impacts? • What were the lessons learned? Summary of Findings and Recommendations Although progress has been made, there are challenges that recur across the RECs and member states. Most of these challenges fall into the following three categories: InsƟtuƟonal Challenges • Lack of EWS working groups in both RECs to coordinate EWS activities. In the IGAD region, the institutionalization of the FSNWG has been slow as the FSNWG has yet to be endorsed as an institution of IGAD. • Policies that outline roles and responsibilities for EWS actors at both regional and national levels are generally weak. Although many ESA countries have sector policies, the sectors still operate in silos because of lack of overarching EWS policies. In addition, although several tools such as the IPC have manuals that guide users, these have not been endorsed by regions and member states to provide guidance on systematic data collection, data sharing, monitoring, and agreed action triggers. • Although the GHACOF report focuses more on the actions that users need to consider as compared with the SARCOF reports, which are still expressed in probabilities, some users such as pastoralists are still excluded as the reports are in English. Whereas the start and end dates for the season are useful, there is still a gap in the intensity and frequency of the information and it has yet to reach some of the intended users. • There is a lack of regular updates of the RFBSs and NFBSs and weak monitoring of grain markets, cross-border trade, and commodity price monitoring. Although FEWS NET and the WFP actively monitor these activities, integration of some of these into regional and national systems is still limited. CHAPTER 5—Best Practices and Recommendations 103 Technical Challenges • Although the SARCOF and GHACOF processes are well established in providing regional climate forecast, there are still challenges to downscal- ing these forecasts to local levels such as districts or villages. Another limitation of GHACOF and SARCOF is that they tend to focus on rain and pay little attention to other parameters. • Limited coverage of the weather observation networks and challenges in crop production forecasts makes agrometeorology data less reliable. The capacity of NMHSs in Africa is not adequate and has considerably degraded in some countries during the past 20 to 25 years. • In some countries, there is a lack of technically qualified professionals such as meteorologists, agrometeorologists, and hydrometeorologists to ensure quality hydrometeorological products. • Although the VAC system has become one of the most useful and reliable EWS tools in the SADC, there are multiple methodologies that need har- monization. The IPC procedures allow for the incorporation of the most reliable relevant information from multiple sources. There is a system for weighing the credibility of the source. As the sources of data become more reliable the IPC estimates become more accurate, contributing toward improving the level of understanding of evolving food security situations. Although the IPC protocols continue to gain currency across the ESA regions, there are challenges to its general adoption by national governments, due in part to differences in the manner some countries account for food security outcome indicators, some using actual metrics whereas others rely on proxies of those indicators. • Weak food security information systems and absence of a framework for sharing EWS data at both regional and national levels makes EWS infor- mation slow to reach users. However, it is key to understand that although regional- and national-level EWSs are complementary, these play entirely different roles at their respective geographic scales. Regional-level EWSs are not required to provide guidance on site-specific interventions, whereas the national EWSs are expected to transform risk knowledge into effective on-the-ground interventions for reducing risk and protect lives and assets. Sustainability and Financial Challenges • There are no clear funding mechanisms for EWSs. Although conceptu- ally part of the emergency preparedness of the country, in many nations EWS programs are perceived as part of the emergency response activi- ties. Consequently, funding for EWS tends to be ad hoc, oftentimes com- peting for funds during the emergency response. In addition, because EWSs rely on international assistance, which tends to be project based, they often face financial sustainability challenges once the