Modeling and investigation of the most important factors affecting the occurrence of floods in the Gorganrood watershed

Abstract

The Objective of this research is to provide a model to calculate flood discharge values with a certain return period by using the geometric features of the watershed and by providing multivariable models.To understand flood occurrence in the Gorganrood watershed, we studied 11 hydrometric stations in Golestan province. Various influential parameters, including physiographic factors, vegetation, land use, soil science, geology, and weather including 31 independent variables were selected in this research. Data analysis and flood modeling were done by the multivariate regression statistical method. In this way, flood regional models using the Enter method have higher accuracy and precision than the models obtained using the Forward method. In general, 13 variables out of 31 independent variables included in the multivariate regression test have been identified as the most important and influential factors in the occurrence of floods in the Gorganrood watershed. The effective variables include: Gravel's coefficient, percentage of the area of permeable formations, watershed slope, concentration time, watershed area, flow velocity, percentage of land area with a north direction, percentage of land area with a south direction, percentage of pasture land area, percentage of agricultural land area, percentage Waterway density is ranked 3, as are average annual temperature and maximum annual rainfall.Conclusions: By using these relationships (models), it is possible to obtain the values of the maximum instantaneous discharge with a return period of 2, 10, 5, 25, 50 and 100 years in the basins without statistics under the sub-basins of Gorganrood.

Keywords

• flood peak discharge
• gorganrood watershed
• modeling
• linear regression