PREDICTION MODEL OF THE DIGESTIBILITY OF NATURAL-HERBACEOUS PASTURES IN THE CANARY ISLANDS

Abstract

Knowing the energy content of pastures is necessary in order to estimate the livestock carrying capacity of grasslands. Taking into account the high cost of analyses for determining the digestibility of herbaceous biomass as an estimate of its energy content, this study proposes a model for predicting digestibility values on the basis of the most significant factors. For that purpose a multifactorial univariate analysis of variance (ANOVA) was carried out on digestibility to find out which factors, being easy to obtain, will allow to predict it with accuracy. From this analysis, we searched for a model which would allow predicting the energy content of samples taken in the field, on the basis of easily collected variables. There are three factors (plant group, vegetation type and month) which determine the digestibility of the herbaceous production in the areas of study, and therefore can be used to predict the quality of herbaceous primary productivity in the calculation of livestock carrying capacity. The model resulted of high predictability value (R2 = 0.76), permitting the estimation of the digestibility of dry matter in a simple way. Using this model the costs of determination of livestock carrying capacity will be considerably reduced, therefore facilitating the development of grazing management in relation to land use zoning, especially in new areas that has not been examined yet.