external funding ceases. However, in general, national government authorities 104 CHAPTER 5—Best Practices and Recommendations and other relevant stakeholders across the ESA regions have a pretty good understanding of the need for effective, timely EWS, as they have joined the Hyogo Framework for Action (2015–15) and the Sendai, and most countries have ratified the Sendai Framework for Disaster Risk Reduction. They have established partnerships with specialized agencies of the United Nations, such UNDP, UNEP, UNICEF, WFP, FAO, and the World Bank, the African Development Bank, EU, and so on. The chronic financial constraints that affect many countries in Africa are the primary reasons why adequate ex ante resources are not allocated for EWS. Most countries are becoming increasingly aware of their contingent liabili- ties and are exploring mechanisms to reduce them, including exploring the establishment of disaster risk financing and transfer strategies. The World Bank’s Africa Program has an ongoing policy dialogue with most countries in the region, aimed to help national governments to find ways to improve regional collaboration and coordination to address food inse- curity through other mechanisms, and to improve EWSs that are ade- quately funded by the government. • There is limited private participation in the provision of climate services. However, it is hypothesized that PPPs could contribute to reducing coun- tries’ reliance on external donors. There is uncertainty, however, over the nature of the relationship between the NMHSs and the private sector. There is also a perception that the private sector may be a threat to data ownership and the ability of governments to fulfill their national and international commitments regarding the provision of basic hydromet services as a public good, including data for EWSs. Recommendations The following are key areas that require investment: Strengthening InsƟtuƟonal Capacity Framework • Strengthen the EWS governance at both the national and REC levels by improving legal and regulatory frameworks, and coordination and ensuring clarity of roles and responsibilities at each EWS phase. This will also include developing common methodologies and procedures for data collection, management, and data sharing across geographic borders. • Develop and strengthen the food security information system at both the national and regional levels to meet the RECs’, African Union’s, and member states’ agendas, including the Comprehensive Africa Agriculture Development Program. The information should serve reg- ular development, EWSs, and emergency preparedness and response. The RECs in the ESA region should consider establishing a food security infor- mation system like that of the ASEAN Food Security Information System (AFSIS) to strengthen the food security EWS. This will be supported by a data-sharing framework and a one-stop food security information hub CHAPTER 5—Best Practices and Recommendations 105 such as an emergency operation center that is accessible to users from governments, the public, and the international community. The regional food security information system should also be replicated in each of the member states. • An effective all-hazards regional technical committee, with a remit on food security EWS, needs to be reviewed, strengthened, or created where one does not exist. This will improve the coordination, commu- nication, packaging, and delivery of information in the region, especially in terms of better joint preparedness and response plans. The mandate should include multihazards and specific hazards such as pest infesta- tions (for example, the FAW). The regional EWS committee should be replicated in the member states. • Develop innovative but carefully designed public-private partnerships that have the potential of improving the sustainability of climate and EWS. PPPs have the potential of breaking the downward trend of relying on external funding, addressing weak infrastructure, deficient services, and low visibility that have a negative impact on efficient and effective EWS services. The increasing access to the Internet has spread rapidly across Africa, which also provides an opportunity to engage the private sector to collect near-real-time data from automatic weather stations and to disseminate this information, alerts, and warnings to improve food and livelihood security. This will require the ESA countries to be proac- tive to engage with private international and local weather