Modeling of the biomass accumulation dynamics in Pennisetum purpureum cv. king grass in the Western region of Cuba

Abstract

Different regression models were adjusted to predict growth of Pennisetum purpureum cv. king grass in the rainy and dry periods in the Western region of Cuba through the accumulated dry matter yield. Information from experiments carried out by the Department of Pastures of the Institute of Animal was used. Data on biomass accumulation of the pasture at 13 regrowth ages in the rainy period and 10 in the dry one with 14 d of difference between ages were utilized. Five regression models (three linear and two non-linear) and five statistical criteria were evaluated. A sample obtained from the Bootstrap was used for selecting the best adjustment in the studied variable. The
maximum growth rhythm, age at which was attained the inflection point and the appropriate moment for the exploitation of the forage area, were estimated. The model of best adjustment was the Gompertz. The highest value of biomass accumulation for the rainy period was of 23.82 t of DM ha-1 and in the rainy period of 7.31 t of DM ha-1. The period of highest forage growth speed was between 14 and 57 d. In
yield terms, the exploitation of this variety under the studied conditions must be from 88 to 90 d in the rainy period and of 78 to 88 d in the dry period. It is recommended to evaluate the curve of biomass accumulation of Pennisetum purpureum cv. king grass until 140 d of age and to confirm results obtained under different environmental and soil conditions.
Key words: growth dynamics, accumulated yield, Gompertz model, Bootstrap.