Evaluation of objective estimates of treatment efficacy for bigleaf maple

An evaluation of six objective estimators of treatment efficacy for bigleaf maple (Acermacrophyllum Pursh) was made on the basis of data from two herbicide screening studies. The estimators were evaluated for sensitivity to statistical comparison of treatment means. Evaluation was made on the basis of (i) overall mean and coefficient of variation for percent change in crown volume, as well as (ii) percent control based on regression-adjusted post-treatment crown height, (iii) crown diameter, (iv) crown profile, (v) crown area, and (vi) crown volume. Estimates of overall mean control were similar among the expressions of crown size, but the estimate based on change in crown volume was substantially lower. The coefficient of variation was smallest when control was calculated from crown height or crown profile and was largest for change in crown volume. Orthogonal contrasts indicated that control calculated from one- and two-dimensional expressions of crown size tended to be more sensitive to treatment differences than other expressions of crown size. Change in crown volume and control calculated from crown volume tended to be the least sensitive to treatment differences, resulting in rejection of the null hypothesis for fewer orthogonal contrasts. These results for bigleaf maple are consistent with the results obtained for several hardwood species in the southeastern United States.