Grantee Research Project Results
2008 Progress Report: Impacts of Global Climate and Emissions Changes on U.S. Air Quality (Ozone, Particulate Matter, Mercury) and Projection Uncertainty
EPA Grant Number: R833373Title: Impacts of Global Climate and Emissions Changes on U.S. Air Quality (Ozone, Particulate Matter, Mercury) and Projection Uncertainty
Investigators: Liang, Xin-Zhong , Wuebbles, Donald J. , Lei, Hang , Zhu, Jinhong , Lin, Jintai , Hayhoe, Katharine , Patten, Ken , Caughey, Michael , Wang, Xueyuan , Tao, Zhining
Current Investigators: Liang, Xin-Zhong , Wuebbles, Donald J. , Williams, Allen , Lei, Hang , He, Hao , Kunkel, Kenneth , Caughey, Michael , Su, Senjian
Institution: University of Illinois Urbana-Champaign
Current Institution: University of Maryland - College Park
EPA Project Officer: Chung, Serena
Project Period: April 15, 2007 through April 14, 2011 (Extended to April 14, 2013)
Project Period Covered by this Report: April 15, 2008 through April 14,2009
Project Amount: $900,000
RFA: Consequences of Global Change For Air Quality (2006) RFA Text | Recipients Lists
Research Category: Climate Change , Air
Objective:
The objective of this study is to quantify and understand the impacts and uncertainties of global climate and emission changes, from the present to 2050 and 2100, on USA air quality, focusing on ozone, particulate matter and mercury. The original contribution of this research will derive from the application of a unique, state-of-the-art, well-established ensemble modeling system that couples a global climate-chemical transport component with a mesoscale regional climate-air quality component over North America. Both components incorporate multiple alternative models representing the likely range of climate sensitivity and chemistry response under the conceivable emissions scenarios to rigorously assess the result uncertainty. Each will be enhanced to contain a fully coupled model to study climate-aerosol interactions, focusing on how they affect USA air quality at the present and in the future.We propose to conduct 3 primary sets of experiments by the ensemble modeling system to achieve the proposed objective. Historical simulations of climate and air quality for the recent past will first be conducted for system validation and bias identification, and also used as the baseline reference for future projections. Future projections for 2050 and 2100 will then be made to quantify the individual and combined impacts of global climate and emissions changes on USA air quality. Finally, sensitivity experiments will refine understanding of relationships with major contributing source regions and types, and uncertainties associated with key conclusions. All experiments will focus on April-October when most air quality episodes occur (except for sensitivity studies on the PM and mercury annual cycle), and integrate for a period of 5-10 years to obtain reasonably robust statistics. Subsequent diagnostic studies will identify possible future changes, and their climate and emissions causes, in the frequency, duration, and extreme pollutant concentrations of adverse air quality episodes over the USA.
Through the proposed application of this unique ensemble modeling system, we will make a major contribution to a key goal of the EPA Global Change Research Program to quantify the effect and uncertainty of global changes on USA air quality. The advanced state of the system components will result in a more complete scientific understanding of complex interactions among global climate and emissions and USA air quality across a full range of spatial and temporal scales. We will build on recent achievements of our ozone study, including a developed modeling system, viable experiment design, effective modeling strategy and objective diagnostic approach, for ozone consolidation, aerosol elaboration and mercury exploration studies for use in designing future effective emission control strategies to meet the national standards.
Progress Summary:
• We have completed the RCM integration driven by the CCSM3 for both 2045-2055 and 2090-2099 under the IPCC A1Fi and A1B emissions scenarios. The result indicates that the RCM downscaling improve the present-day regional climate simulations (more realistic than the driving GCMs) and provides likely more credible (than GCMs) projections of future regional climate changes. We have found that the model biases in the present-day climate simulation depend on geographic regions and climate regimes and can be systematically propagated into future regional climate projections. Compare to surface air temperature, simulated precipitation has larger biases and wider spreads over Texas and the Southeast than the Midwest and California.
• We have investigated future changes in the annual cycles of surface air temperature and precipitation for the broad subdomains of the Northeast, Midwest, Southeast, California, and Texas projected by GCMs and downscaled by RCMs. In 2050s, the projected warming trends are shown clearly in all subdomains throughout the entire year except for the transition months (February and November), while precipitation are projected with slightly dry trends. The warming trends are significantly amplified in 2100s The projected trends of both surface air temperature and precipitation scatter more widely over all subdomains in 2100s as compared with 2050s.
