GEO-AI forecasting of Crops Yield Los Europe using remote sensing and climate data
Faculty of Science
Big Terra Alpha
Timely and reliable estimations of crop production are essential to address climate caused variability, stabilize grain market, and achieving target yields, aligned with sustainable (SDG-2) of ‘zero hunger’ and (SDG-13) ‘Climate Action’. Europe has remained to be the hots events in recent decades, with increased drought’s frequency in western, central, and sout The recent drought of 2022, combined with high temperature impacts in the crop growing sea substantial yield loss in summer crops, causing a total agricultural estimated loss of 13 in EU-27. Since 1992, the Monitoring Agricultural Resources (MARS) Unit of the European Co Research Centre (JRC) has initiated MARS Crop Yield Forecasting System (MCYFS) for accurat and timely crop yield forecasting in EU-27 and neighbouring countries. The system incorpor data sources including agrometeorological from 4000 EU weather stations, WOFOST based crop along with advanced statistical analysis and expert evaluation. Although, existing EU crop system (MCYFS) is well established and operationally mature but exhibits potential limitat capacity to address climate induced future yield loss in EU. The short-term seasonal forec currently adopted system cannot perform yield loss forecasting under climate trajectories SSP scenarios at spatially granular scale. Therefore, this project will urgently provide a solution beyond seasonal to proactive long run yield loss forecasting at NUTS-3 level in f EU states. Moreover, deterministic crop models such as WOFOST with basic agroclimatic inpu well equipped to understand the complex and non-linear relationships between climatic extr response towards changing climate scenarios. From this perspective, current research proje develop a novel Geo-AI framework for high resolution Yield Loss Risk Forecasting (YLRF) of (Maize, Wheat, and Barley) at NUTS-3 scale in Central European countries (Czechia, Slovaki Austria), integrating multisource Earth Observation (EO), and Climate data with crops yiel under present and future (CMIP6-SSPs) climate scenarios. This project leads to develop new Agricultural Drought Indices (AgDI) using EO datasets enabling robust monitoring of agricu across Central Europe. Furthermore, the project will deliver an interactive visualization a structured communication, dissemination, and exploitation strategy to support evidence-b making and climate-resilient agricultural policies aligned with EU priorities. Overall, re will be instrumental in promoting agricultural sustainability by identifying high-risk are climate-resilient crop planning, and informing evidence-based adaptation policies. The dev will serve as scalable decision support tool, bridging the gap between innovative methods implementation in support of national and regional agri-environmental planning. Furthermor can support regional and national authorities in aligning agricultural policies with the E Common Agricultural Policy (CAP) eco-schemes, and long-term climate resilience goals.Sustainable Development Go Meet the Project If you had to explain your project to someone outside your field, how would you describe i sentences? My project explores how climate change and extreme events like droughts are reshaping the crops, including wheat, maize, and barley across Central Europe by combining satellite dat generation AI models. The outcomes will be detailed geospatial maps at the NUTS-3 level, p and robust visual insights into where and why crop losses occur from past to present, enab local-level assessment. Application of advanced crop models with AI algorithms will not on future risks under changing climate scenarios but also provide practical, easy-to-use tool farmers, local communities, and policymakers make smarter and climate-resilient decisions. What fascinates you most about the topic of your research project? My research project directly connects cutting edge geospatial science with real world agri challenges shaped by climate change and European policy priorities. I am particularly inte linking satellite-based observations with climate data across space and time to generate r insights that support climate resilient agriculture. Moreover, application of advanced AI as transfer learning, enables more reliable and accurate forecasting of future yield loss key limitations of traditional statistical methods. Building new drought indices and spati predictions, my project advances towards identifying and forecasting future droughts patte associated Yield Loss Risk at NUTS-3 scale. It directly supports major EU frameworks such Deal and the Common Agricultural Policy (CAP). What makes this project highly compelling i translate complex data into actionable insights through high-resolution maps and predictiv guide farmers, support local communities, and inform policy decisions at multiple scales. of science, technology, and real-world impact is what makes the research both meaningful a How does your research contribute specifically to achieving the UN Sustainable Development My research contributes directly to key United Nations Sustainable Development Goals, part (Zero Hunger), SDG 13 (Climate Action), and SDG 9 (Industry, Innovation and Infrastructure one of the most urgent challenges facing agriculture that is climate-driven crop loss. By satellite-based Earth Observation data with advanced AI models, the project develops high- drought and yield loss risk forecasts for major crops such as wheat, maize, and barley acr Europe. This enables early identification of vulnerable regions at the NUTS-3 level, allow local authorities to take timely actions such as adjusting sowing dates, selecting drought varieties, or optimizing irrigation, thereby reducing crop losses and strengthening food s From a climate perspective, the project provides actionable insights into how extreme even are evolving under future CMIP6 scenarios. These insights support evidence-based adaptatio such as regional crop planning in drought-prone areas of Czechia, Hungary, and Slovakia or management in Austria, directly contributing to climate resilience under SDG 13. At the sa development of an open Geo-AI platform and interactive dashboards advances SDG 9 by transf geospatial data into accessible, decision-support tools for policymakers and agribusinesse Beyond academia, project will directly support local farmers gaining simple, map-based too agricultural decision-making. The local communities would benefit from early warning syste governments can design more targeted agro-climatic policies.
N.B. Funded by the European Union. Views and o are however those of the author(s) only and do not necessarily reflect those of the Europe European Research Executive Agency. Neither the European Union nor the granting authority responsible for them.