Modelling Rainfall Amounts using Mixed-Gamma Model for Kuantan District
An efficient design of flood mitigation and construction of crop growth models depend upon good understanding of the rainfall process and characteristics. Gamma distribution is usually used to model nonzero rainfall amounts. In this study, the mixed-gamma model is applied to accommodate both zero an...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
AIP Publishing
2017
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/17708/1/fist-2017-roslinazairimah-modelling%20rainfall%20amount.pdf http://umpir.ump.edu.my/id/eprint/17708/ https://doi.org/10.1063/1.4982831 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | An efficient design of flood mitigation and construction of crop growth models depend upon good understanding of the rainfall process and characteristics. Gamma distribution is usually used to model nonzero rainfall amounts. In this study, the mixed-gamma model is applied to accommodate both zero and nonzero rainfall amounts. The mixed-gamma model presented is for the independent case. The formulae of mean and variance are derived for the sum of two and three independent mixed-gamma variables, respectively. Firstly, the gamma distribution is used to model the nonzero rainfall amounts and the parameters of the distribution (shape and scale) are estimated using the maximum likelihood estimation method. Then, the mixed-gamma model is defined for both zero and nonzero rainfall amounts simultaneously. The formulae of mean and variance for the sum of two and three independent mixed-gamma variables derived are tested using the monthly rainfall amounts from rainfall stations within Kuantan district in Pahang Malaysia. Based on the Kolmogorov-Smirnov goodness of fit test, the results demonstrate that the descriptive statistics of the observed sum of rainfall amounts is not significantly different at 5% significance level from the generated sum of independent mixed-gamma variables. The methodology and formulae demonstrated can be applied to find the sum of more than three independent mixed-gamma variables. |
|---|