• We have demonstrated that the RCMs’ downscaling reduces significantly driving GCMs’ present-climate biases and narrows inter-model differences in representing climate sensitivity and hence in simulating the present and future climates. Major model present-climate biases are systematically propagated into future-climate projections at regional scales. The result suggests that the nested RCM-GCM approach that offers skill enhancement in representing the present climate also likely provides higher credibility in downscaling the future climate projection.
• We have conducted RCM simulations to evaluate the probability of heat waves of unprecedented severity by the end of the 21st Century if a high emissions path is followed. All of the RCM simulations for high emissions, including the one driven by the low-climate sensitivity CGCM, produce major increases in heat wave intensity and frequency. If a lower emissions path is followed, the outcomes range from quite small changes to substantial increases in intensity. In addition to the direct effects of heat, urban air quality can also be adversely affected by heat waves because of the temperature dependence of relevant chemical reactions, causing secondary impacts on health. The projected more frequent and/or intense heat waves will have increased risks of adverse health consequences.
• We have completed the CAM-Chem global chemistry transport run driven by the NCEP reanalysis II meteorology data during 2001-2002. As compared with the EPA AQS measurements, the model captures the main characteristics of the geographic distributions of surface zone, particulate matter, and three major aerosols (sulfate, ammonium, and nitrate). The CAM-Chem produces large biases in surface ozone, with a systematic overestimation of 8-32 ppb. The biases, however, have been largely reduced as compared with the previous runs driven by the CCSM3 climate. The PM10 biases are generally in the range of -24 to -32μg/m3 as compared with the EPA AQS measurements. That systematic underestimation results because the model currently does not incorporated dust transport. For PM2.5 the CAM-Chem results are 8-24 μg/m3 higher than observations over the eastern U.S. In addition, the simulated current values of wet sulfate depositions agree closely with National Atmospheric Deposition Program (NADP) observations. The simulated peak value of nitrate wet deposition shows strong biases, indicating possible problems in the model representation of the nitrogen chemical cycle.
• We have carried out sensitivity experiments using CAM-Chem to determine the impacts of future changes in anthropogenic emissions on projected US air quality under the IPCC A1Fi, A1B and B1 emissions scenarios representing respectively the upper, middle and lower bounds of climate warming over the 21st century. The projected summer surface ozone changes by 2050 range from 6 to 12 ppb under A1Fi, from -2 to -20 ppb under A1B, and from -5 to -24 ppb under B1. Aerosol changes by 2050 show strong seasonal variability. Sulfate concentrations under all three scenarios show decreasing trends due to the emission control. Ammonium also shows a decreasing trend. However, nitrate concentration varies not only with NOx emissions but also with sulfate and ammonium concentration via sensitive chemical reactions. Under A1Fi, nitrate concentration almost doubled due to predominat role of the NOx emissions increase. In contrast, under A1B and B1 with significant NOx emissions reductions, large nitrate increases are projected in the Northeast for winter. This is because the reduced sulfate coupled with increased ammonia emissions result in more ammonia available to react with nitrate, and thus total ammonia increases and nitrate aerosol concentrations can be more than double.
• We have quantified by sensitivity experiments, from a global CTM perspective, the relative contributions from future changes of local emissions and intercontinental pollutant transport (ICT) to the U.S. ozone changes under A1fi, A1B, and B1 emission scenarios. The result shows that the impact of local emissions is much larger than that of ICT. Under A1Fi, the maximum surface zone change due to local emissions change is around 12 ppb, while transpacific transport contributes less than 5 ppb and limited to the western U.S. Under A1B and B1, the effects from local emissions changes are much larger than those from ICT. It implies that limiting local emissions are still crucial and most effective in controlling future surface ozone quality.
• We have developed air pollution emissions inventories for 2050 and 2100 by scaling the present ones based on the IPCC’s A1Fi and A1B future emissions scenarios. We used the scaling factors derived from the OECD90 regions for the U.S. and Canada, and the factors based on the ALM regions for Mexico. The future inventories were then processed through SMOKE to provide speciated, gridded, and temporalized hourly emissions input for CMAQ. We have completed processing emissions for the periods of 1995–1999 and 2048–2052 using the corresponding meteorological simulations by the CCSM3-driven RCM downscaling with the Grell cumulus parameterization scheme.
