Grantee Research Project Results
2005 Progress Report: Impacts of Global Climate and Emission Changes on U.S. Air Quality
EPA Grant Number: R830963Title: Impacts of Global Climate and Emission Changes on U.S. Air Quality
Investigators: Liang, Xin-Zhong , Wuebbles, Donald J. , Williams, Allen , Huang, Ho-Chun , Zhu, Jinhong , Lin, Jintai , Hayhoe, Katharine , Kunkel, Kenneth , Caughey, Michael , Tao, Zhining
Current Investigators: Liang, Xin-Zhong , Wuebbles, Donald J. , Huang, Ho-Chun , Williams, Allen , Caughey, Michael , Kunkel, Kenneth , Zhu, Jinhong , Patten, Ken , Hayhoe, Katharine , Lin, Jintai , Tao, Zhining
Institution: University of Illinois Urbana-Champaign
EPA Project Officer: Chung, Serena
Project Period: March 23, 2003 through March 22, 2006 (Extended to September 22, 2007)
Project Period Covered by this Report: March 23, 2005 through March 22, 2006
Project Amount: $900,000
RFA: Assessing the Consequences of Global Change for Air Quality: Sensitivity of U.S. Air Quality to Climate Change and Future Global Impacts (2002) RFA Text | Recipients Lists
Research Category: Air , Climate Change , Air Quality and Air Toxics
Objective:
The objective of this research project is to better understand how global changes in climate and anthropogenic emissions affect U.S. air quality, especially tropospheric ozone (O3) and fine particulate matter (PM2.5). The ultimate goal is to account for both effects to enable state and local air quality planners to design realistic and effective emission control strategies to meet the National Ambient Air Quality Standards (NAAQS). Our research objective is to apply a state-of-the-art integrated modeling system that nests a global climate-chemical transport model with a regional climate-air quality model (RCAQ) over North America to quantify the individual and combined impacts on U.S. air quality of global climate and emission changes, from the present to 2020, 2050, and 2100. The RCAQ further includes four high-resolution subdomains over the U.S. Northeast, Midwest, West Coast, and Texas for a more detailed assessment of these impacts on local surface ozone, PM2.5, and their precursors. These are the target areas where high probabilities of exceeding the NAAQS for ozone and PM2.5 are anticipated. The objective will be accomplished by three primary sets of experiments. Historical simulations of climate and air quality will be conducted first for system validation and for use as a baseline reference for future projections. Future projections for 2020, 2050, and 2100 then will be made, where the system incorporates scenarios of global changes in climate and/or emissions (three runs per period) to quantify the individual and combined impacts of global climate and emission changes on U.S. air quality. The third set consists of sensitivity experiments to determine dominant source regions and types (by adding perturbations in the U.S. or global emission inventories), relative roles of episodic transport versus mean background elevation (using transient or mean chemical inflows in RCAQ), as well as uncertainties associated with key conclusions.
Progress Summary:
- We have demonstrated that the regional climate model (RCM) captures the spatial pattern and temporal variation of the dominant patterns that are overall more realistic than the driving R-2 reanalysis for both precipitation and temperature. The RCM downscaling is able to correct important R-2 biases, especially in precipitation. The actual downscaling skill is sensitive to the choice of the cumulus parameterization. In particular, for summer precipitation, the Grell scheme better reproduces interannual variation in the Midwest, whereas the Kain-Fritsch scheme is more realistic over the Southeast. For summer temperature, although the Kain-Fritsch scheme produces higher correlation coefficients of the spatial pattern and temporal variation, its credibility is discounted because of substantial systematic warm biases. These sensitivities will have important consequences for the Mesoscale Model 5-Based Regional Climate Model (CMM5) applications in seasonal-interannual climate prediction, future climate change projection and impacts studies including air quality modeling.
- We have demonstrated that the RCM, when driven by the National Center for Atmospheric Research Parallel Climate Model (PCM) (a coupled ocean-atmosphere general circulation model, CGCM) produces an improved simulation of precipitation and temperature in all seasons. This suggests that the PCM is simulating realistic large-scale circulation patterns, and that the RCM downscaling is able to improve the representation of important regional and local forcing factors. Therefore, the RCM provides a credible tool for downscaling CGCM climate simulations. This is particularly important for climate projection studies as proposed in this U.S. Environmental Protection Agency (EPA) Science To Achieve Results (STAR) project. The results also suggest that the RCM can reduce significantly the uncertainty caused by inadequate resolution and incomplete physics representation of local and regional processes in CGCMs. The RCM downscaling may produce regional climate changes that are different substantially from the driving CGCM projections.
