Satellite data for Weather Index Insurance-AgricuLtural EaRly warning system (SatWIN-ALERT)

Lead Research Organisation: University of Reading
Department Name: Meteorology

Abstract

For the more than 200 million farmers in Africa who depend on rain-fed agriculture, drought is a matter of life and death. Their vulnerability is often aggravated by a lack of access to risk management tools such as insurance, which also limits their ability to take productive risks on their farms. Index insurance, where compensation is based on weather or area yield metrics, rather than on measured losses, is an affordable alternative to traditional insurance. However, index insurance will only increase resilience to climate shocks if the index insured fairly reflects the risk for both farmers and insurers.

Index design processes are becoming increasingly sophisticated, utilizing multiple data-sources and models. A certain degree of basis risk (where compensation doesn't match observed losses) is, however, inevitable as insurance is not designed to target every risk. Anticipating basis risk that ensues from the complexity in the relationship between meteorological drought (rainfall deficit) and agricultural drought (soil moisture deficit) is a key challenge for the agricultural insurance sector.

We propose a new system, SatWIN-ALERT, which can be deployed in real time to detect and predict basis risk events as they happen. As such, it could unlock in-season actions (such as increased monitoring), or allow timely post-season index assessment or action, enabling aid agencies and governments to anticipate basis risk events, and take action to support farmers who are facing uncompensated agricultural losses. In effect, the facility to predict pay outs and losses bridges the gap between post hoc index insurance and forecast-based finance. SatWIN-ALERT provides a means of combining the data issued by existing platforms into integrated assessments of droughts and subsequent compensation. As such, it can be integrated into well-established and trusted insurance design systems, such as the ARC risk viewer (ARV) or the IRI Social Network for Index Insurance Design (SNIID).

SatWIN-ALERT builds on existing state-of-the-art practices to take a sophisticated approach to basis risk management, allowing index insurance to play an improved role within climate risk management and development. We also bring together novel existing participatory and meteorological research to build systems that are suitable for operational use in ODA countries. To do this, we draw on fundamental research on monitoring of environmental conditions, and on recent improvements in forecasts - especially on sub-seasonal to seasonal time scales, working with established partners to bring results to hundreds of thousands of farmers. We will focus on countries identified as important to our partners, specifically Malawi for our participatory research, plus Nigeria, Senegal, Malawi, Zambia and Ethiopia (amongst others) for our basis risk case studies.

In summary, the proposed project will develop a novel operational system (SatWIN-ALERT), which empowers farmers to benefit from robust financial instruments, based on state-of-the-art models, observations and forecasts. Partnership with leading practitioners in Africa will enable SatWIN-ALERT to sit within existing insurance systems to revolutionise basis risk management and build the resilience of millions of farmers to weather-related hazard.

Planned Impact

Who might benefit from this research?

SatWIN-ALERT is designed to benefit some of the poorest farmers in the world. Immediate beneficiaries are the customers of the practitioner partners:

- R4 reaches over 40,000 farmers in Ethiopia, Malawi, Senegal, Zambia, Zimbabwe, Kenya. Uniquely amongst insurance schemes, R4 works with the very poorest subsistence farmers in Africa.

- Africa Risk Capacity (ARC) has 32 member states, including Malawi, Senegal, Kenya, Gambia, Mauritania, Zambia, Nigeria, Burkina Faso, Niger. It provides pooled risk insurance at a governmental level.

- Risk Shield has insured over 80,000 farmer households in over 30 countries, including Zambia, Nigeria, Kenya, Malawi. It has just agreed a major government contract to insure over 1 million farmers in Zambia.

- Pula consulting provides index insurance to more than 400,000 households in Kenya, Rwanda, Nigeria, Uganda, Ethiopia, and Malawi.

It is intended that these practitioners will act as demonstrators for SatWIN-ALERT and that the system will be taken up by parametric insurance schemes throughout Africa and beyond.

How might they benefit?

SatWIN-ALERT allows insurance practitioners to anticipate situations in which pay outs are inconsistent with the severity of drought. This information unlocks remedial actions, such as providing agronomical support to farmers, market intervention and in severe cases, releasing aid. Such safeguards enable farmers to invest in index insurance, secure in the knowledge that 'misfires' will be managed effectively. Ultimately, well run insurance schemes can be scaled out to new territories and markets - enabling even more farmers to benefit from insurance.

