Some modeled responses of forest ecosystems to climate change suggest that forests could increase carbon storage during climate change (Xiao et al., 1997; Prinn et al., 1999). However, recent work has provided new experimental evidence on the response of vegetation uptake (NPP) and ecosystem losses (affecting both NEP and NBP) to observed changes in climate over the past century. The net balance between NPP (usually assumed to increase with warming, though this assumption is challenged by data offered by Barber et al., 2000), heterotrophic respiration (often assumed to increase with warming, though this is challenged, for example, by Giardina and Ryan, 2000, and Liski et al., 1999at least for mineral soil components) and disturbance releases (often ignored but shownfor example, by Kurz and Apps, 1999to be important in boreal estimates of NBP) is no longer as clear as the SAR asserts. Present research continues to improve scaling of localized responses (at the stand level, where increases and decreases in NPP and NEP are observed) to the global scale.
Research reported since the SAR confirms the view that the largest and earliest impacts induced by climate change are likely to occur in boreal forests, where changes in weather-related disturbance regimes and nutrient cycling are primary controls on productivity (Kasischke et al., 1995; Kurz et al., 1995; Yarie, 1999). The impacts are exacerbated by characteristic ecosystem time constants (rotation length, mean residence time of SOM, etc.) that are long compared to other forest ecosystems. In boreal forests, recent warming has been shown to change seasonal thaw patterns (Goulden et al., 1998; Osterkamp and Romanovsky, 1999); increase growing season length (Keeling et al., 1996; Myneni et al., 1997); and, if accompanied by summer drought, reduce NPP (Sellers et al., 1997; Barber et al., 2000). These trends are expected to continue on average (see Table 3-10), although short-term modulations will occur (e.g., in association with ENSO events) (Black et al., 2000).
The SAR (Kirschbaum et al., 1996) concludes that in lowland humid tropics, temperatures already are close to optimum ranges for year-round growth. Hence, an increase in temperature as a result of climate change is likely to have a marginal effect on forest processes. However, research in a lowland tropical forest in Costa Rica has shown that the annual growth in six major species of this forest (with markedly different growth rates and life histories), over a period of 13 years, was highly negatively correlated with annual mean minimum (nighttime) temperatures (Clark and Clark, 1999). Although annual tree growth varied among the six species, there was a highly significant interannual coherence of growth among species.
Other reports in this collection