PakCat catalogue record

A First Orders Nonhomogeneous Markov Chain for the Daily Rainfall Occurrences in Ankara 'Source Title' Modeling Hydrologic Processes, Proceedings Of 3rd International Hydrology Symposium

By AKSIT, A.A

Bibliographic details

Summary

In deciding agricultural strategiesa stochastic model may be used (a) for the simulation of the rainfall sequences, (b) for the forecasting of wet and dry periods, and (cP for the determination of the risks associated with the wet and dry periods. For these purposes a stochastic analysis of the daily rainfall occurrences in Ankara, Turkey has been made by the use of 50 years of data. Hartley's periodicity test showed a very significant annual periodicity in the daily rainfall occurences in Ankara. The stationarity of the daily rainfall occurrences within each week of the year was established by Anderson and Goodman's test of stationarity. Therefore, the probability structure has been analyzed for every week within the year. Independence, first order markov chain and second order Markov chain models have been tested by Anderson and Goodman's chi square goodness of fit and likelihood ratio tests. The resuts show that the first order non stationary Markov chain adequately models the daily rainfall occurrences in Ankara.