RESUMEN
Variability in climatic conditions of low-latitude tropical grass cultivation can affect forage production dynamics. Pasture ecosystems are complex and preferably studied from a multifactorial point of view through multivariate approaches. Therefore, in this study, we characterized different growing conditions for grasses of the Megathyrsus genus through studies conducted in tropical humid and semi-arid conditions. We applied principal component, canonical correlation, and discriminant function analyses to the measurements of agronomic and agrometeorological variables in six studies with Guinea and Massai grasses. The principal component analysis, through the climatic characterization by the first principal component, reflects the contrast between water availability and nitrogen variables and energy supply. Agronomic characterization occurred through the distinction between the density of tillers, forage accumulation, and increase in height, versus the accumulation of stems and dead material. The canonical correlation analysis generated a correlation coefficient of 0.84 between the agronomic and agrometeorological variables. There was a contrast between the dead material accumulation and the other agronomic variables, while the agrometeorological variables showed characteristics similar to the first principal component. Discriminant function 1, with 70.36% separation power, distinguished the cultivation conditions based on the study locations. Grass cultivars were differentiated by discriminant function 2, with a 19.20% separation power. From a multivariate variability analysis, despite the similarities of radiation and temperature in the regions studied, the availability of water and nutrients and measurements of agronomic variables can aid in future modeling studies on forage production.
RESUMEN
The objective was to evaluate the microclimate and animal thermal comfort indexes in silvopastoral system in the Brazilian semiarid region with different woody cover levels within Caatinga trees. The experiment was carried out at the Teaching, Research and Extension Unit of the Federal Institute of Science and Technology of Ceará, Campus Limoeiro do Norte, in Chapada do Apodi, CE, in 2018. The treatments were forage grown under woody cover of 46.15% (SS46), forage grown under woody cover of 30.00% (SS30), and forage grown under woody cover of 17.64% (SS18). The evaluations were carried out during three consecutive days, in the vegetative, transition, and dormancy periods at 6 a.m.; 10 a.m.; 2 p.m. and 6 p.m.. First, the microclimate variables, relative air humidity (%), wind speed (m s-1), dry bulb temperature (°C), wet bulb temperature (°C), black globe temperature (°C), and dew point temperature (°C), were obtained for subsequent calculation of black globe temperature and humidity index and radiant thermal load (W m-2). SS46, SS30, and SS18 systems had no differences of black globe temperature and humidity index and radiant thermal load in the sun. There were significant reductions in black globe temperature and humidity index and radiant thermal load within the trees promoting improvement in the thermal environment at times of higher radiation load. The trees promoted better conditions of animal thermal comfort in all silvopastoral arrangements and periods of the year.