Modelling annual maximum river flows with generalized extreme value distribution
A good understanding of probability distribution of annual maximum river flow is believed to improve water resources planning and design. Based on the annual maximum river flow record over 20-48 years at 9 individual river sites in Sabah, the data set are fitted into generalized extreme value (GEV)...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English English |
| Published: |
SANDKRS sdn bhd.
2019
|
| Subjects: | |
| Online Access: | https://eprints.ums.edu.my/id/eprint/30040/1/Modelling%20annual%20maximum%20river%20flows%20with%20generalized%20extreme%20value%20distribution.pdf https://eprints.ums.edu.my/id/eprint/30040/2/Modelling%20annual%20maximum%20river%20flows%20with%20generalized%20extreme%20value%20distribution1.pdf |
| Summary: | A good understanding of probability distribution of annual maximum river flow is believed to improve water resources planning and design. Based on the annual maximum river flow record over 20-48 years at 9 individual river sites in Sabah, the data set are fitted into generalized extreme value (GEV) distribution with maximum likelihood estimator. Both stationary and non-stationary models are considered. Likelihood ratio test shows that most of the river flows are stationary. Over a homogeneous region, a parent distribution with common shape parameter is found well describing the behavior of selected annual maximum river flow. Hence, 10- and 100-year return levels are estimated using the single model. |
|---|