Dr. Ritvik Sahajpal (Lead) and Dr. Cesar Izaurralde (Co-Lead) have been awarded a grant by the Foundation for Food and Agriculture Research under the Bill and Melinda Gates Global Grand Challenge Initiative to develop tools and technologies for broad-scale pest and disease surveillance of crop plants in low-income countries. Both are members of the NASA Harvest University of Maryland Hub, with existing projects contributing to the development of food security and sustainability practices from Earth observations.
Crop pests and diseases impact crop productivity, farmer profits and regional and national scale food security. These impacts are especially acute in low-income countries, where heavy usage of crop protection products at low crop productivity rates is not economically practicable and is linked to negative human health outcomes. Monitoring and early detection programs can help mitigate losses by providing in-season, actionable insights on the severity and spatial patterns of crop threats. For this project, Ritvik and Cesar will develop a prototype of a crop threat early warning system that attributes pre-harvest losses to potential crop threats by integrating machine learning algorithms, Earth observation (EO) data and a generic crop pest model to estimate crop conditions. The approach is based on the hypothesis that EO data can be used to assess the combined contribution of both abiotic (soil, temperature, precipitation, solar radiation, etc.) and biotic (pest and disease) factors affecting crop growth and a crop pest model can be used to estimate the fraction of pre-harvest losses that can be attributed to biotic factors. The prototype system will be piloted in Tanzania, which is dominated by small-scale intercropped farming systems, is a major agricultural producer in Sub-Saharan Africa and is highly economically dependent on agriculture, which accounts for nearly a third of its GDP.
This project will generate actionable crop threat warning information at a national scale. This information will be shared with the crop monitor systems operated by the GEOGLAM Crop Monitor team, and in turn, the prototype will be validated using the ground-based information from the Crop Monitor.