Carbon trace gas fluxes along a successional gradient in the Hudson Bay lowland

Patterns and controls of carbon trace gas emissions from wetlands may vary depending upon the spatial and temporal scale being examined. The factors affecting these emissions are thought to be hierarchically related according to their respective scales of importance. A hierarchical model of processes controlling methane emissions from wetlands is presented and examined here. During the 1990 Northern Wetlands Study (NOWES) methane (CH 4 ), carbon dioxide (CO 2 ), and non‐methane hydrocarbon (NMHC) fluxes were measured in static chambers along a 100 km transect in the Hudson Bay lowland (HBL). Environmental variables, vegetation abundance, and ecosystem age and structure were also quantified at each sampling site. The findings indicate that CH 4 emissions from peatlands (e.g., bogs and fens) and other wetlands (e.g., salt marshes) in the region were low, and were nil or negative (i.e., CH 4 uptake) in forests and bog forests dominated by aspen and black spruce. Site to site variations in mean CH 4 flux appeared to be most closely related to mean water table and sedge productivity, both of which are intercorrelated. Seasonal changes in CH 4 flux tend to follow soil temperature fluctuations. Instantaneous CO 2 and CH 4 daytime fluxes exhibit a negative correlation, suggesting that photosynthetic assimilation of carbon may be related to CH 4 emissions, although the processes of CO 2 and CH 4 production are occurring at somewhat different temporal scales. No diurnal variations in CH 4 flux could be detected. While soil water pH trends are not fully explored, there is some indication that high CH 4 fluxes are concentrated around p H 4 and p H 7. Soil temperature closely follows the seasonal progression of CH 4 flux. Estimated CH 4 seasonal flux (1.5–3.9 g CH 4 m −2 season −1 ) and estimated aboveground net primary productivity (NPP) (90–400 g dry weight m −2 season −1 ) show systematic changes along a successional sequence which are consistent with patterns predicted from successional theory. Estimated seasonal NMHC emissions (0.5–1.4 g C m −2 season −1 ) exhibit an increase along the succession from salt marsh to Sphagnum bog communities. Data from several studies were combined to estimate seasonal CO 2 flux from three sites. The estimated fluxes range from a net uptake of 23 g CO 2 m −2 season −1 to a net loss of 77 g CO 2 m −2 season −1 , although there are large uncertainties in these estimates. It is inferred from the assessment of ecosystem age and structure that disturbance effects and successional changes occurring over hundreds to thousands of years in the HBL strongly control regional CH 4 , CO 2 and NMHC emissions by influencing NPP, species composition, community structure, soil (peat) development, and landscape hydrology. Given this, it is likely that models of carbon trace gas flux based on succession models may be useful in predicting climate change‐landscape change feedbacks.