Spatial approach for the estimation of average yield of cotton using reduced number of crop cutting experiments

Abstract

In India, cotton yield estimates are done using crop cut-ting experiments (CCEs) conducted within the framework of the general crop estimation surveys (GCES) methodology. In recent times, for obtaining reliable estimates at levels lower than the district, the number of CCEs has increased in comparison to the existing setup of GCES. This puts an additional financial burden on Government agencies. There is a possibility of reducing the number of CCEs under the GCES methodology and predicting the remaining CCE points using an appropriate spatial prediction model. In this article, the predictive performance of different spatial models has been compared. Furthermore, district-level estimate of average productivity of cotton has been determined using the geographically weighted regression (GWR) technique and the results compared with those obtained using the traditional GCES methodology. The proposed spatial estimator of the average yield of cotton obtained using the GWR approach is more efficient and the results are comparable with the estimates obtained using the GCES methodology. The developed methodology can be utilized to reduce the number of CCEs and capture the spatial non-stationarity present in the cotton crop yield.