A Markov chain simulation model for predicting critical wet and dry spells in Kenya: analysing rainfall events in the Kano Plains

✍ Scribed by W. O. Ochola; P. Kerkides

• Publisher: John Wiley and Sons
• Year: 2003
• Tongue: English
• Weight: 344 KB
• Volume: 52
• Category: Article
• ISSN: 1531-0353
• DOI: 10.1002/ird.94

✦ Synopsis

Abstract

The occurrence of wet and dry spells is a phenomenon most often used to identify the arid and semi‐arid lands (ASAL) in Kenya. The use of first‐order Markov processes that are embedded into a computer model to determine the critical climate extremes is presented. The model uses the concepts of conditional probability, Poisson probability distribution function and chi‐square testing to predict the critical spells. The daily rainfall data (1981–2000) for two weather stations in the Kano Plains (Kenya) have been used to illustrate model application. For example, based upon the bimodal rainfall pattern in the study area, the model revealed the length of the critical dry spell to be 14 days in the long rainy season and 12 days in the short rainy season, while the critical wet spell was found to be 12 and 8 days, for the long and short rains respectively for Ahero Irrigation Scheme. It is recommended that a climate and environmental audit in the Kano Plains for the determination of land and water management strategies include critical dry and wet spell determination. This will enhance more sustainable planning and utilization of the crucial water and land resources in the region and in other tropical rangelands. There are possibilities of integrating the model with other agro‐ecosystems models and other natural resource management decision support systems. Copyright © 2003 John Wiley & Sons, Ltd.