Key points are not available for this paper at this time.
Accurate monitoring of cover crop adoption is essential for evaluating conservation investments, yet conventional roadside transect surveys provide incomplete spatial coverage and cannot track field-level persistence over time. We developed a high-resolution remote sensing framework to quantify cover crop adoption dynamics from 2014 to 2024 in the Big Pine Creek Watershed, a conservation priority area in west-central Indiana. Using harmonized 3-5 m PlanetScope imagery acquired in winter (December) and the following spring (April), we trained seasonal Random Forest classifiers with manually interpreted reference polygons and applied a "winter plus following spring" rule to reduce false positives and capture variable termination timing. Field-scale adoption was derived by aggregating pixel classifications to agricultural field boundaries using zonal statistics and majority voting. Polygon-level validation achieved F1-scores of 99. 0-99. 6%, supporting operational field-scale inference. Across the 11-year period, annual adoption ranged from 0. 66% to 12. 49% of eligible corn and soybean cropland, peaking in 2016 alongside a major early funding influx (over 4. 1 million). Adoption was dominated by short tenure: 13. 22% of eligible cropland adopted in only one year and fewer than 3% persisted beyond five years, with longer-term adoption concentrated in the southern watershed. Rotation analysis indicated consistent placement preference, with soybean → cover crop → corn accounting for roughly 40-58% of annual adopted area after 2015. Overall, this satellite-based, field-resolved monitoring approach links adoption extent, persistence, and management context to incentive cycles, providing actionable evidence for targeting support and improving the durability and environmental efficiency of cover crop programs.
Chen et al. (Wed,) studied this question.