Soil carbon dioxide fluxes in northern semiarid grasslands

The high indigenous soil organic carbon content, root biomass, and microbial populations in prairie soils provide a source of carbon dioxide (CO2) that is important in the carbon budget of grasslands. Soil chambers were used to measure soil CO2 flux from a grazed mixed-grass prairie (GP), nongrazed mixed-grass prairie (NGP), and grazed western wheatgrass (WWG) [Pascopyrum smithii (Rybd) Löve] grasslands in the Northern Great Plains, USA. Objectives were to quantify soil CO2 fluxes for each site and to determine the contribution of soil temperature, soil water content, and air temperature to soil CO2 flux. Soil CO2 fluxes were measured on each site about every 21 d at 13:00 h during the 25 April–27 October growing period from 1996 to 2000 for NGP and GP and from 1996 to 1998 for WWG. Dormant period fluxes were measured on the GP from 28 October to 26 April from 1999 to 2001. In addition, five sequential daytime measurements were made on each site for 3 days each year. Fluxes were low in the spring and autumn and peaked concurrent with biomass in late June to mid-July. Maximum fluxes for these dissimilar managed grasslands averaged 5.8 g CO2-C m−2 d−1 for NGP, 6.9 g CO2-C m−2 d−1 for GP, and 6.1 g CO2-C m−2 d−1 for WWG. Soil fluxes measured during the dormant period decreased to near zero during the months of December, January, and February and then increased rapidly in March as soil temperatures increased. Daily soil flux during the growing period averaged 3.5 g CO2-C m−2 d−1 for NGP, 4.3 g CO2-C m−2 d−1 for GP, and 4.0 g CO2-C m−2 d−1 for WWG. Dormant period fluxes for the GP averaged 0.5 g CO2-C m−2 d−1. Regression analysis indicated that soil temperature accounted for 65%, soil water content 5%, and air temperature 3% of flux variability. Growing period soil CO2 flux over years averaged 728 g CO2-C m−2 and dormant period CO2 flux averaged 86 g CO2-C m−2. A predictive relationship describing the response of soil CO2 flux to changes in soil temperature was developed using the minimum, maximum, and optimum soil temperatures for soil CO2 flux. The model provides an estimate of the important dormant period soil flux component in annual ecosystem carbon budgets.