and climate information services providers. • Hydrometeorological monitoring and forecasting, and the mainte- nance and operation of the observational networks, will continue to be an activity that is best carried out by specialized government agencies, as a government monopoly, and the services generated by such a monopoly provided as a public good. The privatization of key government functions that have been traditionally carried out by the NMHSs is a contentious issue because establishing the required obser- vational networks is a capital-intensive endeavor that is not, in general, profitable from a business perspective. Considering that private partici- pation is a relatively new territory for the ESA countries, its integration into the existing structures will require a phased approach that can allow for experimentation, testing, flexibility, evaluation, and sharing of lessons learned as the innovations progress. • PPPs shall be supported by legal frameworks and agreements that out- line the roles and responsibilities of the participating partners, data policy, and intellectual property rights to ensure the public goods ser- vice of EWS is not compromised. It is also key to understand that private enterprises will find the required technical infrastructure unprofitable to maintain for periods of time that can span for generations and for future uses that are as yet unknown, or for which the expected return on invest- ment cannot be estimated. Private enterprises would be discouraged from investing in infrastructure to cover geographic areas where, for example, 106 CHAPTER 5—Best Practices and Recommendations they don’t have markets, or if potential markets exist, such markets are unprofitable. However, private entities can benefit from government ser- vices provided at no cost to all. For instance, hydromet information can be used to develop and promote whole new industries, including para- metric agricultural/flood risk insurance, which, in turn, would contribute to reducing the government’s contingent liabilities resulting from adverse hydrometeorological events. In this regard, certain specialized weather services can be taken up by the private sector, building on the basic ser- vices, provided by the government NMHSs. • As the technical capacity of NMHSs improve, these agencies could also provide specialized fee-based services. For instance, in many countries airlines pay for additional weather services at airports. • NHMS will continue to derive its core funding for operation and mainte- nance and research from the government budget. Improve Technical CapaciƟes and Knowledge Exchange • Invest in technical capacity development to enable the collection of high-quality agrometeorological crop production forecasts and vul- nerability data. EWSs require (a) improved capacity to downscale regional climate forecasts to high resolution for the forecasts to be meaningful at the local level; (b) strong weather observation networks with a wider coverage; and (c) improved data collection for crop assess- ments, livestock assessments, and vulnerability assessments. The SADC’s Vulnerability Assessment and Analysis as well as the IPC methodologies should be harmonized, or at least agree on, minimum indicators to ensure quality assurance and comparison between countries. • Improving capacity building is key for ensuring that NMHSs are capa- ble of discharging their mandates. However, the chronic underfund- ing of many, if not most, NHMSs across the developing world—already struggling to stay financially afloat, and consequentially not very effective or relevant for government decision making—makes it hard to persuade financially constrained government officials to invest in future, often- times unknown or undetermined benefits. For this reason, among others, it is critical to come up with ways to demonstrate the current economic value of effective and efficient NMHSs. • Review and strengthen monitoring of the regional food balance sheets and national food balance sheets, grain markets, cross-border trade, and commodity pricing monitoring and agree on action triggers when thresholds are reached. There is a need to develop market monitors for cross-border trade to help monitor prices and volumes to determine the food security levels in the countries. The monitoring and warning service should be supported by regional standard operating procedures (SOPs). • Improve knowledge of emerging threats such as pest infestations (for example, FAW) by strengthening