ABSTRACT
The physical controlling factors on coastal plant communities are among the most dynamic of known ecosystems, but climate change alters coastal surface and subsurface hydrologic regimes, which makes rapid measurement of greenhouse gas fluxes critical. Greenhouse gas exchange rates in these terrestrial-aquatic ecosystems are highly variable worldwide with climate, soil type, plant community, and weather. Therefore, increasing data collection and availability should be a priority. Here, we demonstrate and validate physical and analytical modifications to automated soil-flux chamber measurement methods for unattended use in tidally driven wetlands, allowing the high-frequency capture of storm surge and day/night dynamics. Winter CO2 flux from Sarcocornia perennis marsh to the atmosphere was significantly greater during the day (2.8 mmol m-2 hr-1 ) than the night (2.2 mmol m-2 hr-1 ; p < 0.001), while CH4 was significantly greater during the night (0.16 µmol m-2 hr-1 ) than the day (-0.13 µmol m-2 hr-1 ; p = 0.04). The magnitude of CO2 flux during the day and the frequency of CH4 flux were reduced during a surge (p < 0.001). Surge did not significantly affect N2 O flux, which without non-detects was normally distributed around -24.2 nmol m-2 hr-1 . Analysis with sustained-flux global potentials and increased storm surge frequency scenarios, 2020 to 2100, suggested that the marsh in winter remains an atmospheric CO2 source. The modeled results showed an increased flux of CO2 to the atmosphere, while in soil, the uptake of CH4 increased and N2 O uptake decreased. We present analytical routines to correctly capture gas flux curves in dynamic overland flooding conditions and to flag data that are below detection limits or from unobserved chamber-malfunction situations. Storm surge is an important phenomenon globally, but event-driven, episodic factors can be poorly estimated by infrequent sampling. Wider deployment of this system would permit inclusion of surge events in greenhouse gas estimates.
