The ability to quantify soil–plant–atmosphere water, heat, and gaseous exchange is important for understanding the linkages between terrestrial ecosystems and Earth’s climate. We combined a one-dimensional vertical soil water flow and heat transport algorithm with a soil CO2 production and transport algorithm to study water, heat, and CO2 exchange for vegetated areas in cold climates that are subject to seasonal snow accumulation and soil water freeze–thaw. The model was applied to a semiarid high-elevation mixed-grass rangeland near Laramie, WY. Calibration was achieved by optimizing seven parameters describing plant growth respiration, soil profile CO2 production due to roots and microbes, and the depth distribution of roots and microbes using global parameter optimization. Modeling efficiency (ME) for the calibration and validation periods was high for soil temperature (ME = 0.93–0.96) and variable for soil water content (ME = −0.93 to 0.74), soil CO2 concentration (ME = 0.01–0.59), ecosystem respiration (Reco, ME = −0.71 to 0.59), and snow height (ME = 0.09). The calculated annual C budget for 2 yr yielded a gross primary production (GPP) of 92 to 107, an Reco of 183 to 188, and a net ecosystem exchange (NEE) of −81 to −90 g C m−2 yr−1. The NEE values, suggesting that the site served as a significant C source, were deemed unrealistic considering the fact that precipitation for the site was about average for both years. Sensitivity analysis was used to explore the impact of N-limitation and soil–plant-respiration parameters on GPP, Reco, and NEE.