Reciclagem de nutrientes e padrão de distribuição de excretas de bovinos em pastagens de Brachiaria decumbens Stapf. consorciadas com leguminosas na Zona da Mata de Pernambuco

Abstract

This study aimed to evaluate the decomposition and spatial distribution pattern of feces, and the nutrient composition of forage and feces of heifers grazing Brachiaria decumbens (Signal grass) intercropped with Mimosa caesalpiniifolia Benth (Sabia), Leucaena leucocephala (Lam.) de Wit (Leucaena), Bauhinia cheilantha (Bong) Steud (Mororó) and Gliricidia sepium (Jacq.) Kunth ex Walp. (Gliricidia) in the Dry Forest region of Pernambuco. Besides these treatments, we tested two additional treatments: monocultured signal grass with the addition of 60 kg N yr-1 and monocultured signal grass without fertilization. The experimental design was a randomized block with four replications. Two experiments were conducted to look at decomposition of feces, first, to determine the evolution of CO2 from a mixture of soil and feces within a 22-day incubation, and the other to determine the decomposition of feces in nylon bags in at times 4, 8, 16, 32 , 64, 128 and 256 days of incubation in soil. The simple negative exponential mathematical model was adequate (P ≤ 0.0001) to quantify the evolution of CO2 from a mixutre of soil with feces, indicating that 78% of CO2 was released at the beginning of incubation, this was especially apparent in the feces collected in the signal grass intercropped with Gliricidia. The curve of the remaining biomass decomposition in the nylon bags was adequate to the simple negative exponential model (P <0.001) for the other treatments. A higher disappearance rate constant (k) of biomass from animals feces (P <0.001) was observed for the fertilized signal grass treatment (k = 0.0031 g g-1 day-1) and a lower constant for the signal grass intercropped with Sabiá ( k = 0.0018 g-1 day-1). To evaluate the distribution pattern of feces and composition of forage and feces only the treatments of monocultured signal grass and the legume treatments with Gliricidia and Sabia were used. The treatments differed only in N content in feces (P = 0.008), and the N content in the animal feces in the Sabia treatment (20 g kg-1) was higher compared to that observed in the feces of the animals in the Gliricidia ( 17.3 g kg-1) and monocultured signal grass (18.3 g kg-1) treatments. Legumes had higher values of N (42 g kg-1 and 33 g kg-1 for Gliricidia and Sabia respectively), however, the Sabia did not fertilize the signal grass. The 15N natural abundance in the fertilized and unfertilized signal grass was 3.29 ‰ and 2.7 ‰, respectively, being lower (P <0.05) than the signal grass – legume treatments (4.2 ‰). Distribution of manure occurred in an aggregated manner with a higher (P <0.05) Dispersion Index (ID) in October, 2010 (ID = 1.54) and August, 2011 (ID = 1.53). The spatial dependence between the manure pies was considered moderate with an index of spatial dependence (IDE) between 25% to 75% in the periods of January 2010 and May 2011, where they were significantly different (P <0.05) then the other periods with strong spatial dependence (IDE ≤ 25%). Nitrogen fertilization of signal grass favored the decomposition of feces of heifers at the end of 256 days of incubation. The onset of decomposition of feces was benefited by the presence of legumes of signal grass pasture with higher initial emissions of CO2. The introduction of legumes promoted an increase in N content in grass. The presence of shrub legumes did not directly influence the distribution pattern of feces, being similar to monocultured grass pastures.