Using Sentinel-1, Sentinel-2, and Planet Imagery to Map Crop Type of Smallholder Farms

Remote sensing offers a way to map crop types across large spatio-temporal scales at low costs. However, mapping is challenging in heterogeneous, smallholder farming systems, such as those India, where field sizes are often smaller than the resolution of historically available imagery. In this study, we examined potential relatively new, high-resolution imagery (Sentinel-1, Sentinel-2, and PlanetScope) identify four major (maize, mustard, tobacco, wheat) eastern India using support vector machine (SVM). We found that trained SVM model included all three sensors led highest classification accuracy (85%), inclusion Planet data was particularly helpful for classifying smallest farms (<600 m2). This likely because its higher spatial (3 m) could better account field-level variations systems. also impact image timing on accuracy, early-season images did little improve our models. Overall, readily Sentinel-1, were able field-scale with high Indian The findings from study have important implications identification most effective ways