The practitioner partners and their customers will benefit as follows:

- Improved processes for monitoring of drought will enable R4 and Risk Shield to expand their schemes, benefiting more farmers.

- Improved targeting of crop cuts will improve the accuracy of Pula Consulting's products, with direct benefit for existing and new customers.

- Timely information on basis risk events will enable ARC to explain the origin of basis risk events to governments, building trust and facilitating early intervention.

Publications

10 25 50
publication icon
Young M (2020) Optimal spatial scales for seasonal forecasts over Africa in Environmental Research Letters

publication icon
Wainwright C (2019) 'Eastern African Paradox' rainfall decline due to shorter not less intense Long Rains in npj Climate and Atmospheric Science

publication icon
MacDonald A (2019) Groundwater and resilience to drought in the Ethiopian highlands in Environmental Research Letters

 
Description We have:
- developed and validated the TAMSAT-ALERT platform for application to financial services and agricultural decision support (Asfaw et al, 2018; Boult et al. 2021)
- identified and outlined a set of case studies for user applications of the TAMSAT-ALERT platform through a stakeholder workshop held jointly with the TAMSAT-ALERT project
- demonstrated that soil moisture outlooks provide valuable information, both for insurance and for wider applications in the agricultural and financial sectors
- shown that crop failure due to agricultural drought can be anticipated up to 3 months before the end of the growing season
- developed our agricultural drought forecasts into a decision support tool, and scaled this out to farmers served by the One Acre Fund, throughout East Africa

Through our extension project, INFORM, we have:
- integrated TAMSAT-ALERT soil moisture forecasts into Red Cross early action protocols throughout East Africa
- provided county level bulletins for the arid counties in Kenya, warning of worsening drought
- combined our forecasts with vulnerability indices to move towards integrated vulnerability forecasts

Through an associated University consultancy, we have:
- extended our services to Pakistan
- developed a novel NDVI-based crop forecasting system that is highly skillful for the Punjab province
In addition, we have initiated an extensive virtual capacity building programme, consisting of online workshops and user forums.
Exploitation Route - The new soil moisture forecasting system developed in this project will implmented by international organisations, such as the Red Cross and World Food Programme, disaster risk management throughout Africa.
- The decision support tool developed for insurance has been applied to other activities, including decision support that enables farmers to decide when to plant.
Sectors Agriculture, Food and Drink,Digital/Communication/Information Technologies (including Software),Environment,Financial Services, and Management Consultancy,Government, Democracy and Justice

 
Description The findings from SatWIN-ALERT are already been used by the financial agricultural sectors for supporting the development of new products and services, targeted at African farmers. Ongoing successful applications: - agricultural insurance provided to >2Million farmers in southern Africa - planting date decision support tool providing information to >500000 farmers in Kenya and Uganda - early warning protocols developed by the Red Cross will, for the first time, be activated for drought in three countries, based on the forecasting system we developed in SatWIN-ALERT (and previous projects). This could benefit tens of millions of farmers in these countries. - early warning protocols now used by the Kenya National Drought Management Authority for all dry counties in Africa Extension of drought monitoring services to Pakistan - TAMSAT-ALERT system now used for drought anticipatory action by the START Network Disaster Risk Financing (DRF) scheme for three provinces in Pakistan
First Year Of Impact 2019
Sector Agriculture, Food and Drink,Digital/Communication/Information Technologies (including Software),Environment,Financial Services, and Management Consultancy,Government, Democracy and Justice
Impact Types Societal,Economic,Policy & public services

 
Description Exploiting environmental data for food security in Africa: a new rainfall dataset for monitoring and early action
Amount £160,351 (GBP)
Organisation University of Reading 
Sector Academic/University
Country United Kingdom
Start 08/2019 
End 07/2021
 
Description GCRF African Science for Weather Information and Forecasting Techniques (African SWIFT)
Amount £7,971,410 (GBP)
Funding ID NE/P021077/1 
Organisation Natural Environment Research Council 
Sector Public
Country United Kingdom
Start 09/2017 
End 12/2021
 
