Effect of Assessment Scale on Spatial and Temporal Variations in CH4, CO2, and N2O Fluxes in a Forested Wetland

Emissions of methane (CH4), carbon dioxide (CO2), and nitrous oxide (N2O) from a forested watershed (160 ha) in South Carolina, USA, were estimated with a spatially explicit watershed-scale modeling framework that utilizes the spatial variations in physical and biogeochemical characteristics across watersheds. The target watershed (WS80) consisting of wetland (23%) and upland (77%) was divided into 675 grid cells, and each of the cells had unique combination of vegetation, hydrology, soil properties, and topography. Driven by local climate, topography, soil, and vegetation conditions, MIKE SHE was used to generate daily flows as well as water table depth for each grid cell across the watershed. Forest-DNDC was then run for each cell to calculate its biogeochemistry including daily fluxes of the three greenhouse gases (GHGs). The simulated daily average CH4, CO2 and N2O flux from the watershed were 17.9 mg C, 1.3 g C and 0.7 mg N m−2, respectively, during the period from 2003–2007. The average contributions of the wetlands to the CH4, CO2 and N2O emissions were about 95%, 20% and 18%, respectively. The spatial and temporal variation in the modeled CH4, CO2 and N2O fluxes were large, and closely related to hydrological conditions. To understand the impact of spatial heterogeneity in physical and biogeochemical characteristics of the target watershed on GHG emissions, we used Forest-DNDC in a coarse mode (field scale), in which the entire watershed was set as a single simulated unit, where all hydrological, biogeochemical, and biophysical conditions were considered uniform. The results from the field-scale model differed from those modeled with the watershed-scale model which considered the spatial differences in physical and biogeochemical characteristics of the catchment. This contrast demonstrates that the spatially averaged topographic or biophysical conditions which are inherent with field-scale simulations could mask “hot spots” or small source areas with inherently high GHGs flux rates. The spatial resolution in conjunction with coupled hydrological and biogeochemical models could play a crucial role in reducing uncertainty of modeled GHG emissions from wetland-involved watersheds.