The Study of Spatial Effect on Some Variables of Water Pollution in Baghdad Regions

Abstract

The Durban Regression Model Spatial is one of the important models in the applications of many phenomena, such as economic, health, and environmental spatial phenomena, that are crucial to research. The spatial variable, which is additive and significant, was examined in this study. The aim of the research is to estimate the parameters of the Durban spatial regression model using the regular and modified spatial adjacency matrices under the Rook adjacency criterion using the parametric method represented by the Maximum Likelihood Method, as it was employed. The model for the study of water pollution shows the relationship between the dependent variable Y, which is the percentage of water pollution in the city of Baghdad, and the explanatory variables (Temp., DO, BOD5, PO4, NO3, Ca, Mg, TH, K, Na, SO4, PH). In light of the effect of spatial proximity, one of the most important conclusions reached was the emergence of significant effects of certain explanatory variables on the dependent variable Y, and it also demonstrated that the modified spatial proximity matrix M* outperforms the normal spatial proximity matrix M when estimating the Durban spatial regression model in data representation. The estimated values of the dependent variable Y are close to the real values of the same variable.