Generally, climatic warming is expected to start a drying trend in wetland ecosystems. According to Gorham (1991), this largely indirect influence of climate change, leading to alteration in the water level, would be the main agent in ecosystem change and would overshadow the impacts of rising temperature and longer growing seasons in boreal and subarctic peatlands. Monsoonal areas are more likely to be affected by more intense rain events over shorter rainy seasons, exacerbating flooding and erosion in catchments and the wetlands themselves. Similarly, longer dry seasons could alter fire regimes and loss of organic matter to the atmosphere (Hogenbirk and Wein, 1991).
Climate change may be expected to have clear impacts on wetland ecosystems, but there are only a few studies available for assessing this. There are some laboratory studies concerned with the responses of individual plant species or groups of species (Jauhiainen et al., 1997, 1998a,b; van der Heijden et al., 1998). Based on these studies, however, it is difficult to predict the responses of plant communities formed by species with somewhat varying environmental requirements.
The response of wetland plant communities to drought has received some attention in temperate freshwater wetlands (Greening and Gerritsen, 1987; Streng et al., 1989). Stratigraphical studies in peatlands have shown hydroseral succession whereby swamp and fen communities gradually develop into bog communities (Tallis, 1983). These changes are largely autogenic, connected to growth of wetland communities and caused by past climatic variability or artificial drainage. An alternative approach to observe the vegetation-environmental change succession has been to use space as a time substitute by mapping different plant communities onto climatic and hydrological surfaces (Gignac et al., 1991). This approach has shown tight coupling between various peatland plants, climate, hydrology, and resultant chemistryand even for trace gas exchange (Bubier, 1995)and has been used to infer certain aspects of peatland development through macrofossil analysis (Gorham and Janssens, 1992; Kuhry et al., 1993).
Much is known about how vegetation changes as a result of water-level drawdown following drainage for forestry in northwestern Europe. Drying of surface soil initiates a secondary succession whereby original wetland species gradually are replaced by species that are typical of forests and heathlands (Laine and Vanha-Majamaa, 1992; Vasander et al., 1993, 1997; Laine et al., 1995). Plants living on wet surfaces are the first to disappear, whereas hummock-dwelling species may benefit from drying of surface soil. In nutrient-poor peatlands, bog dwarf-shrubs dominate after water-level drawdown; at more nutrient-rich sites, species composition develops toward upland forest vegetation (Laine and Vanha-Majamaa, 1992; Minkkinen et al., 1999).
The effect of sea-level rise on wetlands has been addressed in several assessments. In northern Australia, extensive seasonally inundated freshwater swamps and floodplains are major biodiversity foci (Finlayson et al., 1988). They extend for approximately 100 km or more along many rivers but could be all but displaced if predicted sea-level rises of 10-30 cm by 2030 occur and are associated with changes in rainfall in the catchment and tidal/storm surges (Bayliss et al., 1997; Eliot et al., 1999). Expected changes have been demonstrated by using information collected from the World Heritage-listed Kakadu National Park, but the scenario of massive displacement of these freshwater wetlands can be extended further afield given similarities in low relief, monsoonal rainfall, and geomorphic processes (Finlayson and Woodroffe, 1996; Eliot et al., 1999). In fact, the potential outcome of such change can be seen in the nearby Mary River system, where saline intrusion, presumably caused by other anthropogenic events, already has destroyed 17,000 ha of freshwater woodland and sedge/grassland (Woodroffe and Mulrennan, 1993; Jonauskas, 1996).
Mechanisms by which environmental factors and biotic interactions control wetland biodiversity are not well understood (Gorham, 1994b). The effects of water-level drawdown after drainage for forestry indicate that the shift in species composition from bog and fen species to forest species only slightly affects plant species richness of individual sites (Laine et al., 1995). In regions dominated by forests, there would be clear reduction in regional diversity as landscapes become homogenized after drainage (Vasander et al., 1997).
The inherent changeability of wetland communities, resulting from spatial and temporal variability in water supply (Tallis, 1983), may be the key factor in the response of wetland communities to climate change. Because there may be differences between species in adaptation potential, community structures would change, and there would be profound effects on the nature of the affected wetlands, as discussed by Gorham (1994a). The response of wetland plant communities to changing environment may have fundamental effects on the species diversity of these ecosystems.
Because of spatial and temporal variability in ecosystem processes, development of systems models for wetlands is becoming an important assessment tool. A fully coupled peatland-climate model has not yet been developed, but there have been some significant advances in modeling various components of the peatland and/or wetland biogeochemical system (Harris and Frolking, 1992; Roulet et al., 1992; Christensen and Cox, 1995; Christensen et al., 1996; Walter et al., 1996; Granberg, 1998), and several process-level models are now used at the global scale (Cao et al., 1996; Potter et al., 1996; Potter, 1997).
Elevated CO2 levels will increase photosynthetic rates in some types of vegetation (e.g., C3 trees and emergent macrophytes) (Bazzaz et al., 1990; Idso and Kimball, 1993; Drake et al., 1996; Megonigal and Schlesinger, 1997; see also Section 18.104.22.168). Responses of nonvascular vegetation, such as sphagna, have been less clear (Jauhiainen et al., 1994, 1997, 1998a; Jauhiainen and Silvola, 1999). A study in Alaskan tussock tundra found that photosynthetic rates in Eriophorum vaginatum quickly adjusted downward such that rates with elevated and ambient CO2 were similar after 1 year (Tissue and Oechel, 1987). However, a sustained increase in net ecosystem carbon sequestration was observed when elevated CO2 treatments were combined with a 4°C increase in temperature (Oechel et al., 1994).
Many C3 plants respond to elevated CO2 with a decrease in stomatal conductance (Curtis, 1996), which could reduce transpiration rates. Because transpiration is an important pathway for water loss from many ecosystems (Schlesinger, 1997), including wetlands (Richardson and McCarthy, 1994), reductions in transpiration rate could affect the position of the aerobic-anaerobic interface in wetland soils (Megonigal and Schlesinger, 1997).
Other reports in this collection