Small area estimation for spatial correlation in watershed erosion assessment

This article describes the combination of small area estimator and a simultaneously autoregressive (SAR) model applied to the erosion data collected at the Rathbun Lake Watershed in Iowa (USA). The proposed methodology considers an EBLUP estimator with spatially correlated random area effects taking into account the information provided by neighboring areas. The article discusses the gain obtained from modeling the spatial correlation among small area random effects useful in representing the unexplained variation of the small area target quantities. Moreover the estimator of mean squared error of the proposed estimator is presented.