Mapping Crop Emergence at Sub-field Scales Using PlanetScope Imagery

Abstract:

Crop emergence is a critical input for modeling crop development and biomass accumulation. It is determined by both environmental factors and each farmer’s operations. In the United States, crop growth stages are reported weekly at state or district (multiple counties) level by the U.S. Department of Agriculture (USDA) National Agricultural Statistics Service (NASS) based on the field observations. However, the crop progress and condition ground data collection are time-consuming, and the summaries at the state and district level do not reflect the spatial variability within the unit.

Remote sensing data have been used to extract vegetation phenology. Global remote-sensing phenology products are available at coarse spatial resolution. For example, the Moderate Resolution Imaging Spectroradiometer (MODIS) land cover dynamic product provides annual global land surface phenology metrics at 500m spatial resolution since 2001. The remote-sensing phenology products usually are available after the end of the growing season. Although they are successful in mapping annual vegetation phenology, they are not designed for the within-season (or near real-time) mapping at the sub-field to field scales.

Recently, we have developed a within-season emergence (WISE) approach to mapping crop emergence dates at sub-field scales using satellite observations available during early growth stages. Results from Beltsville Agricultural Research Center (BARC) experimental fields observed in 2019 using the Vegetation and Environment monitoring New MicroSatellite (VENµS) research mission imagery (2-day, 10 m) show that crop green-ups can be reliably mapped within two weeks after emergence. In this presentation, we applied the WISE approach to the PlanetScope harmonized Level-3 (L3H) data product (daily, 3m). The multi-sensor L3H data products have been inter-calibrated and harmonized with MODIS, Landsat, and Sentinel-2 data. The gap-filled L3H NDVI data were used to map crop green-up dates in BARC. Initial results show that remote sensing green-up dates are strongly correlated to crop emergence dates. The detected crop emergence dates from the multi-sensor PlanetScope are comparable to the results from the single-sensor VENµS. The harmonized L3H data product provides daily gap-filled data at a very high spatial resolution, which could be used to map crop phenology at sub-field scales. Nevertheless, the capability of near real-time mapping depends on the latency of remote sensing data and the availability of the latest cloud-free observations.