• We have conducted present full 5 years (1995-1999) integration of CMAQ driven by the RCM meteorology downscaled from the CCSM3 simulation to evaluate the regional air quality modeling system performance. The model well simulates the AQS observed MDA8 ozone monthly variation in the Northeast, especially for spring and summer, with biases less than 5 ppb. The model, however, underestimates the summer MDA8 in Texas and the Southeast by more than 10 ppb. As compared with limited observations from CASTNET and IMPROVE, the model captures the key distribution features of annual mean concentrations of ammonium, nitrate, and sulfate aerosols, more realistic than a few global models. As compared with the NADP measurements, the model well reproduces the wet deposition of these aerosols in both spatial distribution and concentration levels. The simulated wet deposition of the aerosols is greater by one order of magnitude than the corresponding dry deposition. This implies that the wet removal process is the major sink of the aerosols. The modeled mercury wet deposition is also in reasonable agreement with the sparse NADP data in the Northeast and Midwest, but underestimated in the Southeast due to the missing sources from Mexico and oceans.
• We have completed the CMAQ simulations for future full 5 years (2048-2052) under A1Fi and A1B scenarios with climate and/or emissions changes. We have found that the two scenarios project substantially different U.S. air quality changes. Below we summarize the major findings in terms of future U.S. air quality changes from a regional air quality model perspective, focusing on the differences between the two scenarios and also the contributions of climate changes only as compared with both climate and emissions changes.
• Under A1Fi with changes of both climate and emissions, summer is projected with substantial increases of MDA8 ozone by 7-11 ppb, largest in the Midwest followed by the Northeast, Southeast, Texas, and California; spring is also projected with systematic increases by 5 ppb in Texas and the Southeast and by 4 ppb in other regions. In case of climate changes only, increases are projected over all seasons with smaller magnitudes; in summer, the increases are about 6 ppb in the Midwest, 4 ppb in the Northeast, Southeast and Texas, and 2 ppb in California. On the other hand, under A1B with changes of both climate and emissions, substantial increases are projected in winter by 6 ppb in the Northeast and Southeast, 5 ppb in Texas and the Midwest, and 2 ppb in California; in contrast, summer is projected with systematic decreases by 2-4 ppb except for Texas of 2 ppb increases. In case of climate changes only, largest increases are projected in summer by 7 ppb in the Southeast, 6 ppb in Texas, 3 ppb in the Midwest and 2 ppb in the Northeast.
• Under the A1Fi scenario with changes in both climate and emissions, substantial PM2.5 decreases are projected to occur in the eastern U.S., especially the Southeast by 2-6 μg/m3 in summer and 2-4 μg/m3 in spring and autumn; PM2.5 increases slightly by below 2 μg/m3 in winter over most of the U.S. and in summer in the western U.S. Under A1B with changes in both climate and emissions, PM2.5 decreases in all seasons due to the significant reduction in aerosol precursor emissions; the decreases are largest in the eastern U.S., by 2-6 μg/m3 in summer and autumn, while relatively small in spring and winter; for all seasons, Mexico is projected with substantial PM2.5 increases by 2-10 μg/m3, especially in the south. In case of climate changes only, PM2.5 responses are about one order of magnitude smaller under both scenarios.
• Under A1Fi with changes in both climate and emissions, nitrate aerosols increases overall in all seasons; the increases are especially large in winter, about 0.85 μg/m3 in the Northeast and Southeast, 0.75 μg/m3 in the Midwest, and 0.60 μg/m3 in Texas; increases are also large in autumn and spring in the Northeast (0.5-0.7 μg/m3) and Midwest (0.3-0.5 μg/m3). Under A1B with changes in both climate and emissions, nitrate aerosols decreases overall in all seasons; the decreases are especially large in autumn, 1.1 μg/m3 in the Midwest, 0.6 μg/m3 in the Northeast, and 0.5 μg/m3 in the Southeast and Texas; decreases are also large in the Midwest for both spring and winter (~0.6 μg/m3). In case of climate changes only, nitrate aerosols decreases under both scenarios, with similar seasonal variations and regional contrasts, the largest decreases in autumn, by 0.6 μg/m3 in the Midwest, 0.5 μg/m3 in the Northeast and 0.3 μg/m3 in the Southeast and Texas.
