Remote Sensing in Southwest Nigerian Rice Fields: Evaluating Yield Variability Assessment Through Monitoring Systems

Remote sensing technologies have been increasingly adopted in agriculture to monitor crop health and yield variability over large areas efficiently. A multi-temporal analysis was conducted using Landsat-8 satellite data to assess yield variability. Ground-truthing involved GPS measurements and field surveys. Satellite-derived Normalised Difference Vegetation Index (NDVI) correlated with ground-truthed yields, showing a strong positive relationship with an R² of 0.75 in the validation dataset. Remote sensing technologies offer promising tools for monitoring yield variability in rice fields but require further refinement and integration into existing agricultural practices. Further research should focus on integrating remote sensing data with local climate, soil, and socio-economic factors to enhance predictive accuracy. Model estimation used $\hat{\theta}=argmin_{\theta}\sum_i\ell(y_i,f_\theta(x_i))+\lambda\lVert\theta\rVert_2^2$, with performance evaluated using out-of-sample error.