Abstract
Multiple regression analysis was conducted to find relations between watershed properties and surface water quality in the Tonegawa River basin and the Arakawa River basin. Data of common bacteria in surface water was prepared from statistics on the water source for water purification facilities during the decade between 1995 and 2011. For the objective variable in the multiple regression analysis, statistical values were calculated from the statistics data including mean and maximum. Cluster analysis was used to provide about 10 appropriate explanatory variables from 132 properties for the multiple regression analysis. As a result of the analysis, we found population, the number of dairy farmers and the number of hen in each watershed had a relation with the mean value of common bacteria. The number of hen has a high relation with the average of annual maximum value of common bacteria. From the results, it can be said that the statistical values of the surface water quality related to pathogenic microorganism index can be expressed by some simple equations.
