Grantee Research Project Results
2009 Progress Report: Ensemble Analyses of the Impact and Uncertainties of Global Change on Regional Air Quality in the U.S.
EPA Grant Number: R833369Title: Ensemble Analyses of the Impact and Uncertainties of Global Change on Regional Air Quality in the U.S.
Investigators: Lamb, Brian , Theobald, David M. , Guenther, Alex , Wiedinmyer, Christine , Mass, Clifford , McKenzie, Donald , Salathe, Eric
Current Investigators: Lamb, Brian , Mass, Clifford , Guenther, Alex , Wiedinmyer, Christine , Salathe, Eric , McKenzie, Donald , Chung, Sandra
Institution: Washington State University , National Center for Atmospheric Research , USDA , Colorado State University , University of Washington
Current Institution: Washington State University , National Center for Atmospheric Research , University of Washington , USDA
EPA Project Officer: Chung, Serena
Project Period: February 1, 2007 through January 31, 2011 (Extended to January 31, 2012)
Project Period Covered by this Report: February 2, 2009 through February 1,2010
Project Amount: $899,987
RFA: Consequences of Global Change For Air Quality (2006) RFA Text | Recipients Lists
Research Category: Climate Change , Air
Objective:
This project builds on results from a previous EPA global change project (RD838309). Our overall goal is to answer questions, as initially posed in our previous project, related to the effects of global change on continental and regional air quality and to include quantitative estimates of uncertainties as part of the answers to our research questions. We will employ an ensemble modeling approach with three specific objectives: 1) to develop a quantitative measure of the uncertainty in our modeling framework using ensemble modeling methods in comparison to current 1995-2004 observations; 2) to project these uncertainties into the future for the period 2045-2054 and quantitatively address the uncertainties that accompany projections of future emissions, both global and U.S., including changes in landcover, urbanization, biogenic emissions, and fire emissions; and 3) to continue to address research questions that will determine the consequences of global change on U.S. air quality.Progress Summary:
Approach: Our work plan begins with Bayesian analyses of GCM/WRF/CMAQ model configurations for a base climate period (1995-2004) to produce weighted ensemble members based upon their skill in representing observed climate and air quality. This analysis will reduce the number of ensemble members for future climate runs to those that provide significant skill to the overall composite. This reduced set will be combined with a range of potential emission scenarios in a factorial design, to predict both expected values and uncertainties in future air quality. The CMAQ model will enable us to estimate future air quality conditions in terms of photochemical gas ambient concentrations, levels of fine and coarse particulates, and the deposition rates of N, S, and Hg species. Modeling analyses using WRF/CMAQ will primarily address the continental U.S. with 36-km grid resolution, with some simulations conducted for the Pacific Northwest with 12-km grid resolution. An important aspect of the latter scale will be examination of future impacts in Class 1 wilderness areas, as specified in EPA’s Regional Haze Rule.
1. WRF Simulations
A suite of regional climate simulations for this project has been made using the WRF mesoscale model. The model uses multiple nests at 108 km, 36 km, and 12 km horizontal grid spacing (Figure 1) and 31 vertical levels. High vertical resolution (~ 20 – 100 m) is used in the boundary layer, and the model top is fixed at 50 mb. Details of the model setup are reported in Salathé, et al. (2010).
This project relies on an ensemble of climate simulations to provide a basis for estimating uncertainty in future projections of US air quality. Currently, four simulations have been completed as part of this project: two simulations forced by reanalysis fields (both for NCEP/NCAR and NCEP/DOE reanalysis data) and two simulations forced by global climate model output (CCSM3 and ECHAM5) using the SRES A1B projection of future greenhouse gas concentrations. Additional simulations using different configurations of the WRF model are under way. These simulations will provide regional scenarios using multiple large-scale boundary conditions and multiple implementations of the regional model.