links with local universities and research institutes, for example, collaborating with botany and CHAPTER 5—Best Practices and Recommendations 107 biological sciences, and linking these with the meteorological depart- ments. Improving agricultural productivity goes hand in hand with improved weather and climate forecasting capabilities, which in turn contributes to enhancing the provision of Agricultural Extension Services (AESs). However, AESs do not necessarily materialize without the con- certed effort of key stakeholders from government agencies, NGOs, and academic/research institutions. AESs have a substantial impact in pro- tecting the livelihoods of households engaged in agriculture and, conse- quently, in reducing or mitigating the risk of food insecurity in vulnerable populations. • Promote south-south knowledge exchanges, for example, exchange of information between AFSIS and RECs in the process of developing their Food Security Information Systems. Inter-REC knowledge exchanges such as Agriculture, Hydrology and Meteorology Regional Centre of the countries of the Permanent Inter-State Committee for Drought in the Sahel could also be considered. • Develop and strengthen community-based VACs, including knowledge and best practices sharing between ESA countries. Leverage Available Financial Resources • Strengthen public commitment and mainstream EWS considerations into agricultural/food security policies, budgetary allocations, and planning frameworks.  This will require evidence-based advocacy to regional leaders and cooperation of development partners on the eco- nomic benefits of EWSs. • In Uganda, for instance, the government has demonstrated its awareness of the existing nexus between weather, climate, and water and food inse- curity levels. Uganda has adopted medium-range and long-term strate- gies for reducing the risk of food insecurity, including the incorporation of key concepts of climate-smart agriculture, which depends on improved hydromet monitoring and forecasting capabilities that also contribute to effective end-to-end EWSs. To achieve similar goals across the ESA regions, national governments need to create the enabling environment for policy dialogue that promotes the paradigm shift from responding to food insecurity crises toward managing food insecurity risks, an endeavor that requires a multidisciplinary engagement across multiple sectors and levels of government administration. • There is a growing body of evidence that demonstrates that the societal value of certain government services, including weather and climate monitoring and forecasting services, far exceeds their direct costs (on par with other government-funded services such as public education and health programs). The BCR of investments in hydromet, EWS, and sim- ilar services, generally paid by the national government and provided as a public good, demonstrates the societal gains of providing such services at no cost to the public, including the private sector. However, there is 108 CHAPTER 5—Best Practices and Recommendations potential for implementing effective and efficient cost-recovery measures in the ESA countries, particularly considering that the NMHSs are essen- tially the only agencies providing weather and climate services. • Scale up technical support and investment under schemes such as the World Bank modernization of hydrometeorological infrastruc- ture and institutional strengthening and capacity building of targeted NMHSs with an added emphasis on strengthening subnational insti- tutional capacity. There is also a need to consider leveraging on climate and development finance and build a “weaning phase” into EWS projects financed by development partners to ensure sustainability of benefits. CHAPTER 5—Best Practices and Recommendations 109 References Abrahams, P., T. Beale, M. Cock, N. Corniani, R. Day, J. Godwin, G. Richards, J. Vos, and S. Murphy. 2017. “Fall Armyworm Status. Impacts and Control Options in Africa: Preliminary Evidence Note (April 2017).” CABI. http://www.invasive-species.org/Uploads/InvasiveSpecies/FAW- inception-report.pdf. Adams, Richard M., Laurie L. Houston, Bruce A. McCarl, Mario Tiscareño, Jaime Matus, and Rodney F. Weiher. 2004. “The Benefits to Mexican Agriculture of an El Niño-Southern Oscillation (ENSO) Early Warning System.” Agricultural and Forest Meteorology 115 (2003): 183–94. http:// www.ukm.my/ipi/wp-content/uploads/2013/07/50.2003The-benefits- to-Mexican-agriculture-of-an-El-Ni%C3%B1o-southern-oscillation- ENSO-early-warning-system.pdf. AFSIS. 