ABSTRACT
Dryland agroecosystems could be a sizable sink for atmospheric carbon (C) due to their spatial extent and level of degradation, providing climate change mitigation. We examined productivity and soil C dynamics under two climate change scenarios (moderate warming, representative concentration pathway [RCP] 4.5; and high warming, RCP 8.5), using long-term experimental data and the DayCent process-based model for three sites with varying climates and soil conditions in the US High Plains. Each site included a no-till cropping intensity gradient introduced in 1985, with treatments ranging from wheat-fallow ( L.) to continuous annual cropping and perennial grass. Simulations were extended to 2100 using data from 16 global circulation models to estimate uncertainty. Simulated yields declined for all crops (up to 50% for wheat), with small changes after 2050 under RCP 4.5 and continued losses to 2100 under RCP 8.5. Of the cropped systems, continuous cropping had the highest average productivity and soil C sequestration rates (78.1 kg C ha yr from 2015 to 2045 under RCP 4.5). Any increase in soil C for cropped rotations was realized by 2050, but grassland treatments increased soil C (up to 69%) through 2100, even under RCP 8.5. Our simulations indicate that reduced frequency of summer fallow can both increase annualized yields and store more soil C. As evapotranspiration is likely to increase, reducing fallow periods without live vegetation from dryland agricultural rotations may enhance the resilience of these systems to climate change while also increasing soil C storage and mitigating carbon dioxide emissions.
