Key Takeaways
- Taylor Geospatial, a nonprofit advancing geospatial artificial intelligence for public benefit, has named three awardees under its Geospatial Innovation for Food Security (GIFS) Challenge, each receiving up to $550,000 in funding over an 18-month deployment period.
- The UN World Food Programme and the REACH Initiative are developing Afghanistan's PULSE platform, a system that integrates climate, food security, nutrition, and market data to forecast supply chain disruptions and support humanitarian response in conflict zones.
- A multi-university consortium including Arizona State University, the University of Maryland, and Washington University in St. Louis, in partnership with NASA Harvest and FEWS NET, is building a GeoAI tool that lets analysts query satellite data in natural language to identify early signals of food system instability.
- Researchers at the University of Missouri are developing an open-source “water first” GeoAI model that maps plant-available soil water at sub-field scales to improve nitrogen application decisions for rainfed arable farms.
- All three projects are designed to be adaptable beyond their initial geographies, with methodologies intended for replication across other conflict-affected or agriculturally similar regions globally.
Taylor Geospatial Launches GIFS Challenge Awardees to Close the Gap Between Research and Field Use
Taylor Geospatial, a nonprofit organization focused on advancing geospatial artificial intelligence for global public benefit, has announced three project awardees under its Geospatial Innovation for Food Security (GIFS) Challenge. The initiative was designed to address a recognized gap between the volume of geospatial research produced and the availability of operational tools that humanitarian agencies, governments, and agricultural specialists can actually use under real-world conditions.
Each selected team will receive up to $550,000 in funding alongside expert guidance and support, transitioning from proof-of-concept to full operational deployment over an 18-month period. Projects were selected through a competitive process with external expert review.
“These projects prioritize execution over theory, ensuring that the work functions under the real-world constraints of time, scale, and uncertainty. The GIFS awardees are not just producing research; they are building tools that can be used to manage resources more efficiently and that humanitarian teams in conflict zones can use to identify food system risks before they become crises,” said Rachel Opitz, GIFS program manager at Taylor Geospatial.
WFP and REACH Initiative Build Afghanistan's AF-PULSE Early Warning Platform
The first selected project is led by the United Nations World Food Programme in partnership with the REACH Initiative. The team is developing Afghanistan's PULSE platform (Platform for Understanding Local Shocks and Emergencies), a system that combines climate, food security, nutrition, and market data to provide a comprehensive picture of on-ground conditions. The platform is designed to help responders track hazards affecting food access and supply routes in environments where ground-level data is often incomplete or unavailable.
Using GeoAI models trained on rapid assessment feeds integrated with community feedback, AF-PULSE is built to forecast supply chain disruptions and identify alternate transport routes to support timely humanitarian response. While the initial development focuses on Afghanistan, the methodology is designed to serve as a template for other conflict-affected countries.
“AF-PULSE reduces the risk of hunger and malnutrition escalating into famine-like conditions during conflict and disasters by helping prioritize limited resources and reaching communities sooner – ultimately saving lives. It is a gamechanger for disaster risk reduction and preparedness, providing crucial insights tailored to local contexts,” said Raul Cumba, Head of Research, Assessment and Monitoring for WFP Afghanistan.
NASA Harvest and University Partners Develop Natural-Language Food System Risk Tool
The second project is a collaboration between Arizona State University, the University of Maryland, and Washington University in St. Louis, alongside NASA Harvest, NASA's Goddard Space Flight Center, and the Famine Early Warning Systems Network (FEWS NET). The team is building a GeoAI capability that aligns in-season satellite-based data with natural-language queries, enabling analysts to generate accessible, question-driven insights about food system conditions while explicitly communicating the system's confidence level in its outputs.
As a practical example, a FEWS NET analyst could query “Which fields have been prepared?” and receive a map of likely prepared fields alongside an estimate of coverage and a confidence indicator. The open-source tool is planned for testing in Sudan, Ukraine, Syria, and Haiti, with findings designed to be applicable to any active conflict region.
“Decision-makers often have to assess food security risks with limited and delayed information about what is happening on the ground. By generating more timely and transparent information, it will help address critical gaps and support organizations working to anticipate emerging food security risks,” said Inbal Becker-Reshef, Director of NASA Harvest.
University of Missouri Team Targets Precision Nitrogen Management With “Water First” GeoAI
The third GIFS awardee is led by researchers at the University of Missouri in partnership with MU Extension. The project applies satellite imagery and machine learning to map plant-available soil water at sub-field scales, providing the basis for more accurate in-season nitrogen application decisions. The approach addresses the absence of reliable tools that can dynamically update crop nitrogen requirements based on local conditions throughout the growing season.
The initial model targets rainfed arable farms in the US Midwest, with a focus on claypan soil regions in Missouri and Iowa. The open-source framework is designed to be adaptable to regions with comparable agricultural systems and soils globally.
“There are no reliable tools that dynamically update estimates of crop nitrogen needed throughout the growing season based on local conditions. Advancing a system that does that is a win for everybody – better harvests, fewer inputs, healthier ecosystems,” said Tim Haithcoat, Associate Professor in Data Science and Analytics at the MU Institute for Data Science and Informatics.
“By designing it to travel — to adapt to different soils and seasons — we're building something the global agricultural community can eventually use for precision agricultural management,” added Jasmine Neupane, Assistant Professor of Agricultural Systems Technology at the Digital Agriculture Research and Extension Center.
