Partitioning of Heterotrophic Soil Respiration in Corn and Switchgrass to Evaluate In Situ Soil Organic Matter Decomposition Dynamics

Abstract:

The vast majority of in situ soil carbon efflux studies measure total CO₂ evolution (R_S) from soil, which consists of both heterotrophic-(R_H) and autotrophic-(R_A) derived carbon sources. Because R_A represents recently fixed carbon that is relatively transient within the plant-soil system, partitioning R_H from R_A is necessary for understanding organic matter decomposition and deriving ecosystem carbon balances. Management practices and vegetation type can influence R_H through the effects of soil tillage, litter inputs, root turnover, root exudates, and soil microclimate. Yet, partitioning R_H and R_A in the field has proven notoriously challenging, and all current methods are subject to certain biases. The root exclusion method is the most common technique for estimating R_H, but removing live roots from soil can alter physical conditions and thus may bias the R_H estimate. Our objective was to derive and compare in situ estimates of R_H in corn (Zea mays) and switchgrass (Panicum virgatum) temperate bioenergy cropping systems. We hypothesized that summer R_H would be greater in corn compared to switchgrass due to tillage practices and generally warmer surface soil conditions in corn. Additionally, we hypothesized that the root exclusion method would provide relatively low estimates of R_H, particularly during high precipitation periods when the root exclusions may hold excess moisture. We evaluated our hypotheses in July 2014 at the Great Lakes Bioenergy Research Center cropping systems trial in Arlington, WI, USA. Total R_H averaged 50% greater in corn than in switchgrass, but the percent contribution of R_H to R_S was statistically similar between cropping systems at 24%. Following an abnormally wet June, R_H estimates were 18% of R_S, which is lower than most studies in similar cropping systems. However, as root exclusions dried out throughout July, R_H gradually increased to 30% of R_S. Continuing efforts will better constrain R_H estimates using the root regression method and will additionally assess the effects of soil temperature and moisture on R_H.