Testing Mean Stage as a Predictor of Alfalfa Forage Quality with Growth Chamber Trials
- Gary W. Fick and
- C. Garth Janson
Estimation of alfalfa (Medicago sativa L.) forage quality as the crop grows in the field provides useful management information. The quantified morphological stage of development defined by mean stage by weight (MSW) can be used to estimate crude protein (CP), in vitro true digestibility (IVTD), neutral-detergent fiber (NDF), acid-detergent fiber (ADF), and acid-detergent lignin (ADL) of standing crop. The purpose of this study was to validate previously published prediction equations for CP, IVTD, NDF, ADF, and ADL based on MSW, including evaluation of general robustness and potential bias in prediction equations. Nine combinations of growthchamber environment and plant age were used to develop a data set independent of the prediction equations to be tested. The tests consisted of linear regression of growth-chamber observations on predicted values. Prediction errors were ≤39 g kg-1 of dry matter for all parameters, and all coefficients of determination were ≥0.61. These compare favorably to calibration statistics for the prediction equations and indicate that MSW is a robust basis for predicting alfalfa forage quality. However, most intercepts and slopes of the linear regressions deviated from 0.0 and 1.0, respectively, indicating bias in model predictions. Therefore, prediction equations should be recalibruted so that the calibration data and the predictive domain overlap as much as possible.
Copyright © 1990.