Description INtegrated FORecasting for Mitigation of risk
Amount £152,782 (GBP)
Organisation UK Department for International Development 
Sector Public
Country United Kingdom
Start 01/2020 
End 12/2020
 
Title TAMSAT-ALERT decision support framework 
Description TAMSAT-ALERT (The TAMSAT Agricultural Early Warning System) outputs community level agricultural risk assessments based on multiple streams of data, including Earth Observation, Reanalysis and meteorological forecasts. In essence, the system addresses the question: 'Given the state of the land surface, the stage in the growing season and the meteorological forecast, what is the chance of some adverse agricultural outcome?'. So far the system has been implemented for seasonal drought risk assessment, planting date decision support and probabilistic crop yield forecasting. 
Type Of Material Computer model/algorithm 
Year Produced 2018 
Provided To Others? Yes  
Impact - Early warning of agricultural drought in northern Ghana during 2017 (system also run in 2018) - Plans for field trials of the planting date decision support 
 
Title Winter wheat yield prediction model for Pakistan 
Description An NDVI-based yield prediction model was developed for the Punjab, Sindh and Balochistan provinces of Pakistan. The model takes high resolution, publicly available NDVI data as input and uses the TAMSAT-ALERT approach to predict the evolution of NDVI at district scale over a season. A machine learning approach is used to relate the NDVI to district crop yields. This enables us to anticipate which districts in the three provicnes are likely to experience low winter wheat yield. The model has proved highly skillful at predicting winter wheat yield from February onwards (winter wheat is planted in November and harvested in April). In 2020-2021 season, the model was successfully implemented for Punjab and Sindh and it was extended to Balochistan in 2021-2022. The model is the basis for the START Networks Disaster Risk Financing scheme for Pakistan winter drought. 
Type Of Material Computer model/algorithm 
Year Produced 2020 
Provided To Others? No  
Impact The model enabled a disaster risk financing scheme to be implemented for Pakistan by the START Network for the first time, with significant impact on the rural populations of Punjab, Sindh and Balochistan. 
 
Description Agricultural decision support 
Organisation One Acre Fund
Country Kenya 
Sector Charity/Non Profit 
PI Contribution We have collaborated with the One Acre Fund to develop a methodology for supporting farmers' decisions on planting date. If field trials are successful, this information will be disseminated to farmers via text message.
Collaborator Contribution The partners have provided extensive datasets of planting date and yield, which have enabled us to carry out an ex ante study of the potential of the decision support to improve yield.
Impact Internal reports on the potential added value of the decision support. These will be developed into field trials.
Start Year 2017
 
Description Collaboration with ICPAC 
Organisation ICPAC
Country Kenya 
Sector Private 
PI Contribution Provision of soil moisture forecasts to support the regional climate outlook forum
Collaborator Contribution Implementation and testing of TAMSAT-ALERT soil mositure forecasts, and provision of forecasts to stakeholders in Africa.
Impact Improved drought early warning for the Greater Horn of Africa
Start Year 2017
 
Description Collaboration with NiMet 
Organisation Nigerian Meteorological Agency
Country Nigeria 
Sector Charity/Non Profit 
PI Contribution Training on processing and evaluation of S2S forecast data, provision of drought forecasts
Collaborator Contribution Validation data and implementation of TAMSAT-ALERT forecasts
Impact Improved drought early warning
Start Year 2017
 
Description Collaboration with the Ghana Meteorological Agency 
Organisation Ghana Meteorological Agency
Country Ghana 
Sector Public 
PI Contribution NCAS/ACREW supports the Ghana Meteorological Agency by providing soil moisture forecasts to inform early action.
Collaborator Contribution GMet provides validation data and staff time for implementation and testing of the system
Impact Implementation of our drought forecasting system within GMet, leading to improved drought early warning and early action.
Start Year 2017
 
