Abstract:
In this paper P-3 (Pearson-3) and LP-3 (Log-Pearson-3) distributions are fitted on the data of flood peaks at Guddu, for 38 years (1962-1999), Using three methods of estimation namely MOM (Method of Moments), KLM (Maximum Likelihood Method) and PWM (Power weighted moment). LP-3 distribution appears to be appropriate for modeling the flood data on the basis of x2-test, S-K (Simirnov-Kolmogrov) test, Cs (Coefficient of Skewness), RMSE (Root Mean Square Errors), PPCC (Probability Plot Correlation Coefficient) and L-Moment Ratios. MLM is found to be the efficient method. 1.52 million cusecs flood is expected to arrive at Guddu barrage, during next 100 years, when LP-3 distribution is fitted by MLM, i.e. 37% more than its designed capacity of 1.1 million cusecs.
Page(s):
331-338
DOI:
DOI not available
Published:
Journal: Mehran University Research Journal of Engineering and Technology, Volume: 26, Issue: 4, Year: 2007