The dominating tropical wetland emissions are governed less by temperature than the northern wetlands. In the tropical regions, the seasonality and length of the flooded periods will be the main factor determining any major changes in the atmospheric burden of CH4. On the contrary, the estimated circumpolar CH4 emissions of 30-60Tg CH4 yr-1 (also reviewed by McGuire et al, 2009) are more directly sensitive to climate warming, and may hold a significant potential for feedback to a changing climate. Large-scale CH4 flux models are currently not as advanced as general carbon cycling models and few allow for climate change scenario-based projections of changes in the future. Early attempts to assess and model tundra CH4 emissions driven by climate change all indicated a potential increase in emissions (Roulet et al, 1992; Harriss et al, 1993; Christensen et al, 1995) but more advanced mechanistic models (Walter and Heimann, 2000; Granberg et al, 2001) are now approaching the stage where they, in further developed forms, will be fully coupled with Global Circulation Model (GCM) predictions to assess circumpolar CH4 emissions in the future (Sitch et al, 2007; Wania, 2007). A critical factor is, as eluded to above, not only the mechanistic responses of soil processes (dominant in the northern wetland response) but also the geographical extent of wetlands (dominant feature for tropical wetlands), and how these may change in the future. Combined predictive hydrological and ecosystem process modelling is needed for an improved certainty in projections of changes in global wetland emissions. There is, however, little doubt that with climate scenarios of warming and wetting of the soils, there will undoubtedly be increases in CH4 emissions, while with warming and drying there will be few changes, or a decline of emissions relative to the current scale.
The model developed by Walter and Heimann (2000) and applied by Walter et al (2001a, 2001b) has seen widespread uses for predictive purposes. Shindell et al (2004) used this model to examine the potential feedback of wetland CH4 emissions on climate change. They simulated a 78 per cent increase in CH4 emissions globally in a scenario with a doubling of the atmospheric CO2. This increase was significant in the tropical regions but also included a doubling of northern wetland emissions from a modelled 24-48Tg CH4 yr-1. Gedney et al (2004) estimated substantial increases in wetland emissions, doubling globally by 2100. This led in their model to an overall increase of approximately 5 per cent in radiative forcing compared with the CO2 feedback effect modelled by Cox et al (2000). But, as noted by Limpens et al (2008), this has to be seen in the light of the latter being a factor of two larger than any other coupled carbon-climate simulation over the 21st century (Friedlingstein et al, 2006). The wetland CH4 emission feedback effect therefore may well be greater than suggested by the model of Gedney et al (2004). This is exemplified by improved process modelling based on the Walter model presented by Wania (2007), which suggests an increase in emissions by over 250 per cent in vast wetland areas by the end of this century.
Was this article helpful?