Answers are grounded in this paper and cite their sources.
In an agricultural setting, the planting dates (PD) and harvesting dates (HD) define the growing season length. This temporal information aids in monitoring yields, annual greenhouse gas budgets, and resource cycling. However, satellite-based estimation of PD and HD across Arkansas, which produces 47.5% of U.S. rice, remains understudied. Phenological and machine learning approaches can support the development of spatially explicit PD and HD datasets. This study used 662 PD and 500 HD field-year observations collected across rice fields in Arkansas from 2015 to 2024. Four phenological approaches were used to evaluate temporal synchrony between vegetation index trajectories and field-observed PD and HD, while the Lag-Informed Meteorological and Phenological Random Forest model (LIMP RF) incorporated phenological metrics, and gridded meteorological and soil variables as predictors. Phenological metrics associated with end-of-season transitions exhibited stronger synchrony with HD ( r = 0.79-0.82) than start-of-season metrics did with PD ( r = 0.49-0.75), reflecting the stronger spectral signature of senescence and biomass removal relative to early crop emergence. Under random split validation, the LIMP RF framework predicted PD with MAEs of 6.23 in test set, while HD prediction achieved error of 4.23 days, respectively. Under leave-two-years-out validation, model performance declined, with PD and HD test MAEs increasing to 9.63 and 6.89 days, respectively, demonstrating the influence of interannual climatic variability on model performance. Climate-anomaly analyses further showed that HD prediction errors were strongly associated with anomalous vapor pressure deficit and soil temperature conditions across years, while PD prediction errors exhibited weaker climatic sensitivity. The ability to robustly detect PD and HD will contribute to greenhouse gas budget accounting and an understanding of the impacts of climate change on plant growth, phenology, and yield.
No takes yet. Share an insight, caveat, or question.
Mahbub et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: