Variabilidade espacial da altura de pastos de capim-marandu sob lotações contínua e intermitente

Abstract

The spatial variability of vegetation, also known as the horizontal structure of the pasture, influences the dynamics of plant growth and animal consumption in the pastoral system. This horizontal pasture structure can be modified by the stocking method and the average pasture height. However, its characterization, although important, is still incipient in tropical pastures. With the aim of characterizing the spatial variability of pasture vegetation under continuous stocking (LC) and intermittent stocking (LI) and with different average heights, an experiment was conducted at the Federal University of Uberlândia, MG, from September 2021 to March 2022, in a completely randomized experimental design with four treatments and three replications per treatment. The experimental area consisted of nine paddocks (experimental units) of 800 m2 with Urochloa brizantha cv. Marandu. The treatments were four pasture conditions: continuous stocking with 15 cm average height (LC-15), continuous stocking with 25 cm (LC-25), intermittent stocking with 25 cm in pre-grazing (LI-25Pre) and intermittent stocking with 15 cm in post-grazing (LI-15Post), with LI-25Pre and LI-15Post pasture conditions being evaluated in the same paddocks under intermittent stocking. Crossbred 50 kg sheep were used for grazing. The pastures were mowed in September 2021 and then managed until March 2022, following the height criteria for each treatment. Plant height was measured at 112 points in each paddock, using a regular 2 x 4 m sampling mesh. In these data sets, position measurements (arithmetic mean, median and mode) and dispersion (coefficient of variation, amplitude and standard deviation) were analyzed. The data was also subjected to geostatistical analysis and interpolation using ordinary “kriging”, to create “kriging” maps. The distributions of plant height data in marandu grass pastures under continuous stocking and under intermittent stocking in pre-grazing with sheep present asymmetry to the right, indicating a non-normal distribution. The mode proved to be a more appropriate approach to represent plant heights, as it represents the most frequent and typical value in the distribution, without the influence of extreme values.