U.S. Environmental Protection Agency Office of Research and Development Projecting Changes in Climate and Air Quality for the Southeastern U.S. Chris Nolte 1, Tanya Spero 1, Russ Bullock 1, Jerry Herwehe 1, Megan Mallard 1, Jared Bowden 2, and Kiran Alapaty 1 1 National Exposure Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 2 Institute for the Environment, University of North Carolina, Chapel Hill, North Carolina Background Recent improvements in air quality in the United States have been due to significant reductions in emissions of precursors of ozone and particulate matter (PM), and these downward emissions trends are expected to continue in the next few decades. To ensure that planned air quality regulations are robust under a range of possible future climates and to evaluate possible policy actions to mitigate climate change, it is important to characterize and understand the effects of climate change on air quality (Jacob and Winner, 2009). Recent work by several research groups using global and regional models has suggested that there is a "climate penalty," in which climate change leads to increases in surface ozone levels in polluted continental regions (Weaver et al., 2009). One approach to simulating future air quality at the regional scale is via dynamical downscaling, in which fields from a relatively coarse global climate model (GCM) are used as input for a higher-resolution regional climate model (RCM), and these regional climate data are subsequently used for chemical transport modeling. In our group, dynamical downscaling techniques developed and evaluated using coarse-scale historical meteorology (Otte et al. 2012; Bowden et al. 2013; Bullock et al. 2014) are being applied to GCM outputs from the Intergovernmental Panel on Climate Change (IPCC) Fifth Assessment Report (IPCC, 2013) and used as input for the Community Multiscale Air Quality (CMAQ) model. We show changes in simulated concentrations of ozone and PM around 2030, with a particular focus over the Carolinas. Current Climate Evaluation (2000) Changes from 2000 to 2030 Conclusions and Future Work Alapaty K, Herwehe JA, Otte TL, Nolte CG, Bullock OR, Mallard MS, Kain JS, Dudhia J, Introducing subgrid-scale cloud feedbacks to radiation for regional meteorological and climate modeling, Geophys. Res. Lett. 39, 24, L Bowden JH, Nolte CG, Otte TL (2013), Simulating the impact of the large-scale circulation on the 2-m temperature and precipitation climatology, Clim. Dynam., 40, Bullock OR, Alapaty K, Herwehe JA, Mallard MS, Otte TL, Gilliam RC, Nolte CG (2014) An observation-based investigation in WRF for downscaling surface climate information to 12-km grid spacing, J. Appl. Meteor. Climatol., 53, Herwehe JA, Alapaty K, Spero TL, Nolte CG (2014), Increasing the credibility of regional climate simulations by introducing subgrid-scale cloud-radiation interactions, J. Geophys. Res., in press. IPCC (2013), Climate Change 2013: The Physical Science Basis, Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, Stocker TF et al., eds, Jacob DJ, Winner DA (2009), Effect of climate change on air quality, Atmos. Environ. 43:1, Otte TL, Nolte CG, Otte MJ, Bowden JH (2012), Does nudging squelch the extremes in regional climate modeling? J. Clim. 25, Weaver CP, et al. (2009), A preliminary synthesis of modeled climate change impacts on U.S. regional ozone concentrations, Bull. Amer. Meteor. Soc. 90, Chris Nolte l l PHOTO PHOOPHOTO Modeling Approach The Weather Research and Forecasting (WRF) model was used to dynamically downscale outputs from the NASA Goddard Institute for Space Studies (GISS) ModelE2 GCM over most of North America using 36 km × 36 km grid cells. Two 11-year periods were simulated: a contemporary climate period centered on the year 2000 and a period centered on 2030, following the IPCC Representative Concentration Pathway (RCP) 6.0. These meteorological fields were then used as inputs to the CMAQ model. Anthropogenic emissions of precursor pollutants (NO x, SO 2, and VOCs) were held constant at 2006 levels throughout both simulations to examine the changes in regional air quality resulting from simulated climate change. Figure 1. Seasonal average temperature bias with respect to North American Regional Reanalysis GISS WRF WinterSpringSummerFall Figure 2. Seasonal average precipitation simulated by GISS, WRF, and in NARR GISS WRF NARR Figure 3. Bias in average daily maximum and minimum temperature modeled by GISS and WRF, relative to observations from GHCN sites in the Carolinas. PM 2.5 SO 4 Figure 4. Multiyear average of 98 th percentile ozone mixing ratios at Clean Air Status and Trends Network (CASTNET) sites (observed, modeled, and model-obs difference, top); observed and modeled summer and winter concentrations of total PM 2.5 and SO 4 at Interagency Monitoring of Protected Visual Environments (IMPROVE) sites (bottom). Results and Discussion Temperatures obtained using WRF as an RCM to downscale the GISS ModelE2 are in good agreement with those from the 32-km North American Regional Reanalysis (NARR). There is an overall cool bias, which is greatest during the summer (Fig. 1). WRF adds value over the driving GCM data, providing greater spatial detail and generally reducing mean temperature biases. In addition, spatial correlations are improved and mean absolute errors are reduced for mean, minimum, and maximum daily temperatures (not shown). Precipitation patterns are well captured (Fig. 2), though there is an overall wet bias, which is greatest during the summer. Recent EPA work (not included here) on the interaction between radiation and parameterized clouds improves precipitation simulated by WRF (Alapaty et al. 2012, Herwehe et al. 2014). Average daily maximum and daily minimum temperatures are shown in Figure 3 for sites from the Global Historical Climatology Network (GHCN). WRF’s daily maximum temperatures are negatively biased, particularly in July, while daily minimum temperatures are positively biased in January and negatively biased in July. Mean maximum daily 8-h ozone (MDA8) levels (Fig. 4) are positively biased by 7 ppb (not shown). Concentrations at the upper tail of the ozone distribution have a lower bias. Particulate matter (PM) concentrations are negatively biased, particularly during summer, though sulfate (SO 4 ) concentrations are unbiased. days/ year WinterSpringSummerFall GISS WRF Figure 5. Changes in seasonal mean temperature over North America simulated by GISS and WRF Figure 6. Changes in seasonal mean of maximum daily 8-h average ozone and frequency of exceeding the National Ambient Air Quality Standard of 75 ppb. GISS ModelE2 and WRF both simulate warming over North America of K during summer (Fig. 5) and up to 3 K during winter. WRF is generally consistent with ModelE2, with sharper gradients and greater spatial details. The changes in regional climate result in increases of ppb in summer MDA8 ozone levels, particularly in parts of the Midwest and Ohio River Valley (Fig. 6). For this climate scenario, changes in air quality over the Carolinas are generally small, i.e., there is no significant “climate penalty.” Results from previous EPA-funded work on the effect of climate change in worsening ozone air quality contributed to the EPA Administrator’s Endangerment Finding for greenhouse gases, setting the stage for possible regulations to mitigate climate change. The present research is part of a larger ongoing study to examine air quality under an ensemble of global climate scenarios, encompassing multiple GCMs and future greenhouse gas trajectories. These results will also contribute to an interagency report on the human health implications of climate change, which is being conducted as part of the National Climate Assessment. References WinterSpringSummerFall