Fighting hunger from space: 3 innovative start-ups using AI and Earth Observation
Discover three innovative start-ups that participated in the EO & AI for SDGs Innovation Initiative — a joint initiative by the World Food Programme Innovation Accelerator and the European Space Agency Φ-lab.
By Amine Baha, Fiona Huang and Gulia Rakhimova
The Earth Observation (EO) & Artificial Intelligence (AI) for SDGs Innovation Initiative, jointly launched by the World Food Programme (WFP) Innovation Accelerator and the European Space Agency Φ-lab (ESA) in 2021, has gone through a very fruitful journey. The programme has supported three start-ups to design and test their solutions with real users and WFP stakeholders in various countries. In this blog post, we share the lessons learned and the latest updates from three innovative projects that participated in the programme.
Human-centred design is essential to creating truly impactful products
Earth observation and artificial intelligence are powerful technologies that can help us advance the work toward food security. With increasingly available open-source machine learning models and access to high-resolution satellite images, we can set up algorithms to process satellite data more efficiently. The true value, however, lies in creating solutions derived from users’ needs. That is why, from the beginning of the EO & AI for SDGs Innovation Initiative, we took the time to explore problems with relevant stakeholders, including ESA, WFP Country Offices, and WFP programme units focusing on climate resilience and monitoring and evaluation. Based on these findings, we identified three start-ups with the technical expertise, flexibility, and coachability to refine their solutions to meet humanitarian and development needs. The programme launched with a four-day workshop where stakeholders co-designed the pilot plans and solutions with selected start-ups.
All start-ups worked on new ideas or incremental features by applying the human-centred design approach throughout the EO & AI for SDGs Innovation Initiative. Read on to learn more about these innovation projects and their achievements.
IBISA in Formentera, Guatemala
IBISA, an insur-tech start-up based in Luxembourg, worked with the WFP Guatemala Country Office to design and test a new Forecast Index Insurance for smallholder farmers in the Chiquimula region of Guatemala. This is the first-ever forecast-based insurance model that triggers action based on climate forecasts and observations. It will enable smallholder farmers to receive two anticipated payouts ahead of a forecasted period of 25 consecutive lack-of-rainfall days and 35 dry days and thus take action to prepare and secure the harvest from damage.
Based on ECMWF ERA5, a dataset providing hourly estimates of a large number of atmospheric, land and oceanic climate variables, and ten-day forecast data and field visits, the insurance product can be sold at US$ 15 per policy. It will provide smallholder farmers with two payouts, first, when there are 25 consecutive lack-of-rainfall days, and second, when there are 35 dry days, amounting to US$ 64 — US$ 77 in payouts. IBISA and WFP are working together to obtain buy-in from stakeholders, including local governments and communities. The team aims to launch the product in 2023.
agriBORA in Hatua, Kenya
agriBORA is an agri-tech start-up that supported WFP’s Farm to Market Alliance (FtMA) programme in managing Farmer Service Centres (FSCs) to monitor and predict crop yield using satellite imagery and on-the-ground data in Kenya. FSCs are one-stop shops retailing inputs, services, information, finance, technology and market connections that help farmers to move from small-scale to commercial farm production.
Within several months, agriBORA onboarded 30 FSCs and 6,775 farmers. The agriBORA platform has been used frequently by FSCs and farmers to communicate with each other. As a result of the pilot, two models for crop yield forecasting and crop classification have been derived based on data aggregated through the platform covering 90 tonnes of agricultural output. The model demonstrated an overall accuracy of 84 percent. The pilot revealed that geotagging data submitted by smallholder farmers has been crucial to achieving the accuracy and performance of the models since this data is usually not readily available to feed the platform. Going forward, agriBORA is looking to build even better predictive features by incentivizing FSCs and farmers to geotag their farms.
Learn more about agriBORA’s journey during the programme.
Hydrological Drought Index
Oxford Earth Observation Limited, a water management start-up based in the United Kingdom, joined forces with WFP to explore the development of a Novel Hydrological Drought Index (HDI). This new index includes precipitation, soil moisture, and surface water availability. It aims to supplement and improve the Standard Precipitation Index, which only includes precipitation to monitor green vegetation growth, as a proxy for food production. We found that in Mozambique and Zimbabwe, the HDI can improve the forecasting of vegetation growth in agricultural areas by an average of 20 percent over a conventional Standard Precipitation Index in the short term. This is because it factors in more complex components of the water cycle, such as surface water runoff and infiltration to groundwater, providing more accurate insights on water availability overall to anticipate dry seasons. WFP aims to validate this new model further and roll it out in drought-prone areas.
“As a technical person, I always want to be developing, experimenting, and coding. For me, spending a full week on project design is quite a lot of time. I ultimately think this time was worth it […] — we got a good idea of the challenges WFP faces and how we could potentially help.”
— Lucas Kruitwagen, CTO, Oxford Earth Observation
Taking stock and looking ahead
Throughout EO & AI for SDGs Innovation Programme, we collected feedback from start-ups, WFP colleagues, and ESA experts that will inform our future innovation programmes in this space. With promising start-ups and top-notch research institutes like ESA, we will continue exploring frontier technologies that can power the efforts to achieve the Sustainable Development Goal 2 — a world without hunger.
Find out more about WFP’s Frontier Innovation Programme
The WFP Innovation Accelerator sources, supports and scales high-potential solutions to end hunger worldwide. We provide WFP staff, entrepreneurs, start-ups, companies and non-governmental organizations with access to funding, mentorship, hands-on support and WFP operations.