Evaluation of parameters estimated by three growth analysis methods for predicting maize (Zea mays l.) productivity

Abstract

Identification of the best methods for growth analysis is critical for accurate prediction of crop productivity. Two important components of growth analysis in crops are growth rate (GR) and relative growth rate (RGR). Although several methods have been proposed for computing these and several other parameters of crop growth, there is paucity of information on the comparison of the methods and identification of the best method in computing GR and RGR especially of maize (Zea mays L.) in tropical environments. Dry-matter samples obtained at 5-day intervals from 9 to 39 days after planting (DAP) from sixteen maize varieties planted in three-replicate randomized complete block design, were used to compute GR and RGR by three methods, in order to identify the best method for computing the growth parameters; and determine the relationship between the growth parameters and maize yield. Statistical analysis of the data showed significant differences among the methods (P<=0.01). The three methods were different in terms of mean GR and RGR. The coefficient of variation showed that the calendar-day and heat unit methods (about 49% and 13% for GR and RGR respectively) were not different from each other while the regression method (44% and 12% for GR and RGR respectively) was more efficient than both methods in computing the growth parameters. Correlation analysis showed that the calendar-day and heat unit methods were better than the regression method in predicting maize productivity. Our results revealed that the regression method was better than the calendar-day and heat unit methods in computing GR and RGR but was not as efficient as the two methods in predicting maize productivity.