Satellite based progressive in-season crop mapping for agricultural scheme governance

Résumé

Abstract Accurate in-season crop information is critical for improving procurement planning and land-use policies in smallholder agricultural regions. Conventional satellite-based crop maps become available only after harvest, limiting their usefulness for real-time decision-making. This study presents a novel progressive in-season crop mapping framework that delivers district-level crop type, sowing windows, and production estimation at three time points (February–April) for 17 districts of Odisha (India) during the 2024–25 Rabi season. Using Google Earth Engine, multi-temporal Sentinel-2 normalised difference vegetation index, unsupervised clustering, and dynamic spectral signature matching, validated with 10 394 ground observations, the method achieved a classification accuracy of 92%. Unlike conventional end-of-season crop mapping approaches, the framework progressively updates crop area, type, and production estimates during the growing season, enabling timely agricultural monitoring and decision-making. The results reveal three distinct rice sowing–harvesting cycles, enabling pixel-level identification of early, normal, and late planting, an advancement over traditional end-season models. District-level analysis shows that 49% of the state’s rice production is harvested in May, indicating a critical period for procurement, storage, and logistical planning. Additionally, extensive rice-fallow and crop-fallow areas were identified, highlighting substantial potential for intensifying cropping systems by incorporating pulses, oilseeds, and climate-resilient crops. The proposed framework offers a scalable approach for real-time agricultural monitoring, enabling government agencies to align minimum support price procurement schedules, optimise warehouse allocation, prioritise irrigation support, and design district-specific crop diversification strategies. This operational, policy-relevant system demonstrates clear potential for integration into state-level digital agriculture missions and climate-resilient food system planning.

Citer ce document

Accès au document

Texte intégral en lecture en ligne, réservé aux abonnés SPHAERO et aux membres de l'institution. Se connecter

Auteur(s)

Statistiques

Consultations : 1

Téléchargements : 0