A comparison of vegetation indices for corn and soybean vegetation condition monitoring

The continuous crop condition monitoring with a full geospatial coverage and sufficient granularity throughout the season is critical to decision making in agricultural policy, production, and food prices. The USDA NASS currently uses bi-weekly 1km AVHRR NDVI composited data to monitor the US crop condition in the growing season. To improve both temporal and spatial resolution, 250m daily MODIS surface reflectance data is used and invalid, cloud pixels are reconstructed. Moreover, to compensate the sensitivity to low vegetation area and saturation at the high vegetation of NDVI, this paper proposes a new vegetation index based on NDVI and simple ratio vegetation index as an alternative for crop vegetation condition monitoring. The initial experiments indicate the proposed vegetation index is robust to the low vegetation and sensitive to high vegetation, and has potential to be an alternative to NDVI for crop condition monitoring.