Improved procedures for estimating rainfall extremes

Abstract

Sri Lanka is highly susceptible to natural disasters due to extreme rainfalls caused by monsoons, cyclones and depressions, associated with the country's geographical location in the Indian Ocean. This gives rise to the occurrence of severe floods, resulting in spending of substantial funds annually for flood relief programmes. Thus effective strategies need to be formulated to mitigate environmental damage, economic losses and loss of life. The objective of this paper is to present an appropriate guideline for estimating extreme rainfall, where a gap exists in the knowledge between research and design office practice. Two procedures recommended by the World Meteorological Organization (WMO) for estimating Probable Maximum Precipitation (PMP) which is indicative of extreme rainfall are considered, and the relevance of these methods in the context of Sri Lanka are discussed. Annual maximum daily rainfall series are analyzed at nine meteorological stations in Sri Lanka to estimate 24-h PMP. The results reveal the importance of incorporating both wind and moisture maximization in the hydro-meteorological procedure. The statistical method, which requires only daily rainfall data, is capable of yielding reliable estimates of PMP rapidly, making it a versatile tool in hydrological design.