Yield estimation of summer maize based on multi‐source remote‐sensing data

Abstract Accurately estimating regional‐scale crop yields is substantial in determining current agricultural production performance and effective land management. The Yuncheng Basin an important grain‐producing area the Shanxi Province. This paper used Sentinel 2A with a spatial resolution of 10 m MODIS temporal 1 d 2020. nonlocal filter‐based fusion model (STNLFFM) was to obtain fused data d, combined Carnegie–Ames–Stanford Approach (CASA) light‐use efficiency achieve summer maize ( Zea mays L.) yield estimation. results showed that normalized difference vegetation index (NDVI) could inherit Sentinel‐2A NDVI details express differences between smaller features more effectively. STNLFFM curve consistent actual growth condition, which accurately reflects trend local abrupt change information during period. Moreover, influenced by topographic artificial irrigation factors, whereas mountainous plateau areas <5,000 kg ha −1 those alluvial plain Sushui River reached 8,000 . accuracy estimation constructed based on (mean absolute percentage error [MAPE] = 5.47%, −13.74% ≤ relative [RE] ≤0.12%) significantly higher than (MAPE 15.65%, −19.67% RE 20.88%), indicating use spatio‐temporal technology can effectively improve accuracy.