Some climatic factors which affect maize yield in Kakamega district, Kenya
MetadataShow full item record
The main objectives of this study were: - 1) To test the accuracy of the Food and Agriculture Organization (FAO) model developed by Frere and Popov (1979) in predicting maize yield in Kakamega District. 2) To determine whether simple weather factors such as rainfall and evaporation can be used to predict maize yield in Kakamega District. 3) To determine the weather variable (s) accounting for most of the variation in the yield of maize in Kakamega District. Regression analyses were performed independently for three agrometeorological stations located in different agro-ecological zones within the district using climatic and maize yield data from 1970-85 period. The weather variables included rainfall and evaporation during the growing period i.e. March-October and some second-degree variables derived from these. Rainfall probability analyses were also performed at the three-agrometeorological stations to determine whether the rainfall received at these stations was above or below the water requirements for maize at different stages of growth. Findings were: (1) The correlations between FAO Model maize yield predictions and Ministry of Agriculture (MOA) were low and insignificant for both the divisions in which the agrometeorological stations providing climatic data used in the FAO Model were located and the physically distant divisions from the agrometeorological stations used in this study. Possible explanations for this trend are suggested. (2) There is an over 90% chance of receiving 750mm of rainfall (the minimum water requirement for maize throughout the season) at the three agrometeorological stations. (3) Simple weather factors like rainfall, evaporation and some second degree variables derived from these two fell short of being useful in predicting maize yield in Kakamega District. The Step-Wise regression (SR) weather based models developed for prediction purposes in this study had large amounts of unexplained variances. Possible improvements in these SR-weather based models for yield estimation procedure are suggested. (4) Rainfall received during the first month of the growing season, and its function, conserved soil moisture reservoir during the same month, were the best correlated variables to the ultimate maize yield in all the agro-ecological zones considered in this study. The two accounted for the largest percentage of variation in the yield of maize in Kakamega District.