2017. “ASEAN Plus Three Food Security Information System.” http:// www.aptfsis.org/index.php/statistics-m. Anaman, K. A., and S. C. Lellyett. (1996). “Contingent Valuation Study of the Public Weather Service in the Sydney Metropolitan Area.” Economic Papers 15(3): 64–77. http://onlinelibrary.wiley.com/doi/10.1111/j.1759- 3441.1996.tb00123.x/abstract. Bailey, Rob. 2012. Famine Early Warning and Early Action: The Cost of Delay. London: Chatham House. http://www.fao.org/fileadmin/user_upload /drought/docs/0712pr_bailey.pdf. Basher, R. 2006. “Global Early Warning Systems for Natural Hazards: Systematic and People-Centred.” Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 364 (1845): 2167–82. doi:10.1098/rsta.2006.1819. Black, John, Nigar Hashimzade, and Gareth Myles. 2017. A Dictionary of Economics. 5th ed. New York: Oxford University Press. https://global.oup. com/academic/product/a-dictionary--of-economics-9780198759430? cc=us&lang=en&. Briones, Roehlano M. 2011. “Regional Cooperation for Food Security: The Case of Emergency Rice Reserves in the ASEAN Plus Three.” ADB Sustainable Development Working Paper Series No. 18, Asian Development Bank, Manila. https://think-asia.org/handle/11540/1416. Chambers, R. 1996. Rural Development: Putting the Last First. London: Longman. Coughlan de Perez, E., B. van den Hurk, M. K. van Aalst, B. Jongman, T. Klose, and P. Suarez. 2015. “Forecast-based Financing: An Approach for Catalyzing Humanitarian Action Based on Extreme Weather and Climate Forecasts.” https://www.nat-hazards-earth-syst-sci.net/15/895/2015/nhess-15-895- 2015.pdf. CIMMYT (International Maize and Wheat Improvement Center). 2017. “Multi-Pronged Approach Key for Effectively Defeating Fall Armyworm in References 111 Africa.” CIMMYT. http://www.cimmyt.org/press_release/multi-pronged- approach-key-for-effectively-defeating-fall-armyworm-in-africa/. DMMU (Disaster Management and Mitigation Unit). 2015. Disaster Manage- ment Operational Manual. Lusaka: Disaster Management and Mitigation Unit, Office of the Vice President. East Africa Food Security Outlook, July 2017. https://reliefweb.int/report /somalia/east-africa-food-security-outlook-july-2017. Ebi, Kristi L., Thomas J. Teisberg, Laurence S. Kalkstein, Lawrence Robinson, and Rodney F. Weiher. 2004. “Heat Watch/Warning Systems Save Lives: Estimated Costs and Benefits for Philadelphia 1995–1998.” Bulletin of the American Meteorological Society 85 (8): 1067–73. ECA (Economic Commission for Africa—Subregional Office for Southern Africa). 2011. “Enhancing the Effectiveness of Food Security Information Systems in SADC.” Issue Paper. https://www.uneca.org/sites/default /files/PublicationFiles/enhancing-the-effectiveness-of-food-security- information-systems-in-sadc_issues-paper.pdf. FAO (Food and Agriculture Organization of the UN). 1996. The Rome Declaration on World Food Security. Rome: FAO. ———. 2001. Food Balance Sheets: A Handbook. Rome: FAO. ———. 2017. “GIEWS—Global Information and Early Warning System.” FAO, Rome. http://www.fao.org/giews/country-analysis/en/. ———. 2006. “Assessment of Food Security Early Warning Systems in sub- Saharan Africa.” Policy Brief, FAO, Rome. ftp://ftp.fao.org/es/ESA/poli- cybriefs/pb_04.pdf. ———. 2015. “The Impact of Natural Hazards and Disasters on Agriculture and Food Security and Nutrition: A Call for Action to Build Resilient Livelihoods.” FAO, Rome. http://www.fao.org/3/a-i4434e.pdf. ———. 2017. “Global Report on Food Crises.” FAO, Rome. http://www.fao. org/3/a-br324e.pdf. FEWS NET (Famine Early Warning Systems Network). 2009. “Markets, Food Security and Early Warning Reporting.” Market Guidance No. 6. https:// www.fews.net/sites/default/files/MT_Guidance_Markets%2C Food Security and Early Warning Reporting No 6_En.pdf. ———. 2016. “Southern Africa Special Report: Illustrating the Extent and Severity of the 2015–16 Drought.” http://www.fews.net/sites/default/files /documents/reports/FEWSNET_Southern Africa 2015_16 Drought Map Book_20160318_0.pdf. ———. 2017. “East Africa Special Report: Illustrating the Extent and Severity of the 2016 Horn of Africa Drought.” http://www.fews.net/east-africa /special-report/february-3-2017. Frei, T. 2010. “Economic and Social Benefits of Meteorology and Clima- tology in Switzerland.” Meteorological Applications (2009). https:// www.wmo.int/pages/prog/amp/pwsp/documents/MetAppl_Frei_ MeteoSwiss.pdf. FSIN (Food Security Information Network). (2017). “Global Report on Food Crisis 2017.” http://documents.wfp.org/stellent/groups/public/documents /ena/wfp291271.pdf?