|
Abstract:
|
Citarum watershed is the largest watershed in West Java Indonesia, and serves as the water supply for communities in some cities around it. Upperstream Citarum watershed serves as a conservation area, therefore, water criticality index in this area should be considered, hopefully in the dry season, the watershed is still able to fulfill the water needs of society along the Citarum river. The objective of this study is to obtain the best model for the water criticality index in the upperstream Citarum watershed using spatial regression models. Based on the result of Moran's I test, spatial autocorrelation in the residual of linear regression model, the dependent variable (water criticality index), and the explanatory variables (population density, the percentage of forest area, farm, and plantations) are significant. The Likelihood ratio test showed that the spatial regression models are better than multiple linear regression model in predicting the water criticality index. Based on the AIC values, the Spatial Durbin model is the best model in predicting the water criticality index in the upperstream Citarum watershed.
|
