Climate change impact assessments must begin with decisions about the scope and scale of the assessment: What are the main policy issues? What and who are exposed to climate change impacts? What is the appropriate scaletime frame, geographical extent, and resolution? Considerable progress has been achieved since the SAR in raising such framing questions at the outset of an assessment cycle, often in conjunction with representative stakeholders (see Carter et al., 1994; Downing et al., 2000).
Methods for identifying policy issues include checklists and inventories, document analysis, surveys and interviews, and simulations. The process of determining the scope of assessment should be iterative. The project design should specify what and who is exposed to climate change impactseconomic sectors, firms, or individuals. Evaluation of adaptation strategies should be cognizant of actors involved in making decisions or suffering consequences.
The choice of temporal scales, regional extent, and resolution should be related to the focus of the assessment. Often, more than one scale is required, under methods such as strategic scale cycling (Root and Schneider, 1995) or multi-level modeling (e.g., Easterling et al., 1998). Linkage to global assessments may be necessary to understand the policy and economic context (e.g., Darwin et al., 1995).
The most common set of methods and tools remains various forms of dynamic simulation modeling, such as crop-climate models or global vegetation dynamic models. A major improvement in impact modeling has been applicaton of process-oriented models, often with geographically explicit representations, instead of models that are based on correlations of climatic limits. Data for running and validating models is a recurrent issue. Intermodel comparisons have been undertaken in some areas (e.g., Mearns et al., 1999), but much remains to be done.
Climate change is likely to have multiple impacts across sectors and synergistic effects with other socioeconomic and environmental stresses, such as desertification, water scarcity, and economic restructuring. Most studies (especially as reported in the SAR) have focused on single-sector impacts. Relatively few studies have attempted to integrate regionally or even identified segments of the population that are most at risk from climate change.
Vulnerability assessment may be one way of integrating the various stresses on populations and regions arising from climate change (see Briguglio, 1995; Clark et al., 1998; Huq et al., 1999; Kaly et al., 1999; Mimura et al., 2000; Downing et al., 2001). There are some areas in which formal methods for vulnerability assessment have been well developed (e.g., famine monitoring and food security, human health) and applied to climate change. However, methods and tools for evaluating vulnerability are in formative stages of development.
Further development of methods and tools for vulnerability assessment appears warranted, especially for the human dimensions of vulnerability, integration of biophysical and socioeconomic impacts, and comparison of regional vulnerability. Conceptual models and applications of the evolution of vulnerability on the time scale of climate change are required. Formal methods of choosing indicators and combining them into meaningful composite indices must be tested. Combining qualitative insight and quantitative information is difficult but essential to full assessments. Finally, improved methods and tools should facilitate comparison of vulnerability profiles between at-risk regions and populations and highlight potential reductions in vulnerability, through policy measures or the beneficial effects of climate change.
Other reports in this collection