- We have demonstrated that enhancing the emissions inventory used as Sparse Matrix Operator Kernel Emissions (SMOKE) input can affect significantly regional ozone simulations. The baseline dataset is the 1999 National Emissions Inventory (NEI99). The enhanced dataset consisted of NEI99, plus the Big Bend Regional Aerosol and Visibility Observational study (BRAVO) emissions inventory for northern Mexico, plus the atmospheric portion of the National Pollutant Release Inventory (NPRI) of major, primarily industrial, emissions sources in Canada. Using the RCM meteorology driven by the R-2 reanalysis for the 1998-199 summers, SMOKE produced separate base-case (United States only) and enhanced case (Canada + Mexico + United States) emissions input for the mesoscale air quality model (AQM). As one would expect, the simulated primary emissions (NOx, volatile organic compounds, etc.) increase when new sources are added. The emissions enhancement causes the AQM to produce important ozone differences by as much as an order of magnitude in certain regions, such as near the eastern Great Lakes.
- To understand the relative contributions of future changes in climate versus emissions on projecting U.S. air quality, AQM sensitivity experiments were conducted where the projected climate and/or emission changes are superimposed on the present conditions. Intercomparisons between these experiments identify the individual and combined impacts of global climate and emission changes on future U.S. air quality. For the A1Fi scenario, given both the climate and emissions changes, the AQM produces 6-12 ppb more ozone in the Gulf States and Northwest States and 4-6 ppb less ozone along the southwest U.S. border and the vicinity of New York. Approximate one-half of these changes are attributed to the emissions changes only, where the anthropogenic and biogenic emissions contribute at a similar magnitude. The responses for the B1 scenario are very different, where ozone generally is reduced from the present to 2050. When both the climate and emissions changes are incorporated, the ozone reduction is 2-8 ppb over the western and southeastern States. As compared with the A1Fi scenario, the impact from the climate changes is small and most of the ozone changes are caused by the decreased emissions.
- A similar sensitivity study has been carried out using the Model of Ozone and Related Tracers (MOZART) (a global chemical transport model). The result shows that, following the A1Fi or the B1 scenario, not only the magnitude but also the spatial distribution of the future ozone change vary in response to global climate and/or emissions changes. When both climate and emissions changes are incorporated, future U.S. ozone changes range from +3 to +27 ppb under the A1Fi scenario to -6 to +9 ppb under the B1 scenario in the mid- and late-21st century. The future A1Fi climate tends to reduce ozone concentrations near the coastal areas but enhance ozone concentrations in the inland areas. The future B1 climate, by comparison, tends to reduce ozone concentrations in the eastern United States but enhance ozone concentrations in the western United States. In addition, great future increases of ozone concentrations in Asia, in China under the A1Fi scenario or in Southeast Asia under the B1 scenario, will contribute to buildup of the U.S. surface ozone and increase the difficulty in attaining the ozone standard.
- We have found that the actual AQM model biases in daily mean ozone concentration are greater over the regions with a larger climate sensitivity. The biases in the Northeast are doubled over those of the Midwest for almost all cases; the biases in California are much larger than those in Texas when the lateral boundary conditions (LBCs) from the R-2 are used, although the differences are neglible when using the PCM. The results indicate that the acuracy of the RCM climate, depending on both the driving LBCs and model physics representation, plays an important role for the AQM performance in simulating air quality. As compared to the R-2 (the proxy of observations), the PCM climate simulation is adequate for a realistic RCM climate downscaling and subequent AQM air quality modeling. It is important, however, to establish a baseline under the present climate condition before the identical RCM/AQM modeling configuration is used to project future air quality changes. Given the unavoidable systematic model biases, it is more appropriate to evaluate air quality changes relative to these known biases for a more credible interpretation of the likely future projection.
