Landscape Soil Respiration Fluxes are Related to Leaf Area Index, Stand Height and Density, and Soil Nitrogen in Rocky Mountain Subalpine Forests

Abstract:

There is a recent multi-agency push for accurate assessments of terrestrial carbon stocks and fluxes in the United States. Assessing the state of the carbon cycle in the US requires estimates of stocks and fluxes at large spatial scales. Such assessments are difficult, especially for soil respiration, which dominates ecosystem respiration and is notoriously highly variable over space and time. Here, we report three consecutive years of measurement of soil respiration fluxes in three 1 km² subalpine forest landscapes: Fraser Experimental Forest (Colorado), Glacier Lakes Ecosystems Experimental Site (“GLEES”, Wyoming), and Niwot Ridge (Colorado). Plots were established following the protocol of the US Forest Service’s Forest Inventory and Analysis (FIA) Program. Clusters of plots were distributed across the landscape in a 0.25 km grid pattern. From 2004 through 2006, measurements of soil respiration were made once monthly during the growing season and twice during snowpack coverage for each year. Annual cumulative soil respiration was 6.10 (+/- 0.21) Mg ha^-1y^-1 for Fraser, 6.55 (+/- 0.27) Mg ha^-1y^-1 for GLEES, and 6.97 (+/- 0.20) Mg ha^-1y^-1 for Niwot. Variability in annual cumulative soil respiration varied by less than 20% among the three subalpine forests, despite differences in terrain, climate, disturbance history and anthropogenic nitrogen deposition. We quantified the relationship between respiration fluxes and commonly-measured forest properties and found that soil respiration was nonlinearly related to leaf area index, peaking around 2.5 m²m^-2 then slowly declining. Annual litterfall (FA) was subtracted from soil respiration (FR) to calculate total belowground carbon flux (TBCF), which declined with increasing tree height, density and soil nitrogen. This landscape analysis of soil respiration confirmed experimentally-derived principles governing carbon fluxes in forests: as trees age and get taller, and in high-fertility areas, carbon flux to roots declines. In addition, this work suggests that 1) soil respiration may be partially predicted from either FIA data or remotely-sensed products of forest structure, improving the accuracy of large-scale regional C assessments, and 2) soil respiration estimates may be improved by considering spatial variability in soil nitrogen.