Yield Assessment from Selected Growth Parameters of Maize Grown in Embu County, Kenya

Abstract

The present study was carried out to determine the growth stage at which the number of leaves and plant height best predict the grain and biomass production of DK8031 maize grown in Embu County, Kenya. The study was an RCBD laid out in a split plot arrangement in which irrigation levels and nitrogen rates were assigned the main and sub-plot treatments, respectively. The MLR models gave very good estimates of grain yield in Season II compared to Season I, with maize grain yield of 3,968 kg/ha that was only just above the observed yield of 3,966 kg/ha, making it promising for use in maize growing in Embu County, Kenya. At harvest, grain yields were measured and data was maintained using Microsoft Excel and the SAS computer program (V9.0). When using single factor regression, the correlation coefficient (r) and coefficient of determination (R²) for the best model fits were in the eighth week; when using multilinear regression functions, the correlation coefficient (r) and coefficient of determination (R²) for the best model fits were in the sixth week. For the latter crop, the coefficients were often substantially lower. The best fit for predicting yield is possible six weeks after sowing the crop, according to multifactor regression models employing the same agronomic characteristics. This is significant because it provides the maize grower with an early opportunity to determine whether to continue increasing, reducing, or stopping additional crop resource inputs. Such a decision helps in maize production for commercialization or domestic use with reasonable resource utilization.