A Simplified Rainfall-Streamflow Network Model on Multivariate Regression Analysis for Water Level Forecasting in Klong Luang (KGT.19 Station) Sub-watershed, Chon Buri Province, Thailand

Abstract

A simplified rainfall-streamflow network model based on multivariate linear regression (MLR) analysis has been proposed. To determine significant coefficients of streamflow network, eleven MLR models were examined. The study’s three objectives were 1) to develop a novel a mathematical model based on MLR analysis for forecasting optimal water levels; 2) to determine the most significant coefficient of rainfall-streamflow network among in the area of interest in the vicinity of Klong Luang sub-watershed KGT.19 station; and 3) to apply the optimal MLR model for water level and flood forecasting maps in Klong Luang Sub-watershed. We used Geographic Information System (GIS) and Remotely Sensed Data (RS) data recorded from Klong Luang (KGT.19 Station) sub-watershed, and Phanat Nikhom, Chonburi, Ban Bueng and Phan Thong districts, in Chonburi Province, Thailand. The findings indicated that the MLR based Model No. 8 is the most applicable and effective. The proposed model also could be applied in water level forecasting, water resource management, flood hazard planning, and flood early warning.