Animal performance in grass monoculture or silvopastures using tree legumes

Abstract

Silvopasture systems using tree legumes have potential to reduce C footprint of livestock systems in the tropics. This 4-year study evaluated productive responses of signal grass (Brachiaria decumbens Stapf.) and the performance of cattle in silvopastoral systems (SPS) in the sub-humid tropical region of Brazil. The experimental design was randomized complete block with three replications. Treatments were: signal grass + Mimosa caesalpiniifolia Benth.; signal grass + Gliricidia sepium (Jacq.); and signal grass monoculture. Cattle were managed under continuous stocking with variable stocking rate. Herbage mass (HM) was greater on signal grass monoculture (5057 kg DM/ha) than in mixed pastures (3292 kg DM/ha, average). There was interaction (P < 0.05) between treatment × evaluation month for HM, and the greatest value was observed in May (2722 kg DM/ha). Herbage allowance varied from 1.29 to 3.07 kg DM per kg of body weight (BW). Herbage accumulation rate varied throughout the evaluations, with the greatest rate observed in July (56 kg DM/ha per day) and the lowest in December (15 kg DM/ha per day). Average daily gain was similar across systems (0.65 kg day−1), varying according to month. Stocking rate ranged from 1.5 animal units (AU; equivalent of a 450-kg animal) per ha in December (dry season) to 2.2 AU/ha in July (rainy season). Animal performance per area tended to decrease in the SPS (411 kg BW ha−1 year−1) when compared with signal grass in monoculture (508 kg BW ha−1 year−1), especially in the Mimosa system. Adding tree legumes reduced animal performance per area due to competition between arboreal and herbaceous components, especially in the Mimosa system that had a more closed canopy.