_ga=1.107527238.551697735.1492614194. 112 References GFDRR (Global Facility for Disaster Reduction and Recovery). 2016. “Striving toward Disaster Resilient Development in Sub-Saharan Africa.” GFDRR, Washington, DC. https://www.gfdrr.org/sites/default/files/publication /disaster-resilient-development-sub-saharan-africa.pdf. Goretti K.K. M. 2013. “Study on the Natural-Hazards Vulnerability and Risk Profiles in Hotspot Areas as a Support to Early Warning, Disaster Preparedness and Risk Reduction (EWDPRR) Measures in Uganda.” http://www.unesco-uganda.ug/files/downloads/Report%20on%20geo hazards%20in%20Uganda.pdf. Hallegatte, Stéphane. 2012. “A Cost Effective Solution to Reduce Disaster Losses in Developing Countries: Hydro-Meteorological Services, Early Warning and Evacuation.” World Bank Policy Research Paper No. 6058, World Bank, Washington, DC. http://www.wmo.int/pages/prog/amp/pwsp /documents/wmo_1153_en.pdf. ICPAC (IGAD Climate Prediction and Applications Centre). 2017. “Forty Fifth Greater Horn of Africa Climate Outlook Forum (GHACOF 45) Bulletin.” ICPAC, Nairobi. IFRC (International Federation of Red Cross and Red Crescent Societies). 2009. World Disaster Report 2009: Focus on Early Warning and Early  Action. Geneva: IFRC. http://www.ifrc.org/Global/WDR2009- full.pdf. IPC (Integrated Food Security Phase Classification). 2017. “The Integrated Food Security Phase Classification.” http://www.ipcinfo.org/ipcinfo- countries/ipcinfo-southern-africa/en/. IPCC (Intergovernmental Panel on Climate Change). 2014. Climate Change, Adaptation, and Vulnerability: Summary of Policy. Cambridge: Cambridge University Press. https://www.ipcc.ch/pdf/assessment-report/ar5/wg2 /WGIIAR5-IntegrationBrochure_FINAL.pdf. Jacobs, K., and D. A. Sumner. 2002. The Food Balance Sheets of the Food and Agriculture Organization: A Review of Potential Ways to Broaden the Appropriate Uses of the Data. Rome: FAO. ftp://ksph.kz/Chemistry_ Food Safety/TotalDietStudies/FBS_Rev.pdf. Jones, A. D., F. M. Ngure, G. Pelto, and S. L. Young. 2013. “What Are We Assessing When We Measure Food Security? A Compendium and Review of Current Metrics.” Advances in Nutrition: An International Review Journal 4 (5): 481–505. doi:10.3945/an.113.004119. Lumbroso, Darren. 2016. “Building the Concept and Plan for the Uganda National Early Warning System (NEWS) Final Report.” https://mafiadoc. com/queue/building-the-concept-and-plan-for-the-uganda-national- early-warning-_599c78031723dd08400c04a9.html. Makela, Antti, Irma Ylikangas, Ramchandra Karki, Kamal Prakash Budhathoki, Adriaan Perrels, Kristiina Säntti, Mikko Partio, and Matti Keränen. 2011. “FNEP: Finnish-Nepalese Project for improving the meteorological readiness in Nepal.” Journal of Hydrology and Meteorology 8 (1): 72–76. http://soham.org.np/wp-content/uploads/2012/08/8.pdf. MIDIMAR. 2013. “Development of Comprehensive (National and Local) Disaster Risk Profiles for enhancing Disaster Management in References 113 Rwanda Project.” https://www.gfdrr.org/sites/default/files/publication /Development_of_comprehensive_Disaster_Risks_Profiles_in_ Rwanda_Project_Brochure.pdf. Mills, Anthony, Onno Huyser, Olga van den Pol, Kim Zoeller, Dirk Snyman, Nicholas Tye, and Alice McClure. 2016. “Revenue-Generating Opportunities through Tailored Weather Information Products,” UNDP Market Assessment, New York. Murphy, S. J., R. Washington, T. E. Downing, and R. V. Martin. 2001. “Seasonal Forecasting for Climate Hazards: Prospects and Responses.” Natural Hazards 23 (2–3): 171–96 http://link.springer.com/article/10.1023/A:1011160904414. National Drought Management Authority. 2017. “Taita Taveta County Drought Early Warning Bulletin for February 2017.” Nairobi. http://www .ndma.go.ke/component/jdownloads/send/3-taita-taveta/20-taita-taveta- february-2017?option=com_jdownloads. Pilli-Sihvola, K., P. Namgyal, and C. Dorji. 2014. “Socio-Economic Evaluation of Improved Met/Hydro Services in Bhutan.” Finnish Meteorological Institute, Department of Hydro-Met Services, Bhutan. https://www .researchgate.net/profile/Karoliina_Pilli-Sihvola/publication/301286763_ Socio-Economic_Study_on_Improved_Hydro-Meteorological_ Services_in_the_Kingdom_of_Bhutan/links/570f72d808ae38897ba0fa35 /Socio-Economic-Study-on-Improved-Hydro-Meteorological-Services- in-the-Kingdom-of-Bhutan.pdf. Rogers, David P., and Vladimir V. Tsirkunov. 