Description Decision support for international organisations 
Organisation International Committee of the Red Cross
Department Kenya Red Cross Society
Country Kenya 
Sector Charity/Non Profit 
PI Contribution We have provided international organisations , including the Red Cross and World Food Programme with impact relevant forecasts, which have enabled them to release funds during droughts in Africa.
Collaborator Contribution Our partners have provided a 'route to impact' for our research through their core activities of enabling poor farmers to weather drought.
Impact Provision of impact relevant forecasts has facilitated the release of WFP funds in Zambia, as part of the R4 insurance programme Soil moisture forecasts developed in these research projects have informed the development of early action protocols by the Red Cross
Start Year 2018
 
Description Decision support for international organisations 
Organisation World Food Programme (Italy, Sudan, Senegal)
Country Italy 
Sector Charity/Non Profit 
PI Contribution We have provided international organisations , including the Red Cross and World Food Programme with impact relevant forecasts, which have enabled them to release funds during droughts in Africa.
Collaborator Contribution Our partners have provided a 'route to impact' for our research through their core activities of enabling poor farmers to weather drought.
Impact Provision of impact relevant forecasts has facilitated the release of WFP funds in Zambia, as part of the R4 insurance programme Soil moisture forecasts developed in these research projects have informed the development of early action protocols by the Red Cross
Start Year 2018
 
Description Partnerships with the insurance industry 
Organisation African Risk Capacity
Country South Africa 
Sector Charity/Non Profit 
PI Contribution 1. The Investigators worked closely with the financial sector to improve insurance provision for African farmers. This has resulted in extending insurance to more than 1M farmer households in Zambia. 2. The Investigators have also collaborated with the Africa Risk Capacity to incorporate NCAS/University of Reading datasets into the Africa Risk Viewer (ARV), and supported ARC with the use of these data for informing the release of aid across Africa. 3. The Investigators have provided software to PULA Advisors for objective diagnosis of the onset of the rainy season in southern Africa. This has improved the quality of PULA's replanting insurance products. 4. Drought predictions and drought monitoring information, provided by the Investigators, are being incorporated into the World Food Programme seasonal monitoring bulletins for southern Africa.
Collaborator Contribution The partners have co-developed datasets and products (as described above), and provided feedback on their utility for decision making.
Impact Over 1 million farmers are now insured in Zambia using TAMSAT data and products. These farmers received pay outs that have helped them weather several years of poor weather.
Start Year 2015
 
Description Partnerships with the insurance industry 
Organisation INFORM GmbH
Country Germany 
Sector Private 
PI Contribution 1. The Investigators worked closely with the financial sector to improve insurance provision for African farmers. This has resulted in extending insurance to more than 1M farmer households in Zambia. 2. The Investigators have also collaborated with the Africa Risk Capacity to incorporate NCAS/University of Reading datasets into the Africa Risk Viewer (ARV), and supported ARC with the use of these data for informing the release of aid across Africa. 3. The Investigators have provided software to PULA Advisors for objective diagnosis of the onset of the rainy season in southern Africa. This has improved the quality of PULA's replanting insurance products. 4. Drought predictions and drought monitoring information, provided by the Investigators, are being incorporated into the World Food Programme seasonal monitoring bulletins for southern Africa.
Collaborator Contribution The partners have co-developed datasets and products (as described above), and provided feedback on their utility for decision making.
Impact Over 1 million farmers are now insured in Zambia using TAMSAT data and products. These farmers received pay outs that have helped them weather several years of poor weather.
Start Year 2015
 
Description Partnerships with the insurance industry 
Organisation Pula Advisors
Country Switzerland 
Sector Private 
PI Contribution 1. The Investigators worked closely with the financial sector to improve insurance provision for African farmers. This has resulted in extending insurance to more than 1M farmer households in Zambia. 2. The Investigators have also collaborated with the Africa Risk Capacity to incorporate NCAS/University of Reading datasets into the Africa Risk Viewer (ARV), and supported ARC with the use of these data for informing the release of aid across Africa. 3. The Investigators have provided software to PULA Advisors for objective diagnosis of the onset of the rainy season in southern Africa. This has improved the quality of PULA's replanting insurance products. 4. Drought predictions and drought monitoring information, provided by the Investigators, are being incorporated into the World Food Programme seasonal monitoring bulletins for southern Africa.
Collaborator Contribution The partners have co-developed datasets and products (as described above), and provided feedback on their utility for decision making.
Impact Over 1 million farmers are now insured in Zambia using TAMSAT data and products. These farmers received pay outs that have helped them weather several years of poor weather.
Start Year 2015
 
