Beyond the Methanogenic Black-Box: Greenhouse Gas Fluxes (CO2, CH4, N2O) as Evidence for Wetlands as Dynamic Redox Systems

Abstract:

Seminal wetland research in the 1990s demonstrated that annual methane (CH₄) fluxes scaled positively with ecosystem production across distinctive wetlands globally. This relationship implies a model of flooded wetland ecosystems as ‘methanogenic black-boxes’; poised at a low redox state, and tending to release a fixed fraction of incoming annual productivity as CH₄. In contrast, recent studies have reported high ratios of carbon dioxide (CO₂) to CH₄ emissions, and are adding to a body of evidence suggesting wetlands can vary more widely in their redox state. To explore this apparent incongruence we used principles of redox thermodynamics and laboratory experiments to develop predictions of wetland greenhouse gas (GHG) fluxes under different redox regimes. We then used a field study to test the hypothesis that ecosystem seasonality in gross primary productivity (GPP) and temperature would drive changes in GHG emissions, mediated by a dynamic - as opposed to static - redox regime.

We estimated wetland GHG emissions from an emergent marsh in the Sacramento Delta, CA from March 2014-2015. We measured CO₂, CH₄ and N₂O emissions via diffusion and ebullition with manual sampling, and whole-ecosystem fluxes of CO₂ and CH₄ using eddy-covariance. Ebullition and diffusive CH₄ fluxes were strongly seasonal, with minimum rates (0.86 and 0.35 mg C-CH^­­₄ m^-2 yr^-1, respectively) during winter, and maximum rates (1.3 and 1.8 g C-CH^­­₄ m^-2 yr^-1, respectively) during the summer growing season. In contrast, winter diffusive CO₂ fluxes (494 g C-CO₂ m^-2 yr^-1) and fall bubble CO₂ concentrations (1.49%) were highest, despite being seasons of lower GPP, temperature, and CH₄ flux. Further, diffusive and ebullition fluxes of N₂O showed zero net flux only during spring and summer months, whereas the wetland was a significant source of N₂O during winter (81.2 ± 24.4 mg N-N₂O m^-2 yr^-1). These seasonal flux dynamics contradict a ‘methanogenic black box’ model of wetland redox, which predicts carbon limitation of, and concurrent maxima in, heterotrophic CO₂ and CH₄ emissions, and no significant N₂O emissions. Rather these results suggest that wetlands can function as dynamic redox environments where GHG emission rate and composition varies predictably in time with seasonal changes in GPP and temperature.