Box 6-2. Northern Wetlands: Effects on the Carbon Cycle and Trace-Gas Emissions

Background: Peatlands are a major store of organic carbon and contain approximately 20% of the total amount of organic carbon stored in soils. The northern peatlands account for a majority of this as carbon is stored in the form of peat. This case study demonstrates how changes in climate may affect the flux of carbon (CO, and CH1) and nitrous oxide between these ecosystems and the atmosphere, generating a feedback on climate warming.

Possible impacts: Current scenarios suggest that climate change is likely to increase the flux of CO2 to the atmosphere because temperatures influence whether carbon litter is accumulated into the peat profile or oxidized. In addition, the position of the water table regulates the extent of oxygen penetration into the peat profile. This means that drainage will cause increased decomposition, leading to increased fluxes of CO2 to the atmosphere, although this effect will decline over time. Further, a decrease in water availability could lead to a decrease in CH, emissions from wetlands. Changes in variables such as the areal extent of wetlands and the duration of the active period will determine whether there will be a change in the total CH, flux from a wetland. A lowering of the water table would probably not affect nitrous oxide emissions from bogs but could lead to an increase of emissions from fens, although emissions of nitrous oxide from wetlands tend to be low.

Conclusion: Changes in the source/sink relationship have already occurred in wetlands in some parts of the world. Both climate change and human (non-climate) factors are likely to further affect the biogeochemical functions of wetlands.

1

230

this effect would be amplified in response to a decrease in soil moisture. Thus, it is conceivable that the peat accumulation rate will decrease to half of the present rate or even less (i.e., <0.025-0.055 Gt C/yr). Boreal peatlands may even become net sources of atmospheric CO2. In concluding his discussion of the response of northern wetlands to predicted climate change, Gorham (1991) gives the extreme example of a 1-cm breakdown of the boreal peat layers worldwide. This would result in 2 Gt C/yr, which corresponds to more than a third of the present release of carbon to the atmosphere via fossil fuel combustion. The response in net primary production in relation to climate change is more difficult to predict and may enhance or reduce the effects caused by the estimated changes in the degradation features of peatlands (see Malmer, 1992).

6.5.3.2. Climatic Controls on Methane Flux

Non-Tidal Wetlands

flux. These studies have shown that the flux of CH, is moderately sensitive to changes in temperature and very sensitive to changes in moisture. Using these relative sensitivities as a guide, a qualitative assessment of CH, flux from northern wetlands according to six possible climate scenarios is made (Table 6-3). At present, it is not possible to obtain reliable quantitative estimates of the change in flux because the surface hydrology of general circulation models is too coarse to adequately represent the small changes in moisture regime that probably affect the CH, flux.

Emissions of N2O from northern wetlands are low. In situ chamber measurements and laboratory experiments with intact peat cores have revealed emissions below 0.025 g N2ON/m2/yr (Urban et al., 1988; Freeman et al., 1993; Martikainen et al., 1993). A lowering of the water table of bogs will not affect their N2O emissions, whereas it could strongly increase emissions from fens. Annual emission rates in the range of 0.05–0.14 g N2O-N/m2/yr have been reported for drained peat by Martikainen et al. (1993) and Freeman et al. (1993). The difference between bogs and fens can be explained partly by the fact that drained peat profiles of fens have the capacity to nitrify (Lång et al., 1994). N2O emissions from drained boreal fens are lower than those from drained agricultural organic soils but 10-100 times higher than the rates from coniferous forest soils (Martikainen et al., 1993).

