Evaluating the dry conditions in Peninsular Malaysia using bivariate copula

Abstract

Prolonged dry condition has become a recurrent phenomenon in Peninsular Malaysia. The country's agricultural sector and water resources are under severe constraints from this condition. The dry conditions are examined using the Standard Precipitation Index based on the data of monthly rainfall for the period of 30 years from ten stations. The duration and severity of the dry conditions, as identified from the distribution of the values of Standard Precipitation Index, are assumed to follow the exponential and gamma distributions respectively. To study the potential risk due to the dry conditions, we model the joint distribution of these variables, upon which the return periods are estimated. The joint density function of the duration and severity are investigated by fitting six types of bivariate copula models. The Algorithmic Information Content measure identifies the most appropriate copula distribution. For the majority of stations under study, the Galambos distribution provides the best fit, and the Gumbel--Hougard distribution the second best. References

• T. B. McKee, N. J. Doesken and J. Kleist. The relationship of drought frequency and duration to time scales, Proceedings of the 8th Conference on Applied Climatology, 1993, 179--184.
• I. Bordi, K. Fraedrich, M. Petitta and A. Sutera. Extreme value analysis of wet and dry periods in Sicily. Theoretical and Applied Climatology, 2007, 87, 61--71. doi:10.1007/s00704-005-0195-3
• D.C. Edwards and T. B. McKee. Characteristics of 20th century drought in the United States at multiple time scales. Climatology Rep 97-2. Colorado State University Dept. of Atmospheric Science Fort Collins Colorado, 1997, 155 pp.
• H. Akaike. A new look at the statistical model identification. IEEE Transactions on Automatic Control AC 1974, 6, 71672.
• C. T. Haan. Statistical Methods in Hydrology. Iowa State University Press: Ames, IA. 1977.