We plan to perform a Bayesian ensemble analysis of the model results to characterize the uncertainties in the climate and air quality projections. An ensemble of at least five members will be required. Two additional regional climate simulations are currently under way and will be completed during summer 2010. The first duplicates the ECHAM5 A1B forcing with a variation of the WRF model and the second will use the WRF variant with another ECHAM5 variant. This set of simulations will explore the ensemble space due to changes in both the regional model and the large-scale forcing model. The Bayesian analysis will be based on the simulations of current climate (2003-2007) for each ensemble member and will be used to weight the future projection for the associated ensemble member (i.e., for the same pairing of regional and global models). The Bayesian analysis is based on comparisons between simulated and observed climate statistics. Preliminary assessment of several model parameters has been performed using observations from the Historical Climate Network (HCN) (Karl, et al., 1990). These results are reported in Zhang, et al. (2009) and Dulière, et al. (2010a). These initial studies provide the basis for the more formal Bayesian analysis.
WRF Validation Experiments
The nested WRF domains (108-36-12 km) were run for a 5-year period (2003-2007), driven by both the NCEP/NCAR and NCEP/DOE reanalysis data, and were compared with the observations from the Historical Climatology Network (HCN) to validate the model simulations and quantify the model biases. The WRF simulations also were compared with the HadRM (Hadley Center Regional Model) simulations driven by the same reanalysis data and time period and similar model domain (25-km grid spacing) to address the effect of the regional model on the simulations. The reanalysis data incorporate all the available observations at the time of processing and are particularly suitable for driving regional climate models and for model validation. Results from this study are reported in Zhang, et al. (2009) and Dulière, et al. (2010a).
WRF Climate Experiments
This project will be based on an ensemble of WRF/SMOKE/CMAQ simulations for the current (1994-2005) and future (2044-2055) climate. We have completed two ensemble members for each period using the base WRF configuration and two forcing models (ECHAM5 and CCSM3). The present-day (1994-2005) simulations will serve as the basis for model validation in terms of seasonal and annual variability, and future (2044-2055) simulations will be examined for regional climate change and local responses in future climate. Simulations for future climate use the IPCC Special Report on Emissions Scenarios (SRES) A1B emissions scenario, which assumes a balanced increase in greenhouse gas concentrations. The Hemispheric WRF domain was run for two 11-year time slices, 1994-2005 and 2044-2055, driven by the ECHAM5 global climate model simulations. These runs have been completed and results are reported in Salathé, et al. (2010), Zhang, et al. (2010), and Dulière (2010b).
Maps of the change in winter (DJF) and summer (JJA) 2-m air temperature and precipitation from current to future climate are shown in Figure 2. Warming is generally greater during the winter than summer, but in both cases is on the order of 2~3°C over the continental U.S. except for the Pacific Northwest, where warming is on the order of 1~2°C. Large warming (> 5°C) is identified over the Hudson Bay area of Canada during the winter. Precipitation changes are generally small over the U.S. and Canada except for the Cascade Range, where negative changes are evident.
Figure 1. Hemispheric domain at 220-km grid spacing (top) and continental domain at 36-km grid spacing (bottom).
Figure 2. WRF simulated annual mean changes between 2050s and 1990s of surface air temperature (°C) for (a) winter (December-January-February) and (b) summer (June-July-August) and of precipitation (mm/day) for (c) winter (December-January-February) and (d) summer (June-July-August) over the continental domain at 36-km resolution.
2. CMAQ Simulations
To address all of the perturbations represented in the future simulations, we have constructed a matrix of simulations and are in the process of completing each simulation in this matrix. The intent is to build on the type of attribution analysis we completed in our first grant with a more complete analysis using an ensemble of global model and scenario configurations.
Table 1. Matrix of simulations for current and future climate; chemical boundary conditions; US anthropogenic, biogenic, and fire emissions; and land use for the ECHAM5 A1B global model/scenario configuration. The number (#) column for the 36-km simulations corresponds to the 220-km simulation used to drive the 36-km chemical and meteorological boundary conditions. Simulations investigating the impact of fires on air quality are highlighted.
There is a similar matrix of runs for the ECHAM5 B1 and CCSM A1B and B1 configurations.
