Validation of Flowering Models for Classifying Maize into Maturity Groups in the Different Agro-ecologies of West and Central Africa

Abstract

Aims: This study was conducted with a view to validate and compare flowering models (indices) containing temperature, heat units, photoperiod and their combinations developed to predict the number of days to flowering in maize in different agro-ecological zones of West and Central Africa. Study Design:  Simple and multiple linear regression models developed on 100 maize varieties as the inverse of the number of days from planting to flowering (1/f) on values of temperature, heat units and photoperiod were validated. Place and Duration of Study: One hundred maize varieties of different maturity groups were evaluated for days to tasseling, anthesis and silking at the Teaching and Research Farm (TRF) of Obafemi Awolowo University Ile-Ife, Nigeria during late and early cropping seasons of 2007/2008 and 2008/2009 respectively. The 2007 developed models were validated with 2008 early, 2008 late, 2009 early, 2012 late, 2013 early seasons and flowering data of the other agro-ecologies.  Methodology: The developed flowering models were validated with the flowering data of the 100 varieties obtained during the 2008 early, 2008 late, 2009 early, 2012 late and 2013 early seasons. Furthermore, flowering dates and the weather data from research stations in the Forest, Guinea savanna and Sudan savannah agro-ecologies were used to validate the flowering models. Results: The thermal model best predicted the time from planting to the expression of anthesis. The thermal model accurately predicted the anthesis of about 30-40% of the varieties in the 2008 early and late seasons, trial in environment different from the one in which it was developed in the rainforest agro-ecology. Heat units models closely predicted 10 varieties from the hundred varieties evaluated in this study in 2012 late and 2013 early cropping seasons. In the Southern Guinea savannah and Sudan savannah, none of the models predicted the number of days to the flowering traits of some of the 100 varieties correctly. Conclusion: Predicted days to anthesis due to thermal and heat units models are the best models to classify maize varieties into maturity groups in the rainforest agro-ecology The thermal model correctly classified over 40% of the 100 maize varieties into maturity groups, with the coefficient of determination (r2) ranging from 0.94 to 0.97.