2013. Weather and Climate Resilience. Washington, DC: World Bank. https://www.gfdrr.org/sites/gfdrr /files/publication/Weather_and_Climate_Resilience_2013.pdf. SADC (Southern African Development Community). 2016a. “SADC Regional Humanitarian Appeal.” Gaborone. ———. 2016b. “State of Food Security and Vulnerability in the Southern African Development Community.” Regional Vulnerability Assessment and Analysis Synthesis Report. http://www.sadc.int/files/9014/7911/5767 /SADC_RVAA-August-Final-Web.pdf. Scoones, Ian. 1998. “Sustainable Rural Livelihoods: A Framework for Analysis.” IDS Working Paper 72, Institute of Development Studies, Brighton. Sen, Amartya. 1981. Poverty and Famines: An Essay on Entitlement and Deprivation. Oxford: Clarendon Press. http://staging.ilo.org/public/libdoc /ilo/1981/81B09_608_engl.pdf. Snow, John T., Bonizella Benchwick, Greg Benchwick, George Georgie, Joost Hoedjes, Allan Millaer, and Jeremy Usher. 2016. A New Vision for Weather and Climate Services in Africa. New York: UNDP. https://www.thegef.org /sites/default/files/publications/WeatherAndClimateServicesAfrica.pdf. Srinivasan, G., K. M. Rafisura, and A. R. Subbiah. 2011. “Climate Information Requirements for Community-Level Risk Management and Adaptation.” Climate Research 47 (1–2): 5–12. http://www.int-res.com/abstracts/cr/v47 /n1-2/p5-12/. Swaziland. 2011. “National Disaster Risk Management Policy.” http://ndma. co.sz/wp-content/uploads/2016/04/National-Disaster-Management- Policy-2010-22.06.12.pdf. 114 References Tadesse, Tsegaye, Menghestab Haile, Gabriel Senay, Brian D. Wardlow, and Cody L. Knutson. 2008. “The Need for Integration of Drought Monitoring Tools for Proactive Food Security Management in Sub-Saharan Africa.” Drought Mitigation Center Faculty Publications 32 (January): 265–79. http://digitalcommons.unl.edu/droughtfacpub/1. Tefft, James, Mark McGuire, and Nick Maunder. 2006. “Planning for the Future: An Assessment of Food Security Early Warning Systems in Sub- Saharan Africa.” FAO, Rome. UNDP (United Nations Development Programme). 2016. “A New Vision for Weather and Climate Services in Africa.” UNDP, New York. https://www.thegef.org/sites/default/files/publications/WeatherAnd ClimateServicesAfrica.pdf. UNECA (United Nations Economic Commission for Africa). 2011. “Enhancing the Effectiveness of Food Security Information Systems in SADC.” Addis Ababa: UNECA. UNISDR (United Nations International Strategy for Disaster Reduction). 2006. “Developing Early Warning Systems: A Checklist.” UNISDR, Bonn. ———. 2009. “Terminology on DRR [Disaster Risk Reduction].” http://www .unisdr.org/eng/terminology/terminology-2009-eng.html. United Nations. 2006. “Global Survey of Early Warning Systems: An Assessment of Capacities, Gaps and Opportunities towards Building a Comprehensive Global Early Warning System for All Natural Hazards.” A report prepared at the request of the Secretary-General of the United Nations, Geneva. http://www.unisdr.org/2006/ppew/info-resources/ewc3 /Global-Survey-of-Early-Warning-Systems.pdf. United Nations Office for Disaster Risk Reduction (UNISDR). 2 February 2017. “Early Warning System.” http://preventionweb.net/go/478. USAID. 2012. “Food Security Monitoring and Early Warning Systems: International, regional and national examples, and Indonesia’s applica- tion of best practices.” http://pdf.usaid.gov/pdf_docs/PA00J8PV.pdf. Ververs, M.-T. 2012. “The East African Food Crisis: Did Regional Early Warning Systems Function?” The Journal of Nutrition 142 (1): 131–33. http://www.fao.org/fileadmin/user_upload/drought/docs/The%20 Journal%20of%20Nutrition_EAfrica%20food%20crisis%20(Nov%20 2011).pdf. WMO (World Meteorological Organization). 1992. International Meteorological Vocabulary. Geneva: WMO. https://openlibrary.org/books/OL1136281M /International_meteorological_vocabulary. ———. 2010. “Guidelines on Early Warning Systems and Applications of Nowcasting and Warning Operation.” WMO/TD No. 1559, WMO, Geneva. https://library.wmo.int/pmb_ged/wmo-td_1559_en.pdf. ———. 2012a. “Convention of the World Meteorological Organization International Environmental Agreements (IEA) Database Project.” International Environmental Agreements (IEA) Database Project. https:// iea.uoregon.edu/treaty-text/1947-worldmeteorologicalorganizationentxt. ———. 2012b. “The Guide to Hydrological Practices.” WMO No. 168, WMO, Geneva. http://www.whycos.org/hwrp/guide/. References 115 ———. 