This study addresses a critical gap in agro-meteorological planning: the absence of reliable probabilistic models for predicting monthly rainfall regimes. Specifically, we aim to quantify the long-term distribution of dry, wet, and rainy months in Owerri, Imo State, Nigeria, using a three-state, first-order Markov chain model. Based on 36 years (1987–2022) of monthly rainfall data, the model classifies months into (dry ≤50 mm, wet 51–150 mm, and rainy >150 mm) and were defined based on agro-climatic thresholds used in southeastern Nigeria to reflect minimal, moderate, and high rainfall conditions. Our results reveal that in the long run, the probabilities of experiencing dry, wet, and rainy months are 23%, 18%, and 59%, respectively. The dominance of rainy months indicates a persistent wet climate, which has strong implications for agricultural scheduling. This work offers a valuable stochastic framework that can be used to inform crop planning and water-resource management in rain-fed agricultural systems. This study contributes new evidence to the stochastic rainfall literature by demonstrating the long-run behavior of monthly rainfall regimes in Owerri using a Markov chain framework that incorporates 36 years of updated rainfall data—one of the longest continuous datasets analyzed for this region.
Ohaegbulem et al. (Thu,) studied this question.