Premium
Point processes of seasonal thunderstorm rainfall: 2. Rainfall depth probabilities
Author(s) -
Smith R. E.,
Schreiber H. A.
Publication year - 1974
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/wr010i003p00418
Subject(s) - thunderstorm , storm , homogeneity (statistics) , probability distribution , markov chain , environmental science , probabilistic logic , statistics , mathematics , climatology , meteorology , atmospheric sciences , geography , geology
As a sequel to an earlier study of the stochastic properties of daily rainfall occurrence in a summer thunderstorm season, the probabilistic nature of daily rainfall depths is examined. Daily rainfall depths are found to be well described by what may be termed a compound exponential distribution. Some 1600 to over 2000 daily rainfalls from three stations are partitioned in various conceptual ways to evaluate homogeneity with respect to the pattern of occurrence with sequences or seasons as well as the annual variance of population properties. Although some small but statistically significant inhomogeneities are found, the statistical description of depths in arbitrary intervals, such as annual total depths, is shown to be treatable as the sum of a random number of (independent) random variables by using the model for rainy day occurrence. Simulations from this model match well the measured data from the stations treated. The effect of truncation of the rainfall sample on both depth distribution and Markov chain dependence is discussed. It is also indicated how daily depths are functions, in turn, of individual storm occurrence probability (number per day) and storm depth distributions.