This project evaluates the precision and suitability of rotorcraft-type small unmanned aircraft system (UAS), fixed-wing aircraft (Cessna), and satellite (accessed through the GEOGLAM and Harvest networks) for observing crop stresses and predicting yield of rice (and corn) via integration into a crop simulation model.
This project collects ground, UAS, and airborne data over cotton and sorghum fields in Texas. These ground and airborne measurements are compared with high-spatial resolution satellite data. Spatial resolution of orthomosaic images are 1cm, 10cm, and 10/30m (UAS, Aircraft, and Satellite (Sentinel 2 and Landsat 8, respectively). With the extraction of crop phenotypic data from remotely sensed imagery, there is cross-validation of crop phenotypic data with ground data. All remote sensing data is compared to the ground data. Calibrated satellite data is then used to predict yield using EPIC/APEX models.