Evaluating Yield Variability of Corn and Soybean Using Landsat-8, Sentinel-2 and Modis in Google Earth Engine

Accurate estimation of crop yield before harvest is critical for sustaining agricultural markets and ensuring food security. Remote sensing data have been demonstrated as a useful tool for estimating crop yields. Previous studies show that high spatial and temporal resolution vegetation index from remote sensing data fusion better explained spatial variability of yield in central Iowa. This paper extends the study area over 10 major agricultural states in the United States and further examines the added value by using Landsat-8, Sentinel-2 and MODIS images from 2016 and 2017. The large area data processing with Landsat and Sentinel-2 has been enabled by using the Google Earth Engine (GEE) technology. Results show that the combination of Landsat-8 and Sentinel-2 provides spatial variability of yield at field scales and explains yield variability better than using single data source alone even at the state-level.