Markov chain model and its application to annual rainfall distribution for crop production

Abstract

A stochastic process with a first order dependence in discrete state and time is described as Markov chain. This principle was used to formulate a four state model for annual rainfall distribution in Minna with respect to crop production. The model is designed such that if given any of the four state in a given year, it is possible to determine quantitatively the probability of making transition to any other three states in the following year(s) and in the long-run. The model was used to study the data of annual rainfall in Minna. The results show that in the long run 14% of annual rainfall shall be low rainfall, 34% annual rainfall will be moderate rainfall also well spread, 47% of the annual rainfall shall be high rainfall and 5% of the annual rainfall shall be moderate rainfall not well spread respectively. The model provides some information about rainfall in relation to crops cultivation that could be used by the farmers and the government to plan strategy for high crop production in Minna and the immediate environment.