Open AccessMarkov-dependent geometric models for weather spells and weather cycles -A study
Author(s) -
N. Sundararaj,
Sarathy Ramachandra
Publication year - 1975
Publication title -
mausam
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.243
H-Index - 12
ISSN - 0252-9416
DOI - 10.54302/mausam.v26i2.2166
Subject(s) - markov chain , meteorology , environmental science , markov model , climatology , statistics , markov process , weather forecasting , mathematics , econometrics , geography , geology
The repetitive behaviour of seasonal weather has been studied earlier by many using a variety or mathematical and probability models. The present paper proposes to investigate the empirical validity of Markov-dependent geometric models for wet and dry spells, and weather cycles. The daily rainfall of 50 years duration serves as the data-base for the study.
The necessary formulae based on the geometric probability models using Markovian probabilities for dry and wet days, are given for lengths of dry spells, wet spells and wet dry weather cycles.
The data, as judged by x2-test, seems to fit better for wet-spells than for dry spells.