Remote Sensing based Crop Yield Monitoring and Forecasting

Accurate and timely information on rice crop growth and yield helps governments and other stakeholders adapting their economic policies, enables relief organizations to better anticipate and coordinate relief efforts in the wake of a natural catastrophe, and provides technical backbone of an insurance solution where risks of yield losses from the rice smallholders are transferred to the insurance market. Such delivery of rice growth and yield information is made possible by regular earth observation using space-born Synthetic Aperture Radar (SAR) technology combined with crop modeling approach to estimate and forecast yield. Radar-based remote sensing is capable of observing rice vegetation growth irrespective of cloud coverage, an important feature given that in incidences of flooding the sky is often cloud-covered. Rice yield forecast is based on a crop growth simulation model using a combination of real-time and historical weather data and SAR-derived key information such as start of growing season and leaf growth rate. Results from pilot study sites in South and South East Asian countries suggest that incorporation of remote sensing data (SAR) into process-based crop model improves yield estimation for actual yields and thus offering potential application of such system in a crop insurance program. Remotesensing data assimilation into crop model effectively capture responses of rice crops to environmental conditions over large spatial coverage, otherwise practically impossible to achieve with crop modeling approach alone. This study demonstrates the two angles of uncertainties reduction in forecasting crop yield: (1) minimizing model uncertainties, in this case by assimilation of remote-sensing data into crop model to recalibrate model parameters based on remotely sensed crop status on the ground, and (2) minimizing uncertainties in seasonal weather conditions by incorporating real-time throughout the forecasting dates. Key Terminology: Crop Yield Monitoring, Crop Yield Forecast, Remote Sensing, Synthetic Aperture Radar (SAR), Crop Growth Modeling, ORYZA2000