2.1 Biogenic Emissions
2.1.1 Landuse Changes Our previous future biogenic emission scenarios included a substantial drop in terpenoid emissions due to cropland expansion. The cropland data used for these scenarios appears to have overestimated future croplands. The distribution shown in Figure 3 reflects a more realistic estimate of cropland distributions and was generated by combining three datasets: the IMAGE 2100 global cropland extent dataset, the SAGE maximum cultivable land dataset, and the MODIS current cropland data. The IMAGE 2100 dataset was created from the output of a land cover model (LCM, Zuidema, et al., 1994) that forms part of a sub-system of the IMAGE 2.0 model of global climate change (Alcamo, 1994). The LCM is forced by projected economic and climatic factors as output from a terrestrial vegetation land cover model (TVM) and an agricultural demand model (ADM). The ADM generates projected economic data including regional consumption of commodities, projected population growth, changes in income, and regional trade. The TVM is forced by local soil characteristics and predicted changes in temperature and precipitation, and provides information regarding land suitability for a range of different agricultural activities. When combined with the regional demands for agricultural production simulated by the ADM, the TVM drives predicted land cover conversions within the LCM.
The SAGE cultivable dataset (Ramankutty, et al., 2002) was created using a ~1992 global cropland dataset (Ramankutty and Foley, 1998) modified by characterizing limitations to crop growth based on both climatic and soil properties. The future global cropland extent distribution was generated by analyzing predicted changes in agriculture on a continent-by-continent basis (using the IMAGE data), and then applying these changes to the MODIS based cropland map (used for present day MEGAN simulations), using the SAGE maximum cultivable dataset as an upper limit to cropland extent. The rationale for this procedure was that the expansion of agricultural activities predicted by the IMAGE study resulted in a map of future cropland extent that was somewhat unrealistic (i.e., there was expansion of agriculture into unsuitable areas) even though the IMAGE study provided some intriguing predictions of agricultural growth driven by future socioeconomic trends. We considered the SAGE maximum cultivability dataset to be a realistic upper-bound to agricultural expansion, as the soil and climate data used to create the SAGE dataset were more reliable than those used in the IMAGE study. However, we replaced cell values within the SAGE dataset with values from the 2001 MODIS cropland map for all cells where the MODIS crop cover value was higher than what was indicated by the SAGE dataset. This adjustment was made to reflect cultivable irrigated regions not included in the SAGE dataset.
In addition to generating a future crop cover dataset to simulate potential biogenic VOC emissions using MEGAN, future datasets representing several other MEGAN driving variables were developed, including geo-gridded potential future emission factor (EF) maps for isoprene and several other terpene compounds as well as future-extent maps of four non-crop plant functional types (PFTs): broadleaf trees, needle-leaf trees, shrubs, and grasses. For regions outside of the U.S., the non-crop PFT distributions were generated by reducing the current extent of each non-crop PFT map by an amount that would appropriately offset the predicted cropland expansion for a given continent. For the U.S., future non-crop PFT maps were generated using Mapped Atmosphere-Plant-Soil System (MAPSS) model output (http://www.fs.fed.us/pnw/corvallis/mdr/mapss/, Neilson, 1995), based on three GCM future scenarios. Present-day MAPSS physiognomic vegetation classes were associated with current PFT fractional coverage estimates by dividing the US into sub-regions and averaging existing (MODIS-derived) geospatially explicit PFT data within each sub-region as a function of MAPSS class. Sub-regions were created based on Ecological Regions of North America (http://www.cec.org/). After every current MAPSS class had been assigned PFT-specific fractional coverage estimates, future cover was determined by re-classifying future distribution maps for the 3 MAPSS datasets, using the fractional PFT cover estimates for each MAPSS class (within each ecological region), and averaging the three resultant future datasets into a single estimate of future non-crop cover for each PFT.
For the eastern U.S., future isoprene and monoterpene EF maps were constructed using changes in tree species composition as predicted by the USDA "Climate Change Tree Atlas." (CCTA, http://nrs.fs.fed.us/atlas/tree/). The CCTA data was based on the average of 3 GCMs, and we selected the CCTA datasets that represented the most conservative emissions scenarios available.
Using existing species-specific emission rate (ER) data, we applied anticipated changes in species compositions to ERs in our current EF maps on a state-by-state basis (since the Tree Atlas data were organized by state). As data were lacking on predicted species-level changes for areas outside of the eastern U.S., we did not attempt to alter our EF maps in these regions. The future isoprene emission factor distribution map, which reflects changes in PFT distribution and (for the eastern U.S.) species-related changes in EFs, is shown in Figure 4.