2015. “Valuing Weather and Climate: Economic Assessment of Meteorological and Hydrological Services.” WMO, Geneva. http://www .wmo.int/pages/prog/amp/pwsp/documents/wmo_1153_en.pdf. World Bank. 2008. “Weather and Climate Services in Europe and Central Asia; A Regional Review.” World Bank Working Paper No. 151. http://www .preventionweb.net/files/7578_WeatherClimateWBWP151.pdf. ———. 2012. Public-Private Partnerships: Reference Guide Version 1.0. Washington, DC: World Bank. https://openknowledge.worldbank.org /handle/10986/16055. ———. 2015. “Increasing Agricultural Production and Resilience through Improved Agro-Meteorological Services.” World Bank, Washington, DC. http://documents.worldbank.org/curated/en/246621468167041502 /Increasing-Agricultural-Production-and-Resilience-Through-improved- Agrometeorological-Services. World Bank Group. 2016. “Climate Information Services Providers in Kenya. Agriculture.” Global Practice Technical Assistance Paper, World Bank, Washington, DC. http://documents.worldbank.org/curated/en /706021467995075539/Climate-information-services-providers-in-Kenya. 116 References Appendix A Early Warning Systems and Their Attributes Name Host Description Link ICPAC IGAD ICPAC provides (a) timely climate EW http://www.icpac.net/ information and support to specific sector applications for the mitigation of the impacts of climate variability and change for poverty alleviation, management of the environment, and sustainable development; (b) technical capacity building of producers and users of climatic information to enhance the use of climate monitoring and forecasting products in climate risk management and environment management; (c) proactive, timely, broad-based system of information/product dissemination and feedback, at both the subregional and national scales through national partners; and (d) support to maintain quality-controlled databases and information systems required for risk/ vulnerability assessment, mapping, and general support to the national/regional climate risk reduction strategies. CSC SADC The CSC (a) provides operational and http://www.sadc.int/ regional services for monitoring and sadc-secretariat/ predicting extremes in climate condition and services-centres/ (b) develops and disseminates climate-services-centre/ meteorological, environmental, and hydrometeorological products, and hosts the SARCOFs, which are designed to develop regionwide consensus on climate outlooks in the near future. The Real Time Extreme Weather and Climate Monitoring System is the key tool used to gather and visualize all meteorological data for analysis and EW. Regional Integrated RIMES RIMES provides regional EW services and http://www.rimes.int/ Multi-Hazard Early member builds capacity of its member states in the Warning System states end-to-end EW of tsunami and (RIMES) for Africa hydrometeorological hazards. and Asia Locust Watch FAO The Food Security Warning System provides http://www.fao.org/ag/ timely information on the movement of locust locusts/en/info/info/ swarms and the potential impacts these index.html swarms may have on food security. table continues next page Appendix A—Early Warning Systems and Their Attributes 117 (continued) Name Host Description Link AGRHYMET CILSS AGRHYMET is a specialized agency of the http://www.agrhymet.ne/ CILSS of 13 countries. AGRHYMET has established itself as a regional center of excellence in areas/services including (a) agrometeorological and hydrological monitoring at the regional level, (b) agricultural statistics and crop monitoring, (c) regional data banks, (d) management and dissemination of information on the monitoring of natural resources in the Sahel, and (e) strengthening of interstate cooperation through the exchange of technology and methodology. Flood Forecasting Mekong Core River Basin Management Functions are http://www.mrcmekong. System River performed jointly at the regional and national org Commission levels and include (a) data acquisition, exchange, and monitoring; (b) analysis, modeling, and assessment; (c) planning support; and (d): forecasting, warning, and emergency. AFSIS ASEAN AFSIS is designed to facilitate access to http://www.aptfsis.org/ commodity outlooks, EW, and agricultural index.php statistical data at the subnational level from the ASEAN members plus China, Japan, and Korea, Rep. Regional Météo- The Multi-Hazard Early Warning System http://www.meteofrance. Specialized France provides real-time weather advisories and is re/accueil Meteorological responsible for tracking tropical cyclones in Centre La Réunion the southwest Indian Ocean. 118 Appendix A—Early Warning Systems and Their Attributes