Rice yield estimation using vegetation indexes derived from Sentinel-2 imagery for sustainable agriculture

Organic farming has emerged as a sustainable alternative in agricultural production, supported by a growing market demand for organic products. Leveraging advanced remote sensing technology, particularly Sentinel-2 optical imagery, this study aimed to estimate rice yield and monitor crop conditions in organic rice fields. Two vegetation indices—Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI)—were evaluated across 30 agricultural parcels in Candipuro District, Lumajang Regency, East Java, Indonesia, using data from two harvest periods. Statistical analyses were performed using several methods such as Spearman’s rho, linear regression, coefficient of determination (R2) and Root Mean Square Error (RMSE). For estimation accuracy, the best model fit achieved using EVI; notably for the second harvest period with R2 = 0.70 and RMSE = 0.03 t ha− 1. However, NDVI performed as well as NDVI but it was slightly less accurate woth R2 = 0.61 and RMSE = 0.05 ton/hectare. Our findings indicate that Sentinel-2-derived EVI can be used to estimate yields in organic rice farming with reasonable accuracy. This approach may support farmers and policymakers in making more informed decisions toward sustainable agricultural practices.