Potential of integrated trees-pasture-based systems for GHG emission mitigation and improving soil carbon dynamics in the Atlantic forest biome, Southeastern of Brazil

Brazilian cattle production is mainly carried out in pastures, and mitigating the livestock's greenhouse gas (GHG) emissions has become an essential prerequisite. The importance of planted forests in beef cattle pasture-based integrated systems has been widely recognized in the context of GHG emission mitigation. Adopting the intensification and integration with forest and crops of pasture-based livestock production systems are alternatives to optimizing the soil and tree carbon (C) sequestration; however, research on tropical conditions is still necessary. This long-term trial evaluated the potential of C sequestration (soil + trees) of the forest and the grazing systems with two replicate areas of 3 ha each: 1. continuous stocking rate (CLS) – pasture under continuous stocking; 2. agropastoral or crop-livestock system (AGROP) – pasture under rotational stocking in rotation with corn; 3. agrosilvipastoral or crop-livestock-forestry system (AGROSIL) - similar to AGROP but including eucalyptus; 4. silvipastoral or livestock-forestry system (SIL) - pasture under rotational stocking + eucalyptus and 5. Atlantic Forest (FOR). Soil samples were analysed for texture, C and bulk density in eight soil layers from three to eight trenches for each treatment, then C stock and C accumulation rates were calculated. Data from Eucalyptus urograndis of the SIL and AGROSIL systems with 333 trees. ha−1 during five first years and 166 trees. ha−1 in subsequent years were collected (five- and eight-years old trees) and used to define the wood volume and acquire wood rings. Elemental analyzer CHN performed soil and trees C content. Equations for determining stem volume and tree biomass and, afterward, the C pools of the tree trunks were proposed. The potential of GHG emission mitigation (t CO2 e.. ha−1. year−1) by annual C sequestration (soil or soil + trees) was calculated, including the wood used for the lumber-mill scenario. The participation of the C stock from the 0–30 cm soil layer was between 45 % and 53 % of the total C stock in the 0–100 cm soil layer. Regardless of the elevated C soil accumulation rates, 0.88, 2.54, and 1.74 t C. ha-1. year−1 in the conversion stages CLS → AGROP; AGROP → SIL; AGROP → AGROSIL, respectively, it was not possible to compensate for the negative result of the Forest → CLS conversion (-2.55 t C. ha-1. year−1) because this stage lasted 12 years, while the implementation of AGROSIL and SIL systems had only six years. However, when C tree accu