FREQUENCY ANALYSIS OF CENSORED DATA

Abstract

Flood frequency Analysis is one of the most important aspects of hydrological studies from its requirement in the design and analysis of water resources structures point of view. This is because of the fact that, any water resources structures design requires estimation of hydrologic and meteorological events for different return periods based on extrapolation. This can be achieved through frequency analysis approach in which in general, a past record is fitted with a statistical distribution which in turn is used to make inferences about frequency of future flows. In the study of this flood frequency analysis, estimation of parameters of a distribution plays an important role, for it is required in the estimation of floods of a given return period. Here in this paper, the "FREQUENCY ANALYSIS OF CENSORED DATA", using partial probability weighted moments (PPWMs) method has been introduced for dealing with censored samples. By censored samples we mean observed data sets containing values above or below the analytical threshold of measuring equipment. This method is a very useful technique in studying the undesirable effects of low outliers of river flows, which usually occurs in semi-arid and and zones. It is also useful in analyzing frequency of river discharges, which are reported as zero for values below some measurement threshold, which in actual case be zero or between zero and the measurement threshold, yet reported as zero. The input data used for this study have been considered under two cases: those which are generated artificially using the mixed congruent or modular method and those 113 actual data of gauged sites of central India, known as Subzone 3(c) or Upper Narmada and Tapi. This study has shown that, censored samples could yield high quantile estimates that are almost as efficient as those obtained from uncensored samples. This result could also be a very useful guide for dealing with the undesirable effects of low outliers, which commonly occur in semiarid and arid zones. iv