Downscaling Climate Models: Sharpening the Focus on Local-Level Changes

Abstract

Most climate change projections are developed using global-scale models that generate average temperature changes that can be expected to occur over decades and far into the future. These global models are unable to represent granular atmospheric features such as cloud cover, airborne particles, and local pollution sources. Yet these smaller details can have a big impact on local climate, which is one reason the effects of climate change are expected to vary depending on geographic location.1,2 Downscaled regional climate models (RCMs) provide grist for climate change adaptation planning at the local and regional level. © 2012 Joseph Tart/EHP; Map Resources; Ray et al.26 This graphic is for illustration purposes only and does not represent an actual RCM. “Downscaling” climate models are an attempt to bridge the gap between global and local effects by layering local-level data over larger-scale climate models. Downscaled modeling examines relatively small areas in detail—in some cases down to 25 square kilometers,3 a far higher resolution than that offered by global climate model simulations. The goal is to generate more locally relevant projections of long-term weather patterns for regions, states, and cities. In 2000, in one of the first projects to apply downscaling to simulate local effects, a small group of researchers designed and tested a system to estimate the health impacts of climate change on individuals in a specific location: New York City. Their analysis, developed under the banner of the New York Climate and Health Project (NYCHP),4 incorporated local heat and air-quality data as well as land-use data such as new development and roads, since those also impact surface temperature and air quality. Using these data, the team projected that higher-than-normal temperatures and resulting increases in ground-level ozone production brought on by climate change could result in a 4.5% increase in ozone-related deaths across the metropolitan area by the 2050s.5 They also calculated that summer heat–related mortality across the area might increase 70% on average over the same period.6 “Downscaling work provides a view of how climate change may impact health in the future, it begins to describe for us the range of possible answers to some of the public health questions that we have today, and it gives us a sense of different possible alternative futures,” says Kim Knowlton, a senior scientist with the Natural Resources Defense Council, who worked on the NYCHP. “At this point, downscaled modeling results are invaluable to adaptation planning, and the expertise in the modeling community is improving all the time.” Thorsten Wagener, an associate professor at The Pennsylvania State University who specializes in hydrology, says there is not a major research university in the United States that doesn’t have someone working on downscaled regional climate modeling (RCM) because the impact of climate change for planners and managers is a crucial question at the moment. Wagener says, “Once the IPCC [Intergovernmental Panel on Climate Change7] used models to project climate change in the future, scientists quickly moved into asking, ‘What does global warming actually mean for a person in the street, for the power supply, for our economy, for our health, for all sorts of things?’” In a few cases public health officials and resource managers have already begun using these downscaled data to develop climate-change adaptation plans.8