Age-age correlation models for early selection of Azadirachta indica provenances planted in semi-arid region

Abstract

An age-age correlation indicates the gain sacrificed to obtain the time advantage, a strong correlation implies that time advantage is met with a small reduction in absolute gain per cycle or vice-versa. A study was carried out to develop age-age correlation models, which could predict the phenotypic age-age correlations and determine the age of early selection. Ten years growth data of Azadirachta indica provenance trial on height and diameter at breast height (dbh) were compiled and analyzed in SYSTAT 11.0 software. Pearson correlation coefficients for tree traits viz. height, dbh and D²H (dbh² x height) were computed between ages 5 to 10 years. Based on these age-age  correlations, empirical models of the form r_{age, age} a+ b LAR; where, LAR- Log (younger age/older age) and r_{age, age}-correlationbetween two ages were fitted for all three traits. Empirical model based on 'dbh' was found best and thus used for estimating selection efficiency in terms of gain per unit time. It suggested that for the plantation age 9 and rotation period of 30 years, the selection efficiency comes out to be 1.179, which means selection at 9 years would yield 1.179 times more gain than selection at age of 30 years. Correlation between 7 & 9 years was significantly high (r=0.829, p<0.05) in case of 'dbh', thus A. indica provenances may be selected at an early age of 7 years on the basis of this trait in semi-arid region.