Description Partnerships with the insurance industry 
Organisation World Food Programme (Italy, Sudan, Senegal)
Country Italy 
Sector Charity/Non Profit 
PI Contribution 1. The Investigators worked closely with the financial sector to improve insurance provision for African farmers. This has resulted in extending insurance to more than 1M farmer households in Zambia. 2. The Investigators have also collaborated with the Africa Risk Capacity to incorporate NCAS/University of Reading datasets into the Africa Risk Viewer (ARV), and supported ARC with the use of these data for informing the release of aid across Africa. 3. The Investigators have provided software to PULA Advisors for objective diagnosis of the onset of the rainy season in southern Africa. This has improved the quality of PULA's replanting insurance products. 4. Drought predictions and drought monitoring information, provided by the Investigators, are being incorporated into the World Food Programme seasonal monitoring bulletins for southern Africa.
Collaborator Contribution The partners have co-developed datasets and products (as described above), and provided feedback on their utility for decision making.
Impact Over 1 million farmers are now insured in Zambia using TAMSAT data and products. These farmers received pay outs that have helped them weather several years of poor weather.
Start Year 2015
 
Description Capacity building workshop 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact We held a three-week virtual workshop to build the capacity of professional meteorologists in Africa to use TAMSAT drought monitoring products.
Year(s) Of Engagement Activity 2020
 
Description International workshop on TAMSAT-ALERT climate services for the insurance industry 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Third sector organisations
Results and Impact Attendees at the workshop included the major players in African drought insurance and climate services, from the World Food Organisation (R4), Pula, Risk Shield, Blue Marble, CIMMYT, One Acre Fund , the Ghana Meteorological Service and the Africa Risk Capacity. The chosen organisations were a mix of international, national and third sector organisations. This is the full list of attendees:

Emily Black (overall PI) University of Reading
Rahel Diro (co-PI, participatory lead) IRI (Columbia)
Helen Greatrex (co-PI) IRI (Columbia)

Federica Carfagna ARC
Tom Philp Blue Marble/XLCatlin
Thabbie Chilongo Center for Agricultural Research Development (CARD) - LUANAR
Peter Craufurd CIMMYT
Michael Tanu Ghana meteorological service
Patrick Lamptey Ghana meteorological service
Eric Asuman Ghana meteorological service
Dan Osgood IRI (Columbia)
Markus Enekel IRI (Columbia)
Bristol Powell IRI (Columbia)
Melody Braun IRI (Columbia)
Lisette Braman IRI (Columbia)
Step Aston One Acre Fund
Steven Kogo PULA Advisors
Rose Goslinga PULA Advisors
Kalie Gold PULA Advisors
Agrotosh Mookerjee Risk Shield
Ross Maidment University of Reading
Matthew Young University of Reading
Katie Cooper University of Reading
Daniela Cuellar WFP/R4
Hussein Madih WFP/R4
Jyothi Bylappa Maralenhalli WFP/R4
William Dick WFP/R4
Mathieu Dubreuil WFP/R4

At the workshop we demonstrated the new TAMSAT-ALERT tools and explored applications. The outcomes were:
- Progress towards applying the TAMSAT-ALERT for supporting planting date for >500,000 farmers (CIMMYT and 1AF)
- TAMSAT-ALERT soil moisture forecasting services for >1M Zambian farmers (Risk Shield)
- Objective rainy season identification code passed to Pula Consultants (reaching >500,000 farmers) (Pula)
- Inclusion of TAMSAT data in the Africa Risk Capacity portal (national level forecast based finance for most of Africa) (ARC)
- New research projects agreed with Ghana Meteorological Agency (GMet)

The workshop will thus hugely expand the ODA impact of the TAMSAT-ALERT system developed during TAMSAT-ALERT and SatWIN-ALERT.
Year(s) Of Engagement Activity 2018
 
Description TAMSAT user forum 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact We held a user forum for those engaged in using satellite-based data and products for climate services in Africa.
Year(s) Of Engagement Activity 2020