The net emission of CH, from peatlands is dependent on how
much CH, is formed in the anaerobic parts of the profile and
the amount oxidized in the oxic zones. Because the position of
the water table and the associated capillary fringe determine
the thickness of the zones of production and oxidation of CH1,
the flux of CH, is intimately tied to the surface hydrology of
the wetland-which in turn is controlled by climate (precipita-
tion and evaporation) and the topographic and geologic setting
(surface and subsurface water flow). A decrease in water avail-
ability in the peat can lead to a decrease in CH, emissions
(Whalen et al., 1996; Sundh et al., 1994a, 1994b; Martikainen
et al., 1995; Roulet et al., 1993). Deeper penetration of oxygen
into the peat also will enhance the capacity of the peat to act as
a CH-oxidizing filter for CH, diffusing from the CH,-forming 6.5.4.
sources below. Changes in the direction or magnitude of any or
all of the controlling variables discussed above will affect the
CH, flux. A change in the total CH, flux from northern wet-
lands can be expected if the areal extent of wetlands changes,
the duration of the active period changes, and/or the per-unit-
area production or oxidation of CH, changes.

The relations among moisture content, temperature, and CH, flux in individual wetlands have received much attention (Bartlett et al., 1992; Crill et al., 1988; Dise et al., 1992; Moore and Knowles, 1989; Moore and Dalva, 1993; Moore and Roulet, 1993; Svensson, 1976; Svensson and Rosswall, 1984). These relations have been used to estimate qualitatively the year-to-year variation in the flux and the possible direction of change based on changes in temperature and precipitation obtained in 2 x CO2 scenarios (Table 6-3). Four different approaches have been used to address this issue: (1) correlation of the time series of CH, fluxes with the time series of temperature and moisture using interannu ́al data sets; (2) direct observations of changes in CH, flux in manipulation experiments that simulate expected changes in wetlands due to climate change; (3) modeling of variability of CH, flux using existing climate records and regressions between temperature and CH, flux; and (4) modeling of thermal and hydrological regimes of wetlands in 2 x CO2 climate scenarios and then modeling of change in CH, flux using regressions relating CH, flux to temperature and moisture in order to predict a change in

Case Study: Kalimantan

6.5.4.1. Background

Kalimantan is one of the largest islands (539,460 km2) in the Indonesian archipelago (see Box 6-3). The region has a humid tropical climate, with high temperatures and high precipitation. The peatlands of Kalimantan probably play a major role in determining local climate at the present time, although there is no substantial evidence to confirm this.

The largest wetland areas are found in low-lying alluvial plains and basins and flat-bottomed valleys. Most of the freshwater wetlands in the area are forested swamps, specifically either freshwater swamp forests or peat swamp forests (Silvius, 1989). The freshwater swamp forests are rich in epiphytes, rattans, and palms. They provide shelter for a range of rare and endangered species of wildlife, including numerous bird species. The peat swamp forests are a further developmental stage of the freshwater swamp forest. Deep peats are found in the central and western parts of the island (Sieffermann et al., 1988, 1992; Rielly et al., 1992). The peat swamp forests have a relatively high diversity of tree species, but the variety of wildlife tends to be poorer than in freshwater swamp forests (Whitten et al., 1987). Because of the high acidity of the peats and the fact that they are difficult to drain, peat swamp forests are of limited agricultural value (Silvius, 1989). Both swamp types are important watershed areas

Table 6-3: Potential changes in CH, flux from northern wetlands due to changes in the thermal and moisture regime (adapted with additions from Matthews, 1993).

Notes: AT change in temperature; AWT = change in water table; P = precipitation.

232

Non-Tidal Wetlands

Box 6-3. The Forested Swamps of Kalimantan: Effects on Habitat, Hydrology, and Carbon Cycling Background: The wetlands of Kalimantan are found in low-lying alluvial plains and basins and flat-bottomed valleys. Most of the freshwater wetlands in the region are classified as forested swamps, specifically as either freshwater swamp forests or peat swamp forests. The freshwater swamp forests are important in providing shelter for rare and endangered species; both types are important as watersheds and habitats for valuable tree species. They also are carbon sinks but release carbon into the atmosphere when water declines. The Kalimantan wetlands currently are stressed by the loss of the natural ecosystem through deforestation, drainage, and agriculture. These activities limit the buffering capacity of developed wetlands, causing changes that tend to be long-term and irreversible. However, these inland wetlands have escaped some interference because most of the human settlements are located along the coastal zones.