- We have answered, from a global chemistry and transport model (CTM) perspective, several important questions relative to the U.S. ozone air quality, including ozone background, the impacts of short- and long-range pollutant transports, and future ozone change, recognizing and quantifying model biases and several possible causes for these biases including inaccurate emission inputs. Model evaluations show that MOZART generally shows a positive bias relative to surface ozone observations of 20-40 ppb over most of the eastern United States because of inaccurate emission and meteorological inputs, coarse model resolution, and likely problems in the treatment of chemical/physical processes, especially in the surface layer. By using the regional anthropogenic/biogenic emission data over North America, MOZART produces much lower surface ozone concentrations over the Northeast, the Midwest, and the southern Great Plains, and produces 5-10 ppb higher ozone concentrations over much of the southeast and the northwest. With appropriate modifications to eliminate the biases over the eastern United States, afternoon background ozone is estimated to be less than 45 ppb, mainly 20-40 ppb, over the United States. Contributions of anthropogenic emissions from Canada/Mexico decrease from 1-6 ppb near the political boundaries to less than 1 ppb in many inland locations. Contributions of anthropogenic emissions from Europe/Asia generally are less than 3 ppb over most U.S. regions.
- We have shown that the differences between the MOZART and AQM simulations under the present climate are systematic and substantial. The AQM significantly improves the ozone simulation over most of the continental United States. The small differences in both model results as driven by the PCM simulation versus R-2 assimilation suggest that the PCM meteorology is adequate for the RCM and AQM to produce realistic downscaling of the U.S. climate and air quality. Given its general overestimation of ozone concentration, the MOZART simulation must be improved before providing accurate chemical LBCs to the AQM for realistic downscaling. It also presents a challenge for how to use the MOZART result embedded with the AQM to understand the impacts on long-range transport on the U.S. air quality. This will be one of our future research foci.
- We have shown that substantial differences between the MOZART and AQM simulations also exist in their projections of the future ozone changes. These differences are highly region dependent and strongly vary under different emissions scenarios. It is not possible to determine which model projection, if any, is the more accurate outcome under a specific emissions scenario. On the other hand, the AQM, along with the improved meteorology (via the RCM downscaling) and emissions (by detailed SMOKE representation) inputs, has been demonstrated to more realistically reproduce the observed ozone distribution, whereas substantially larger biases (mainly overestimation) are identified with the MOZART. As such, we may anticipate that the AQM result is more credible, especially at the local-regional scales, than the MOZART.
- We have demonstrated that the long-range pollutant transport plays an important role on the U.S. air quality. Given the substantial biases in the MOZART and most current global CTMs, direct incorporation of the chemical LBCs from these model outputs may not necessarily improve the regional AQM simulation. In the MOZART case, the AQM performance actually is worsened over most of the United States except California. Much of the degradation results from the long-range transport across the northern and southern boundaries, where pollutants originating in Canada and Mexico are important contributors. More rigorous validation is warranted for global CTMs in reproducing the observer vertical distributions along the boundaries before their outputs can be credibly used to provide chemical LBCs driving the AQM for downscaling. Only then can the actual contribution of the long-range pollutant transport to the USA air quality can be quantified. For the same reason, extra caution must be taken in conducting sensitivity experiments and interpreting the results on how the long-range transport affects the U.S. air quality under the future projection scenarios of global climate and emissions changes.
- We have concluded that the uncertainty of future air quality projection is substantial, especially for the 2090s. The uncertainty results from the estimate of the emissions changes, the climate sensitivity of the global GCMs projection and regional RCMs downscaling, as well as the chemistry mechanisms of the global CTMs and regional AQMs. The biggest source of uncertainty, however, is identified with the emissions estimates, which can totally change the sign of ozone change, from substantial increases under the A1Fi scenario to large decreases under the B1 scenario. Our simulations demonstrate that, if no control action is taken to avoid the A1Fi emissions scenario, the Northeast and Midwest are likely to suffer a greater risk of future air quality degradation than California and Texas. Unexpectedly, our modeling results also identify the broad Southeast and southern Rockies as the areas with the most severe potential of violating the national ozone standard if emissions are not controlled. These later regions demonstrate a great sensitivity to both climate and emissions changes. The most effective solution is to control the future emissions toward the B1 scenario.
- Progress in completing project tasks is in line with our original proposal and in many cases we have accomplished more than originally proposed. With the support of this EPA STAR project, we have published five articles, submitted four articles, and almost completed four manuscripts and are preparing several others for publication in peer-reviewed journals. We also have made 10 presentations (for this reporting year alone) at major conferences including the U.S. Climate Change Science Program workshop and the annual meetings of the American Geophysics Union and the American Meteorology Society.