Additional activities to be completed with respect to future biogenic VOC emission estimates include the processing and analysis of additional GCM-driven MAPSS and CCTA datasets to better understand the variability and potential range of predicted of future biogenic VOC emissions.
Figure 3. Cropland distribution for present (top) and future (bottom) biogenic VOC emission scenarios. Red indicates low fraction of cropland and green indicates a high fraction.
Figure 4. Isoprene emission factor (EF) distribution for present (top) and future (bottom) biogenic isoprene emission scenarios. Red indicates lower EFs and green indicates higher EFs.
2.1.2 Current and Future Emissions The MEGAN biogenic emission model (Guenther, et al., 2006) was used to estimate current and future biogenic VOC and NOx emissions using current and future meteorology and current and future land use. Results for summertime isoprene emissions are shown in Figure 5 in terms of average summertime isoprene emissions and the associated standard deviation based upon 10 years of simulation for each decade. As expected, isoprene emissions occur at relatively high rates (>50 mg C/m2/day) in the eastern US and at much lower rates in the western US (<10 mg C/m2/day). The variability in these emissions among summers is on the order of 50%. When the emissions are projected to future climate conditions with current land use distributions, isoprene emissions are projected to increase, with the most noticeable increases occurring in the southeastern US. The average increase across the US is approximately 25% of the current decade's emissions. When future climate is combined with future land use, there are still increases in the eastern US, but the spatial extent of these increases is reduced and replaced with areas of decreased isoprene emissions. In this case, the average increase was approximately 12% of the current decade's emissions, compared to the 25% increase when changes in land use are not included. Thus, reductions in forest cover in the future are projected to lessen the overall increase in isoprene emissions due to a warmer climate. Similar results are shown in Figures 6-8 for summertime monoterpene, sesquiterpene, and soil NO emissions; however, for the areas where cropland extension is predicted, such as in eastern and central Texas, monoterpenes, sesquiterpenes, and NOx emissions are all predicted to increase, whereas isoprene emissions are predicted to decrease.
Figure 6. Monoterpene emissions for June-July-August: (a) average for the current decade; (b) standard deviation for the current decade; (c) difference between future and current decades without considering changes in land use; and (d) difference between future and current decades considering changes in land use.
Figure 7. Sesquiterpene emissions for June-July-August: (a) average for the current decade; (b) standard deviation for the current decade; (c) difference between future and current decades without considering changes in land use; and (d) difference between future and current decades considering changes in land use.
Figure 8. Biogenic NO emissions for June-July-August: (a) average for the current decade; (b) standard deviation for the current decade; (c) difference between future and current decades without considering changes in land use; and (d) difference between future and current decades considering changes in land use.
2.2. Fire emissions
Biomass burning emissions were included in the hemispheric emissions from the POET emission inventories (Granier, et al., 2005) and the black and organic carbon inventories described by Bond, et al. (2004). For the continental U.S. in our previous project, we employed historical fire data for the current decade (1990-1999) and the stochastic Fire Scenario Builder (FSB) to predict fire location and size for future conditions in 2045-2054. In the present project, we plan to continue our efforts to develop the FSB by compiling current decade fire emission data (now in 1995-2004) from historical records and also by using the FSB with WRF current decade meteorology. Using these two different sets of fire emissions data will provide an estimate of the uncertainties associated with the FSB based on current decade conditions. We will continue to use the FSB to estimate fire occurrence for future decade WRF simulations.
In a related way, we plan to update the way we estimate emissions for a given fire (whether from historical or FSB results) by using the new BlueSky fire framework, which recently has been introduced by the AirFire group working in conjunction with Sonoma Technology, Inc. (Larkin, et al., 2008). In the new framework, a SMARTFIRE module has been developed that estimates fire size and progression based on a combination of data sources, including satellite detects. The new framework produces SMOKE ready emission files and provides a more modular approach that allows different fire behavior, consumption, emission, and plume rise modules to be selected for a given application. One aspect that still needs to be completed is to allow the new BlueSky framework to work with MCIP meteorological files directly, rather than WRF files, to make it easier to use with CMAQ. This is simply a logistical issue because we are archiving MCIP files in a readily accessible form but have less ready access to WRF files.