Possible impacts: Increased temperatures are likely to result in a longer period of reduced rainfall because higher temperatures will cause evapotranspiration to exceed precipitation. This is likely to have deleterious effects on the vegetation and hydrology of these wetlands. Climate change also could enhance peat losses in the region that currently result from human interference. On the other hand, increased precipitation in the dry season is likely to be beneficial, as the lower water levels that are typical in this season lead to a net loss of carbon into the atmosphere and an increased risk of fire (one of the greatest threats to their functioning).

Conclusion: It is possible that some measures may be taken in this region in the near future to abate the detrimental impacts caused by human (non-climate) stresses on these wetlands. The extent to which adaptations will be sufficient to counteract changes imposed by a changing climate as well cannot be determined. However, human activities that reduce the resiliency of these wetlands, as well as planned development of previously undisturbed areas of swamp forest, will likely lead to considerably enhanced carbon transfer from the wetlands to the atmosphere, even without climate change.

higher temperature and longer dry periods combine to produce a longer period when evapotranspiration exceeds rainfall and effective rainfall is greatly reduced. This, linked to increasing human activity on peatlands-such as timber extraction, agricultural development, and construction work-could have serious consequences.

Much of the Kalimantan lowland area is subjected to a distinct dry season from July to September or October in which there are high water losses from wetlands as a result of direct evaporation and evapotranspiration. Thus, water levels drop and peat oxidation occurs, with a net loss of carbon to the atmosphere. The spread of fire is one of the greatest threats to the functioning of these wetlands. Peat fires occur frequently in the region, creating palls of smoke sufficiently heavy to close local airports. In 1983-84, fires destroyed 3.5 Mha of both dipterocarp and peat swamp forest in Kalimantan, resulting in a direct economic cost of $2-12 million and an incalculable ecological cost (Maltby, 1986). The risk of fires spreading from cultivated to forest areas increases during the dry season and would increase if the dry season were extended.

6.5.4.3. Effects of Precipitation Change

Estimated precipitation changes for the area range from -20% during winter to +40% during the summer. The higher "summer" (i.e., dry-season) values are beneficial to the peatlands because this is the time when they experience the greatest drawdown of the water table. A decrease of precipitation in the wet season could be significant if, as a result, the dry season

Large-scale removal of the forests and drainage of the underlying peats could prevent further peat formation. The high peats of Kalimantan already appear to be degrading (Sieffermann et al., 1988) and losing carbon directly through oxidation to the atmosphere or indirectly in surface drainage waters, followed by oxidation of carbon compounds at a later stage. Climate change would most surely exacerbate this degradation, leading to peat losses in this region.

6.5.4.4. Remediation Possibilities

A forest-management project in Kalimantan currently being sponsored by the British Overseas Development Administration will include suggestions for the sustainable management of the peat swamp forests. The resulting guidelines should help preserve the forest cover on the deeper peats, particularly if extraction methods do not intensify. Indonesian authorities also are introducing stricter regulations and controls on unnatural fires resulting from illegal land clearance and settlement. However, in 1986-89, a feasibility study was carried out to investigate the potential of using deep peat to generate electricity.

The extent to which these potential management changes will be sufficient to counteract the effects of climate change cannot be determined. However, utilization of the forests on these peats and planned development of previously undisturbed areas will likely lead to considerably larger carbon transfer from wetlands to the atmosphere even without

Non-Tidal Wetlands

6.5.5.

Case Study: The Florida Everglades

6.5.5.1. Background

233

hydroperiod and nutrient regimes in the Everglades have
resulted in changes in the rate of peat accretion. Areas experi-
encing reduced hydrology (caused by overdrainage) exhibit
lower rates of accretion (1.6–2.0 mm/yr) compared to areas of
extended hydroperiod (2.8-3.2 mm/yr) (Craft and Richardson,
1993). Likewise, pollen analysis of peat cores indicates a
decrease in the extent of wetland vegetation such as sawgrass
and slough and a concurrent increase in terrestrial "weedy"
species (ragweed and pigweed) since drainage activities were
initiated (Bartow et al., 1994). Thus, future changes in the
Everglades ecosystem caused by global warming must be
interpreted in the context of recent anthropogenic alterations of
hydrology and nutrient regimes (see Box 6-4).

