Regression models to predict herbage production and digestibility in a non-regular sequence of cuts

Publisher Website
Google Scholar
Abstract
Regression models were constructed to predict the yields and digestibilities of herbage cut in different sequences of harvests.The yield model used a seasonal production curve modified by the effects of defoliation. Values for the model parameters were obtained by fitting the model to experimental data using a non-linear regression procedure. When these parameters were used, to predict treatment effects in another series of experimental data, good agreement was obtained. The digestibility model incorporated the effect on digestibility time of year, regrowth time and yield.Using the models to interpolate between the results of cutting experiments, annual yields were shown to increase with the date of first cut and also with the interval between subsequent cuts, whereas the average digestibility of the harvested material ecreased with the date of first cut and with the subsequent cutting interval. Finally, a procedure was devised for combining the two models in order to find an optimum cutting strategy for a hypothetical animal production system.