2.3 Anthropogenic Emissions
Global emissions of gases (ozone precursors) from anthropogenic, natural, and biomass burning sources have been estimated for the period 1990-2000 (applied to 1995-2004) using the POET emission inventory project (Granier, et al., 2005). Anthropogenic emissions (containing 15 sectors) are based on national activity data, emission factors, and grid maps (e.g. population maps) for spatial distribution of the emissions within a country. ATSR active fire maps are used for the spatial and temporal distribution of the emissions.
The global tabulation for black and organic carbon (BC and OC respectively) was obtained from Bond et al. (2004), who uses emission factors on the basis of fuel type and economic sectors alone. The inventory includes emissions from fossil fuels, biofuels, open biomass burning, and burning of urban waste. The dependence of emissions on combustion practices is covered by considering combinations of fuel, combustion type, and emission controls, as well as their prevalence on a regional basis.
Emissions for the year 2000 from the POET, MEGAN, and Bond, et al. (2004) inventories were coupled, and the 16 gas phase POET and MEGAN species, along with the BC-OC aerosol species, were adapted to the SAPRC99 chemical mechanism. Diurnal patterns were developed and applied to the gridded emission inventories and processed using SMOKE. For the future decade hemispheric domain simulations, current decade emissions were projected to the year 2050 based on the Intergovernmental Panel for Climate Change (IPCC) A1B and B1 emission scenarios.
For the continental US simulations, US anthropogenic emissions for the current decade were based on the US EPA 2002 National Emissions Inventory (NEI). The 2002 NEI emissions were projected to 2050 using the MARKAL Allocation Model for a Baseline Scenario (Business as Usual).
MARKAL is a dynamic data-driven systems model that maps the energy economy from primary energy sources through their refining and transformation to the point at which a variety of technologies (e.g., classes of light-duty personal vehicles, heat pumps, or gas furnaces) service end-use energy demands (e.g., projected vehicle miles traveled, space heating). The Electric Generation Sector in the EPA National MARKAL Database characterizes existing and new technologies available for electricity generation. Based on estimated sector-specific electricity demand (residential, commercial, industrial, and transportation), fuel prices, technology costs, and environmental and operational constraints, MARKAL determines the most cost-effective way to meet the system’s electricity demand. MARKAL then quantifies the system-wide effects of changes in resource supply and use, technology availability, and environmental policy to determine the least-cost pattern of technology investment and utilization required to meet specified demands and model constraints before calculating the resulting criteria pollutant and greenhouse gas emissions (EPA, 2006).
The projected change in summer emissions for the future decade in comparison to the current decade is presented in Table 1. Emissions of NOx and SO2 are projected to decrease by roughly 15-20%, while emissions of CO, NH3, PM, and VOCs are projected to increase by as little as 1%.
Table 1. Projected increase in anthropogenic emissions in the US calculated using the Market Allocation Model (MARKAL) Baseline Scenario (Business as Usual)
2.4 CMAQ Simulations and Evaluation for Current Conditions
For current climate conditions, we have begun to evaluate the performance of the modeling system for summertime ozone and PM2.5. In these initial results, we obtained air quality observations from the AIRNOW network as indicated in Figure 9, and we have compared the distribution of these observations in Figure 10 to the distributions predicted for these sites for the 1996 simulation year, which is the warmest year in our current decade record.
For 8-hr daily maximum ozone, the simulation results agree reasonably well with the observed distributions. In particular, the peak values are well represented in five of the US regions, but in the Midwest and Northeast, the predicted peak ozone is higher than observed. This result is reasonable, considering 1996 is the warmest of our current decade simulations, and the modeled peaks are likely to be reduced as we add more summer simulations to this evaluation. For 24-hr daily average PM2.5 concentrations, the simulated distributions appear to underestimate the observed distributions in every region. While this may be attributed in part to 1996 being a particularly warm simulation year and the mismatch between 36-km grid cell values with point monitoring sites, the comparison is not as good as we obtained in our previous work (Chen, et al., 2009; Avise, et al., 2009) using a previous version of CMAQ (v4.4). We are still examining the reasons for these differences, and additional simulation years will be incorporated into this evaluation in the near future.