The Everglades is a 500,000-hectare freshwater peatland domi-
nated by vast expanses of sedge and sawgrass, interspersed with
shallow-water aquatic communities (sloughs), wet prairies, and
tree islands (Loveless, 1959; Gunderson, 1994). Peat accumula-
tion and the subsequent formation of the Everglades began
approximately 5,000 years ago as sea-level rise slowed after an
initial rapid rise during deglaciation (Gleason and Stone, 1994).
As recently as a century ago, the Everglades encompassed more
than 1,000,000 hectares, but drainage for agriculture and urban
development has resulted in the loss of more than half of the
ecosystem (Kushlan, 1989; Davis et al., 1994). The remaining
area has been dramatically altered by construction of impound- 6.5.5.2. Effects of Sea-Level Rise
ments, canals, levees, and water-control structures; the system
is managed, primarily, as a water source (Light and Dineen,
1994). During the wet season (June-November), excess water
from agricultural land and suburban areas is pumped into the
Everglades; during the dry season (December-May), the
Everglades serves as a water source (DeGrove, 1984). In addi-
tion, approximately 50% of the water from the Kissimmee
River/Lake Okeechobee complex-the "headwaters" of the
Everglades-is diverted by canals to the Atlantic Ocean and the
Gulf of Mexico before recharging the wetland (Light and
Dineen, 1994). Thus, the present-day Everglades is character-
ized by a general reduction in the hydroperiod (Fennema et al.,
1994; Stephens, 1984; SFWMD, 1992; Walters et al., 1992).
Long-term rates of peat accretion in the Everglades average
0.8-2.0 mm/yr, based on 14C dating of the basal peat
(McDowell et al., 1969) and 210Pb dating of peat cores (Craft
and Richardson, 1993). However, alterations of the natural

The most immediate effect of climate change will be acceler-
ated sea-level rise, resulting in saltwater intrusion into the
lower part of the glades from Florida Bay (Wanless et al.,
1994). Increased salinity would result in encroachment of salt-
tolerant wetland communities such as mangroves and salt
marshes. The areal extent of freshwater communities such as
sawgrass, slough, and wet prairie will decrease, and the amount
of organic carbon sequestered also will decrease. Another bio-
geochemical consequence of saltwater intrusion is a shift in
anaerobic decomposition away from methanogenesis toward
nitrate and sulfate reduction.

6.5.5.3. Effects of Temperature Change

Temperature is expected to increase from 0.5 to1.5°C, likely resulting in an increase in evapotranspiration—which may

Box 6-4. The Florida Everglades: Effects on Water Supply and Critical Habitats

Background: The Everglades is a freshwater peatland dominated by sedge and sawgrass, with sloughs, wet prairies, and tree islands. The Everglades is important as a habitat for wildlife, fish, and plant species and as a water source for the neighboring community. It is estimated that drainage for agriculture and urban development in the past century has resulted in a loss of more than half of the ecosystem, and the remaining wetlands have been altered by other construction and so forth.

Possible impacts: Sea-level rise is expected to be perhaps the most important variable that will affect the Everglades, causing saltwater intrusion that is likely to result in an encroachment of salt-tolerant wetland communities. This would decrease the areal extent of the freshwater wetlands, with some effects on anaerobic decomposition. The increased evapotranspiration expected in some seasons would exacerbate this saltwater intrusion. Increased temperature is likely to cause a northward migration of some introduced species but could be conducive for other species. Climate change also is expected to affect the hydrology of the wetlands, causing higher water levels in the winter and lower levels in the summer. This could result in the loss of critical habitats such as sawgrass and wet prairie communities, although these losses are likely to be offset by an increase in woody shrubs and trees.

Conclusion: Overall, the impacts that are projected as a result of climate change would adversely affect the end-users of the ecosystem: waterfowl, fish, and other wildlife, hunters, fishers, and tourists; and surrounding populations that rely upon the Everglades for freshwater resources. However, some action is currently underway to modify the water-control