Combining eddy-covariance and chamber measurements to determine the methane budget from a small, heterogeneous urban floodplain wetland park

• Eddy covariance and chamber data combined for the methane budget of a wetland park.
• Development of strategies to quantify methane budget of heterogeneous ecosystems.
• Heterogeneity of methane fluxes a large uncertainty when up-scaling measurements.

Methane (CH4) emissions and carbon uptake in temperate freshwater wetlands act in opposing directions in the context of global radiative forcing. Large uncertainties exist for the rates of CH4 emissions making it difficult to determine the extent that CH4 emissions counteract the carbon sequestration of wetlands. Urban temperate wetlands are typically small and feature highly heterogeneous land cover, posing an additional challenge to determining their CH4 budget.The data analysis approach we introduce here combines two different CH4 flux measurement techniques to overcome scale and heterogeneity problems and determine the overall CH4 budget of a small, heterogeneous, urban wetland landscape. Temporally intermittent point measurements from non-steady-state chambers provided information about patch-level heterogeneity of fluxes, while continuous, high temporal resolution flux measurements using the eddy-covariance (EC) technique provided information about the temporal dynamics of the fluxes. Patch-level scaling parameterization was developed from the chamber data to scale eddy covariance data to a ‘fixed-frame’, which corrects for variability in the spatial coverage of the eddy covariance observation footprint at any single point in time. By combining two measurement techniques at different scales, we addressed shortcomings of both techniques with respect to heterogeneous wetland sites.We determined that fluxes observed by the two methods are statistically similar in magnitude when brought to the same temporal and spatial scale. We also found that open-water and macrophyte-covered areas of the wetland followed similar phenological cycles and emitted nearly equivalent levels of CH4 for much of the year. However, vegetated wetland areas regularly exhibited a stronger late-summer emission peak, possibly due to CH4 transport through mature vegetation vascular systems. Normalizing the eddy covariance data to a fixed-frame allowed us to determine the seasonal CH4 budget of each patch and the overall site. Overall, the macrophyte areas had the largest CH4 fluxes followed by the open water areas.Uncertainties in the final CH4 budget included spatial heterogeneity of CH4 fluxes, the tower footprint, measurement in the data to be scaled, and gap-filling. Of these, the spatial placement of the chambers provided the largest source of uncertainty in CH4 estimates. This reinforces the need to utilize site-level measurements when estimating CH4 fluxes from wetlands as opposed to using only up-scaled point measurements.