Effect of downscaling methodology on decision-making

Abstract:

There is increasing demand from decision makers for fine scale climate information that is relevant and useful for regional and local adaptation planning. While global climate models (GCMs) are vital for understanding large-scale changes in global circulation patterns, the horizontal resolution of a typical GCM is too coarse for use in local impact studies. A number of methods have been implemented to translate coarse GCM climate projections down to the regional and local scale. These range from the simplest delta approach to complex dynamical downscaling models. With so many diverse methods of downscaling now available, there is a need to perform robust comparisons and evaluations of the different techniques. In this study we explore how the choice of downscaling method may influence the climate change response of important impacts related variables. Our goal is to identify the uncertainty in future climate change associated with different downscaling methods. We then examine how the uncertainty associated with downscaling can affect vulnerability assessments and adaptation planning.

We focus on the impact of climate change to extremes in three sectors: forest fire risk management, heat stress and human health, and energy consumption by buildings. For each sector, an impacts relevant index is used to assess current and future risk. The Keetch-Byram Drought Index (KBDI) is used for fire, the Wet Bulb Globe Temperature (WBGT) is used for heat stress, and heating and cooling degree-days are used for energy consumption. Local climate changes have been calculated for each sector using four downscaling techniques: the delta method, a bias correction method (KDDM), the statistical downscaling model (SDSM), and dynamical downscaling with NARCCAP. Climate response surfaces (e.g. response of KBDI to changes in temp. and precip.) are generated at four locations in the United States. Response surfaces are a useful tool to help decision makers estimate the vulnerability to climate change. The downscaled climate changes are then layered over each response surface to see how likely future projections are and if the downscaling methods result in different levels of uncertainty. Differences in downscaling method may be a critical source of uncertainty that is currently overlooked by decision makers.