Identification of potential dam sites using OLS regression and fuzzy logic approach

Abstract In this paper potential dam sites were identified using remote sensing and GIS. Determinant factors viz., precipitation, slope, flow accumulation, soil texture, land use, geology analyzed in the GIS domain. Each factor was reclassified assigned suitable fuzzy membership values depending on their influence site potential. All fuzzified layers overlaid "Fuzzy Overlay" tool platform. Initially, a total of 26 proposed. Only seven selected proximity to nearby dams, settlements, accumulation. The with elevation, stream order, maximum storage capacity, time concentration calculated. determinant subjected ordinary least squared (OLS) regression understand relationship sites. OLS model statistics showed that all are positively correlated except slope (as low slopes more for construction). diagnostics multiple R squares adjusted R-square found be 0.835894 0.872153, respectively. study, Koenker’s (BP) statistic statistically insignificant ( p > 0.01), proving is consistent. Jarque–Bera conducted also 0.01) indicating Gaussian distribution residuals. This proves logic approach coupled powerful deciphering can applied at regional continental scale.