PS 34-246 - Nutrient availability and soil depth influence soil carbon and nitrogen stocks in a long-term grassland restoration

Abstract: Belowground carbon (C) and nitrogen (N) pools in formerly cultivated soil restored to grassland change over time in response to cessation of tillage and establishment of perennial vegetation. Available inorganic N declines rapidly with increased NPP, root biomass, root C:N ratio, and microbial biomass during grassland reestablishment. Nevertheless, the dynamics of soil C and N pools are better understood in native grasslands than in restored systems, where these pools change dynamically over time. We measured belowground C and N pools in response to long-term manipulations imposed to increase soil heterogeneity in a 24-year-old grassland restoration experiment. Total soil C and N stocks, root biomass, and root %C and % N were quantified in four plots, each containing 12 subplots with a factorial combination of three levels of nutrient availability (reduced-N, ambient-N, and enriched-N) and two levels of soil depth (shallow or deep soil). Data were analyzed using mixed models and a split-block design with nutrient availability and soil depth as fixed affects and block, nutrient strip, and soil depth strip as random effects. We hypothesized that enriched-N soils would have higher root biomass with lower root C:N ratios and higher soil C and N stocks with lower C:N ratios than ambient-N soils after two decades of restoration and annual fertilization. Contrary to our hypotheses, total soil C and N stocks were affected by an interaction between nutrient availability and soil depth, with shallow soils under reduced N having higher soil C and N stocks than deep soils in the reduced N treatment. Long-term C additions used to reduce soil N availability resulted in a C:N ratio greater than enriched-N soils. Root biomass and C:N ratios did not differ among soil treatments. Soil N stock was positively correlated with root biomass, and the soil treatments exhibiting this relationship included the reduced N treatment over all soil depths (r=0.62; P = 0.01) and the shallow soil treatment over all levels of nutrient availability (r=0.54, P < 0.01). Where root access to N supply was limited (e.g., restricted access to mineralized N in deep soil or where there were high rates of immobilization of available and exogenous N), root biomass and total soil N stocks were correlated. Additional analysis of microbial biomass and mineralization rates will provide a more holistic understanding of long-term changes in C and N dynamics in restored grasslands under contrasting manipulations of nutrient availability and soil depth.