Friday, 16 August 2019

Assessing the Adequacy of Probability Distribution for Extreme Value Analysis of Rainfall and Temperature using Goodness-Of-Fit And Diagnostic Tests


Volume 13 Issue 3 February - April 2018
Research Paper
N. Vivekanandan*
Scientist-B, Central Water and Power Research Station, Pune, Maharashtra, India.
Vivekanandan, N. (2018). Assessing the Adequacy of Probability Distribution for Extreme Value Analysis of Rainfall and Temperature using Goodness-Of-Fit And Diagnostic Tests. i-manager’s Journal on Future Engineering and Technology, 13(3), 1-8. https://doi.org/10.26634/jfet.13.3.14225
Abstract
Assessment of rainfall and temperature for a desired return period is of utmost importance for planning and design of hydraulic structures and also for installation of nuclear power projects. This can be achieved through Extreme Value Analysis (EVA) by fitting of probability distribution to the observed series of meteorological data. This paper illustrates the study on EVA of rainfall and temperature recorded at Mandla observatory adopting Gumbel and Frechet distributions. The annual series of 1-day maximum rainfall, maximum and minimum temperature derived from daily data are used for EVA. The parameters of the probability distributions are determined by order statistics approach and used in EVA. The adequacy of probability distribution is quantitatively assessed by using Goodness-of-Fit (GoF) test viz., Kolmogorov- Smirnov and Diagnostic test, using correlation coefficient and root mean squared error. The GoF and Diagnostic tests’ results show that Gumbel is better suited probability distribution for EVA of rainfall and temperature for the region under study.

No comments:

Post a Comment