Land use-land cover gradient demonstrates the importance of perennial grasslands with intact soils for building soil carbon in the fertile Mollisols of the North Central US

Abstract

Listen

Translate article

The impact of land use change and agricultural management on the cycling of soil organic carbon (SOC) is not well understood, limiting our ability to manage for, and accurately model, soil carbon changes at both local and regional scales. To address this issue, we combined long-term soil incubations with acid-hydrolysis and dry combustion to parse total SOC (Ct) into three operationally defined SOC pools (active, slow, and recalcitrant) from 9 long-term sites with varying land uses on current and former tallgrass prairie soil. Land uses represented a gradient of soil disturbance histories including remnant prairie, restored prairie, grazed pasture, annual crop rotations, and continuous maize. Dry combustion was used to estimate total carbon (Ct, physical), while acid hydrolysis of both the active (Ca) and slow (Cs) pools was used to estimate a recalcitrant carbon pool (Cr, chemical). Non-linear modeling of CO2 efflux data from the long-term incubations was then used to estimate Ca, and the decomposition rates of both Ca and Cs (ka and kr, biological). The size of the slow pools Cs was then defined mathematically as Ct-(Ca + Cr). Remnant prairie had the highest Ct, while cool-season pasture and a 35-y-old restored prairie had higher Ct than the other agricultural systems. All agricultural systems, including pasture, had the highest fraction of Ct as Cr (∼50%), whose mean residence time (MRT) in these soils is ≥500 years (Paul et al., 2001a) demonstrating that this fraction persists, while the more labile fractions were lost over the course of a few months (Ca) to a few decades (Cs) as a result of tillage-intensive agriculture. The two- to four-decade MRT time of Cs indicated a pool likely to be more responsive to the 20 to 40 years of land-use practices used at some of the sites. The Cs pool was largest in the remnant- and 35-y-old prairies indicating significant C accrual and stabilization compared to the agricultural ecosystems. Interestingly, the remnant prairie maintained the highest Ca pool as well, demonstrating the strong connection between the quantity of fresh C inputs and the potential for long-term C stabilization and accrual. The accumulation of C in active (≈labile) pools as a first step toward long-term stabilization highlights the tenuous nature of early carbon gains, which can be quickly lost in response to climate change or poor management.