IPCC Fourth Assessment Report: Climate Change 2007
Climate Change 2007: Working Group II: Impacts, Adaptation and Vulnerability Consequences of changes in ecosystem structure and function for feedbacks to the climate system

Climate warming will decrease the reflectivity of the land surface due to expansion of shrubs and trees into tundra (Eugster et al., 2000); this could influence regional (Chapin et al., 2005a) and global climate (Bonan et al., 1992; Thomas and Rowntree, 1992; Foley et al., 1994; Sturm et al., 2005; McGuire et al., 2007).

Measurements show great spatial variability in the magnitude of sink (net uptake) or source (net release) status for carbon, with no overall trend for the Arctic (Corradi et al., 2005). In contrast, models suggest that overall the Arctic is currently a small sink of about 20 ± 40 g carbon m2/yr (McGuire et al., 2000; Sitch et al., 2003, 2007). The high uncertainties in both measurements and model projections indicate that the Arctic could be either a sink or source of carbon. Thus, currently circumpolar Arctic vegetation and the active layer are unlikely to be a large source or sink of carbon in the form of CO2 (Callaghan et al., 2005; Chapin et al., 2005a). They are, however, most probably a source of positive radiative forcing due to large methane emissions; even in tundra areas that are net sinks of carbon, significant emissions of methane lead to positive forcing (Friborg et al., 2003; Callaghan et al., 2005).

Higher temperatures, longer growing seasons and projected northward movement of productive vegetation are likely to increase photosynthetic carbon capture in the longer term, whereas soil warming is likely to increase trace gas emissions in the short term. Drying or wetting of tundra concurrent with warming and increased active-layer depth (see Section 15.3.4) will determine the magnitude of carbon fluxes and the balance of trace gases that are involved. Drying has increased sources in Alaska (Oechel et al., 2000), whereas wetting has increased sinks in Scandinavian and Siberian peatlands (Aurela et al., 2002; Johansson et al., 2006).

Models project that the Arctic and sub-Arctic are likely to become a weak sink of carbon during future warming (an increase in carbon storage in vegetation, litter and soil of about 18.3 Gt carbon between 1960 and 2080), although there is high uncertainty (Sitch et al., 2003; Callaghan et al., 2005). Increased carbon emissions from projected increases in disturbances and land use, and net radiative forcing resulting from the changing balance between methane and carbon dioxide emissions (Friborg et al., 2003; Johansson et al., 2006) are particular uncertainties. Wetting, from increased precipitation and permafrost thawing, is projected to increase fluxes of methane relative to carbon dioxide from the active layer and thawing permafrost (Walter et al., 2006).

Changes in forest area will lead to both negative and positive feedbacks on climate. According to one coupled climate model, the negative feedback of carbon sequestration and the positive feedback of reduced albedo interact. This model predicts that the central Canadian boreal forests will give net negative feedback through dominance of increased carbon sequestration, while in the forests of Arctic Russia decreased albedo will dominate, giving net positive feedback (Betts and Ball, 1997; Betts, 2000).