Simulating climate change impacts on soil carbon storage in agroecosystems from Brazilian drylands

Dryland regions of Brazil are increasingly threatened by climate change, which intensifies aridity and reduces agricultural productivity. In this context, soil organic carbon (SOC) plays a critical role in sustaining agroecosystem resilience. This study used the DayCent ecosystem model to simulate long-term SOC dynamics (2024-2100) under current and projected climate scenarios (SSP2-4.5 and SSP5-8.5) across three representative dryland sites: Betânia do Piauí (PI), Petrolina (PE), and Sobral (CE). Field data, including SOC and N stocks, were used to calibrate and evaluate the model for a range of land-use systems, including native vegetation, conventional and intensified agroecosystems (e.g., fertilization, no-tillage, integrated crop-livestock - CLI and crop-livestock-forestry - CLFI systems). R² ranged between 0.97 and 0.73, while root mean square error (RMSE) values varied between 2.09 and 0.55 for SOC and N, respectively. Results showed that land-use conversion often reduced SOC (5-20 %, compared to native areas), especially following fire or under low-input systems. However, system intensification consistently enhanced SOC stocks - 36 to 46 %, relative to CLI-tillage - particularly in no-tillage and fertilized systems. Under future climate scenarios, SOC losses were projected at all sites, especially in sandy soils. Nonetheless, integrated agricultural systems (IASs), as CLI and CLFI, associated with intensified management, partially mitigated these losses up to 2100. While the adoption of intensified management practices improved system resilience, they could not fully offset the adverse effects of increased aridity. These findings underscore the need for targeted adaptation strategies (such as soil conservation, improved nutrient management, and the adoption of IASs) to maintain soil carbon and ensure long-term sustainability in Brazilian drylands.