Assessing Impacts of Seasonal Climate Variability on Maize Yield in Kenya

Abstract

Maize (Zea Mays) is the major food crop in Kenya. Its production variation has devastating consequences on people’s basic food availability. This study will investigate the relationships between climate variability and maize yield using observed weather data from Kenya Meteorological Department and national annual maize yield data from the Ministry of Agriculture for the period 1979–2012. Mann–Kendall test was used to detect a trend in precipitation, minimum, and maximum temperature. Location-wise correlation method was performed between each climate variable and maize yield in every station. Stations which had significant correlations were aggregated to form climate indices which were used to build multiple linear regression model. The results revealed that maize yield in Kenya was significantly decreasing at a rate of 0.07 tons/ha/decade at the 95% confidence level accompanied by high inter-annual variation, while world average was increasing at a rate of 0.6 tons/ha/decade. This reduction was accredited to a significant increasing temperature and reduction in seasonal rainfall. Empirical relationship derived from multiple regression models indicates that 67.53% of yield variance was attributed to varying seasonal climate indices. Precipitation is the dominant predictor accounting to 49.73% of yield variance. There is a significant correlation of 0.78 between the modeled and observed yield hence high credibility of the statistical model. A Continuous decrease of maize yield is expected under the influence of climate change which threatens national food security if effective measures to raise maize production are not endorsed. These findings form a framework for designing policies geared towards the reduction of climate-related vulnerability in many parts of the world.