In-Season Yield prediction using crop modeling, remote sensing, and machine learning Crop and Region of Interest: Soybean for Mato Grosso, Brazil.
- Historical Yield, planted area and production (2011-2022)
- Planting Progress (2015-2023)
- Harvest Progress (2018-2022)
- Weather (2000-2023)
- NDVI (2013-2023)
- Soybean Area 2022 (raster)
- Regions (shp)
Used APSIM Next Generation and the R package APSIMX for running simulations The analysis was done using python. The output of the crop model was combined with ground truth data (weather, remote sensing, planting date) to train a Catboost Machine Learning Model. In-season predictions for yield and harvest progess curve were obtained.
Presentation: Slides