Detecting Winter Cover Crop Termination using Harmonized Landsat and Sentinel-2 Satellite Imagery to Support Ecosystem Assessment
Abstract
Cover crops are planted to reduce soil erosion, increase soil fertility, and improve watershed management. In the Delmarva Peninsula of the eastern U.S., winter cover crops are an important component for reducing nutrient and sediment losses from farmland. Cost-share programs have been created to encourage the use of cover crops to achieve conservation objectives. The programs require that cover crops be planted and terminated within a specified time window. Usually, farmers report cover crop termination dates for each enrolled field, and conservation district staff confirm the report with field visits to >20% of fields conducted within 2 weeks of termination. This verification process is labor-intensive, time-consuming, and became quite restricted in 2020 due to the COVID-19 pandemic. In this study, we refined and extended the within-season termination (WIST) algorithm for application to the routinely archived Harmonized Landsat and Sentinel-2 (HLS) dataset over the Delmarva Peninsula, and applied it to fields (n=28,190) enrolled in the Maryland Department of Agriculture (MDA) winter cover crop program for 2020-2021. We extracted an average normalized difference vegetation index (NDVI) value for each low-cloud HLS scene for the enrolled winter cover crop fields. The WIST algorithm then detected and updated termination date estimations based on the NDVI time series. Results were conveyed to MDA on a bi-weekly basis to assist with program management. The estimated remote sensing termination dates were compared to field observations and to farmer-reported termination dates from the MDA database. Our results show that the WIST algorithm detected ~80% of terminations for the enrolled fields. Among the detected terminations, over 70% of termination dates were within two weeks of farmer-reported dates. Remote sensing provides a fast and consistent method to map cover crop terminations over a large set of fields. However, near-real-time monitoring of cover crop terminations is still challenging and can be affected by many factors such as the frequency of clear observations, the latency of satellite data products, data quality and data consistency. We will discuss those factors, the uncertainty of cover crop termination estimation, and the potentials for operational winter cover crop assessment over large areas.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2021
- Bibcode:
- 2021AGUFM.B35P1603G