Aims: Geomorphological parameters are signature for the conversion of rainfall into runoff. The morphometric analysis using GIS gives platform for deriving the geomorphological parameter of the river basin. Due to intercourse between stream network and landscape, geomorphological parameters are driving force for the conversion of rainfall into runoff. For finding parameter which is more significant contribution to runoff, statistical analysis gives better solution. The morphological parameters of the watershed are derived and then use landform equation for getting geomorphological parameters. Derived parameters are subjected for the factor analysis using Minitab 18 software.
Place and Duration of Study: Study is conducted on Thuthapuzha River basin which is tributary of Bharathapuzha River located in Kerala, India has been selected and it is conducted on Department of Soil and water Conservation Engineering, Kelappaji College of Agricultural Engineering (KCAET), Tavanur, between June 2017 and July 2018.
Methodology: From seventeen Micro-watershed the geomorphological parameters are derived and subjected to factor analysis. Interpretation of factor loadings, communalities, variance, % variance, factor score gibes more influencing parameters on runoff.
Results: The total percentage variance explained by all the factors is about 83.9%, which gives the analysis is a good approach. Runoff factor exhibits the highest percentage variance of about 37% among the percentage variance explained by each factor. The highest correlation coefficient is obtained between runoff factor and the length of overland flow, and the constant of channel maintenance which are more significant parameters of river basin. Total variance explained by all three factors is 11.247 where runoff factor exhibits the highest variance of 5.192. Highest factor score is obtained from length of overland flow having factor score 0.190 followed by elongation ratio, form factor having factor score of 0.165 and 0.152 respectively which can be used as independent variable in case of regression models.