• When both changes in climate and emissions are included, sulfate aerosol responses resemble closely in both seasonal variations and regional contrasts between A1Fi and A1B; both project large decreases in the Northeast, Southeast, and Midwest with similar magnitudes; these decreases are largest in summer (4.5, 3.5, 2.5 μg/m3), followed by autumn (2.5, 2.6, 1.5 μg/m3) and spring (2.0, 2.0, 1.0 μg/m3). In case of climate changes only, sulfate aerosol responses differ largely between the two scenarios; in summer, sulfate aerosols are increased by 0.2 (Southeast, Northeast) to 0.4 (Midwest) μg/m3 under A1Fi whereas decreased by 0.4 (Midwest, Southeast) to 0.7 (Northeast) under A1B; the autumn trends, however, are similar, both having increases in the Midwest (0.4 μg/m3) and Northeast (0.2-0.3 μg/m3), but decreases in the Southeast and Texas (0.2 μg/m3).
• When changes in both climate and emissions are incorporated, the two scenarios project opposite changes of nitrate wet deposition in the Midwest and Northeast: increases under A1Fi but decreases under A1B with similar magnitudes of 400-1000 mg/m2/year. The wet deposition changes resemble dry deposition in spatial pattern but with substantially larger magnitudes by a factor of 40-50. Thus the wet deposition is the predominant sink for nitrate aerosols. In contrast, both scenarios project sulfate wet deposition to decrease in the eastern U.S., with peaks of 1500 mg/m2/year in the Northeast and Midwest. The projected sulfate wet deposition changes are about 5-7 times bigger than those of dry deposition and the location of the peak changes also differ between the two.
• Under A1Fi with changes in both climate and emissions, the mercury wet deposition increases by 0.4-2.0 μg/m2/year in the Midwest and Northeast, while decreases by a similar magnitude in California. In contrast, under A1B, the mercury wet deposition decreases by 0.4-2.0 μg/m2/year in the Midwest and Northeast, while increases by a similar magnitude in the southern California, cross States and northwestern Mexico. In case of climate changes only, the spatial pattern of the projected changes remains the same, although the magnitude is somewhat weaker in the Midwest and Northeast. This suggests that the future total mercury wet deposition trend is mainly driven by climate changes, while emissions changes may cause enhancement in the Midwest and Northeast with heavy pollutants.
• Progress in completing project tasks is in line with our original proposal and in some cases we have accomplished more than originally proposed. With the support of this EPA STAR project, we have so far published 5 articles, submitted 2 articles and almost completed 1 manuscript for publication in peer-reviewed journals. We have also made 12 presentations in meetings and seminars.
Future Activities:
• Continue to conduct projection experiments to study impacts of global climate and emission changes, focusing on particular matters
• Continue to conduct sensitivity experiments to explore impacts of global climate and emission changes on mercury
• Diagnose outputs to quantify relative roles of global climate and emission changes
• Diagnose outputs to study the impacts of climate and emission changes on U.S air quality
• Publish the results in peer-reviewed journal articles, including:
- Future USA ozone projections due to global climate and/or emissions changes
- Future USA PM2.5 projections due to global climate and/or emissions changes
- Future USA dry and wet deposition projections for nitrate
- Future USA climate change projections: uncertainty assessment and reduction
- Wang, X., X.-Z. Liang, W. Jiang, Z. Tao, J.X.L. Wang, H. Liu, Z. Han, S. Liu, Y. Zhang, S. Peckham, and G. Grell, 2009: WRF-Chem simulation of East Asian air quality: Sensitivity to temporal and vertical emissions distributions. J. Geophys. Res. (to be submitted shortly).
Journal Articles:
No journal articles submitted with this report: View all 21 publications for this projectSupplemental Keywords:
climate change, emission, uncertainty, pollutant transport, ozone, particulate matter, mercury, aerosol, cloud, radiation, regional climate model, air quality model, RFA, Air, climate change, Air Pollution Effects, Atmosphere, air quality modeling, environmental monitoring, particulate matterRelevant Websites:
http://www.sws.uiuc.edu/atmos/modeling/caqims/Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.
Project Research Results
- Final Report
- 2011 Progress Report
- 2010 Progress Report
- 2009 Progress Report
- 2007 Progress Report
- Original Abstract
21 journal articles for this project