Unmanned aerial vehicle in optimizing nitrogen fertilizer use and estimating yield of okra varieties

Okra ( Abelmoschus esculentus L.), an important vegetable crop, is grown in tropical and sub-tropical regions for its nutritional fruits. Conventional field-based monitoring of okra is time-consuming and expensive. Unmanned aerial vehicles (UAVs) integrated with sensors offers a convenient, low-cost precision agriculture technology with high spatial and temporal resolution. Despite its potential, the technology remains unexplored in okra nitrogen fertilizer optimization and yield estimation. To overcome this limitation, this study assessed spectral responses under varying nitrogen treatments and developed UAV-based multispectral yield estimation model. Two local varieties, V1: LalTeer OK 285 and V2: BARI Okra 2, were treated with four nitrogen treatments. Multispectral imagery acquired between August 17 and October 16, 2023 focusing maximum growth stage. Vegetation indices (VIs), such as NDVI, NDRE, CVI, Cl (red-edge), MTCI, PSRI, and ARI were calculated and used for spectral analysis and change detection. V1 performed better than V2 in all treatments. V1 showed higher sensitivity to nutrient application with yield increases of 83.56 % to 123.66 % under enhanced nitrogen management. Cow dung with 120 kg/ha urea fertilizer achieved 20 % to 84 % higher yields than other treatments. Maximum yield (15.97 ton/ha) occurred with V1 under optimal nitrogen treatment. A stepwise multiple linear regression model was developed using vegetation indices from flowering-stage multispectral data. Best fitted model explained 83 % yield variation (RMSE = 1.82). These results demonstrate the potential for UAV-based yield estimation which can provide valuable insights for timely and precise decision making in okra cultivation.