The calibration of a rainfall-runoff model

• Main Authors: Yusop, Zulkifli, Ling, Lloyd
• Format: Conference or Workshop Item
• Published: 2015
• Subjects: HD Industries. Land use. Labor
• Online Access: http://eprints.utm.my/id/eprint/61448/; http://10times.com/acee-penang
• Institution: Universiti Teknologi Malaysia

Since the inception of SCS (now USDA) runoff prediction model in 1954, worldwide researchers reported inconsistent study results. Non parametric inferential statistics was used to guide the numerical analysis to search for optimum results for the model calibration in this article. The technique was tested in a case study and a significant improved runoff prediction model was formulated with Nash-Sutcliffe of 0.82 and 16% less residual sum of squares (RSS) than the conventional model. The methodology proposed herewith also addressed the common selection dilemma between mean and median. It identified optimum value used to calibrate the conventional model and also formulated a better runoff predictive model with statistical significance than those by either mean or median. An adjustment equation was also introduced to amend the conventional USDA runoff prediction model.