Future Activities:
The main activity during Year 4 of the project is to disseminate our results from this STAR project to the public, through publications in peer-reviewed journals and books and also presentations at major conferences and institutions. The effort will focus on diagnosis of all the model simulations, from the SMOKE for regional emissions, RCM for regional climate, AQM for regional air quality, and MOZART for global chemical transport, that have been completed and additional integrations that are necessary for better understanding of the result. In particular, we plan to publish our findings in the following topics:
- RCM downscaling skill on interannual precipitation and surface air temperature.
- RCM versus GCM simulations of heat waves.
- Climate conditions conducive to air quality problems.
- RCM downscaling uncertainties caused by climate sensitivity of the GCMs and RCM.
- Impacts of emissions in Canada and Mexico on U.S. air quality.
- Regional versus global model simulations of ozone today and tomorrow.
- Impact of the MOZART versus clean LBCs on U.S. air quality.
- Summertime U.S. background ozone climatology.
- Future ozone change: upper and lower limits.
- Future ozone change: contributions of climate and emission changes.
- Impacts of continental and intercontinental transports on U.S. ozone.
- Impacts of emission inputs on the performance of the MOZART CTM.
- Individual and combined impacts of global climate and emission changes on the U.S. air quality as well as associated uncertainties.
Journal Articles on this Report : 6 Displayed | Download in RIS Format
Other project views: | All 37 publications | 18 publications in selected types | All 18 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Huang H-C, Liang X-Z, Kunkel KE, Caughey M, Williams A. Seasonal simulation of tropospheric ozone over the Midwestern and Northeastern United States:an application of a coupled regional climate and air quality modeling system. Journal of Applied Meteorology and Climatology 2007;46(7):945-960. |
R830963 (2005) R830963 (2006) R830963 (Final) |
Exit Exit Exit |
|
Kunkel KE, Liang X-Z. GCM simulations of the climate in the central United States. Journal of Climate 2005;18(7):1016-1031. |
R830963 (2003) R830963 (2004) R830963 (2005) R830963 (2006) R830963 (Final) |
Exit Exit Exit |
|
Kunkel KE, Liang X-Z, Zhu J, Lin Y. Can CGCMs simulate the twentieth-century "warming hole" in the central United States? Journal of Climate 2006;19(17):4137-4153. |
R830963 (2005) R830963 (2006) R830963 (Final) |
Exit Exit Exit |
|
Liang X-Z, Li L, Dai A, Kunkel KE. Regional climate model simulation of summer precipitation diurnal cycle over the United States. Geophysical Research Letters 2004;31(24):L24208 (4 pp.). |
R830963 (2003) R830963 (2004) R830963 (2005) R830963 (2006) R830963 (Final) |
Exit Exit |
|
Liang X-Z, Pan J, Zhu J, Kunkel KE, Wang JXL, Dai A. Regional climate model downscaling of the U.S. summer climate and future change. Journal of Geophysical Research--Atmospheres 2006;111(D10):D10108 (17 pp.). |
R830963 (2005) R830963 (2006) R830963 (Final) |
Exit Exit |
|
Zhu J, Liang X-Z. Regional climate model simulation of U.S. soil temperature and moisture during 1982-2002. Journal of Geophysical Research-Atmospheres 2005;110(24):D24110 (12 pp.). |
R830963 (2004) R830963 (2005) R830963 (2006) R830963 (Final) |
Exit Exit |
Supplemental Keywords:
climate change, emission, pollutant transport, scale interaction, ozone, nitrogen oxides, sulfate, particular matter, regional climate model, air quality model,, RFA, Scientific Discipline, Air, Ecosystem Protection/Environmental Exposure & Risk, particulate matter, Air Quality, Air Pollutants, climate change, Air Pollution Effects, Chemistry, Monitoring/Modeling, Atmospheric Sciences, Atmosphere, Environmental Engineering, anthropogenic stress, aerosol formation, ambient aerosol, atmospheric particulate matter, atmospheric dispersion models, environmental monitoring, environmental measurement, meteorology, climatic influence, emissions monitoring, future projections, global change, ozone, air quality models, climate models, greenhouse gases, airborne aerosols, atmospheric aerosol particles, atmospheric transport, environmental stress, regional emissions model, ecological models, climate model, greenhouse gas, aerosols, atmospheric models, Global Climate Change, atmospheric chemistry, ambient air pollutionRelevant Websites:
http://www.sws.uiuc.edu/atmos/modeling/caqims/ Exit
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.