Climate Change 2001:
Working Group II: Impacts, Adaptation and Vulnerability
Other reports in this collection Production Forestry

Evaluation of effects of climate change on production forestry are constrained by uncertainties discussed in Section (i.e., state-of-the-art forest models and GCM predictions produce equal probabilities of "considerable forest dieback" and "enhanced forest growth"). Moreover, in many cases, site-specific conditions as well as history, human management, air pollution, and biotic effects (e.g., herbivory) are much stronger controllers of forest productivity than climate change or CO2 enrichment (Eamus and Jarvis, 1989; Aber and Driscoll, 1997; Ollinger et al., 1997; Goodale et al., 1998), especially in mid-latitudes. Finally, lands managed for timber production are likely to be less susceptible to climate change than unmanaged forests because of the potential for adaptive management (Binkley and Van Kooten, 1994).

In a broader assessment for the United States, Sohngen and Mendelsohn (1999) used several GCMs, a variety of ecological models, and a dynamic economic model (with adaptation) to assess climate change impacts on the U.S. timber market. Under a broad range of climate and ecosystem model predictions, economic changes were positive, as a result of generally positive impacts of climate change on U.S. forest production and the ability of producers to adapt. Disturbances from insects and fire were assumed to increase, but the study also assumed that there would be salvage logging followed by planting of the right species for a new climatic regime. In a sensitivity analysis, in which the possibility of replanting with incorrect species was considered, economic impacts became negative as a result of reductions in available stocks and increased regeneration costs.

The foregoing uncertainty raises questions about evaluating impacts and developing adaptation strategies. For example, Woodbury et al. (1998) used a climate change scenario derived from four GCMs, results from experimental studies, and a probabilistic regional modeling approach and estimate that there is a high likelihood that loblolly pine (a major timber production species) growth is likely to decrease slightly over a 12-state region of the southern United States. However, they also estimate that there is a substantial chance of either a large decrease or a large increase in growth. How can this information be used by the timber industry? Should managers assume that there would be no problems with loblolly pine plantations? Should they increase the area of these plantations? Should they convert plantations to a more mixed plantation community? Should plantations be converted to "natural succession"? Crippling uncertainty of this type may lead this production industry (as well as others) to disregard climate change as a factor in planning.

Similar questions have been raised in northwest Canada (see Section Within the Mackenzie Basin Impact Study (MBIS), debate about forest management concerned the scenario of reduced spruce yield and increased risk of losses from fire and insect damage (Hartley and Marshall, 1997; Rothman and Herbert, 1997). In the short term, a large number of pressing issues divert attention from long-term climate change (e.g., land-use planning, British Columbia's Forest Practices Code, treaty negotiations with aboriginal people, trade with the United States, protected area strategy). Adaptation to climate change requires information that is relevant to the context of the industry, particularly if there are implications for harvesting (Barrett, 1997; Fletcher, 1997).

Other reports in this collection