2.5 CMAQ Base Case and Attribution Results
In this section, we provide initial results from various attribution runs in comparison to the base case current simulation. This comparison is in terms of concentration distributions for daily maximum 8-hr ozone (Figure 11). Note that the current decade results are in red (E_CD_Base) and the future decade results are in blue (E_A1B_Base). The other individual members represent current conditions except for future chemical boundary conditions (BC), future meteorology (M), future meteorology with future land use (M_LU), and future US emissions (US). Changes in biogenic emissions in response to future climate and land use are reflected in the (M) and (M_LU) cases.
Figure 11. Daily maximum 8-hr ozone concentration distributions for base case and various future attribution members.
There is no consistent pattern among all of the regions except that ozone in the current base case is always slightly higher than in the future base case. Future chemical boundary conditions lead to an increase in median ozone in all regions except the Southeast, where no change is observed. In contrast, peak ozone increases only in the Northwest and Central regions, and decreases everywhere else. In regions with large sources of biogenic emissions (primarily the eastern half of the US), changes in climate enhance emissions, leading to higher ozone. In the West, climate change has the opposite affect and tends to reduce ozone. Results are very similar in all regions for the case with future meteorology and future land use compared to the case with only future meteorology. The case with future US emissions is always less than the current base case, and in all regions except the Northwest it is the lowest of all the attribution cases.
References:
Alcamo, J. (ed.), (1994), IMAGE 2.0: Integrated Modeling of Global Climate Change. Dordrecht, The Netherlands: Kluwer Academic Publishers.
Bond, T. C., D. G. Streets, K. F. Yarber, S. M. Nelson, J.-H. Woo, and Z. Klimont (2004), A technology-based global inventory of black and organic carbon emissions from combustion, J. Geophys. Res., 109, D14203, doi:10.1029/2003JD003697.
Granier, C., J.F. Lamarque, A. Mieville, J.F. Muller, J. Olivier, J. Orlando, J. Peters, G. Petron, G. Tyndall, S. Wallens (2005), POET a database of surface emissions of ozone precursors, available on internet at http://www.aero.jussieu.fr/projet/ACCENT/POET.php .
Guenther, A., T. Karl, P. Harley, C. Wiedinmyer , P. I. Palmer, C. Geron (2006), Estimates of global terrestrial isoprene emissions using MEGAN (Model of Emissions of Gases and Aerosols from Nature), Atmos. Chem. and Phys., 6, 3181-3210.
Larkin, N. K, S. O’Neill, R. Solomon, S. Raffuse, T. Strand, D. C. Sullivan, C. Krull, M. Rorig, J. Peterson, and S. Ferguson (2008). The BlueSky Smoke Modeling Framework. International Journal of Wildland Fire, in press.
Neilson, R. P. (1995). A Model for Predicting Continental-Scale Vegetation Distribution and Water Balance. Ecological Applications, 5(2), 362-385.
Ramankutty, N.; Foley, J.A.; Norman, J.; and K. McSweeney, (2002) The global distribution of cultivable lands: Current patterns and sensitivity to possible climate change, Global Ecology and Biogeography, 11, 377-392.
Ramankutty, N. and J.A. Foley, (1998) Characterizing patterns of global land use: An analysis of global croplands data, Global Biogeochemical Cycles, 12(4), 667-685.
U.S. EPA (2006), MARKAL Scenario Analyses of Technology Options for the Electric Sector: The Impact on Air Quality, Environmental Protection Agency, EPA #600-R-06-114 https://www.epa.gov/NRMRL/pubs/600r06114/600r06114.pdf, (accessed May 2010)
Zuidema, G.; van den Born, G.J.; Alcamo, J.; and G.J.J. Kreileman, (1994), Simulating changes in global land cover as affected by economic and climatic factors. Wat. Air Soil Pollut., 76(1-2), 163-198."
Journal Articles on this Report : 8 Displayed | Download in RIS Format
Other project views: | All 16 publications | 12 publications in selected types | All 12 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Avise J, Chen J, Lamb B, Wiedinmyer C, Guenther A, Salathe E, Mass C. Attribution of projected changes in summertime US ozone and PM2.5 concentrations to global changes. Atmospheric Chemistry and Physics 2009;9(4):1111-1124. |
R833369 (2008) R833369 (2009) R833369 (2010) R833369 (Final) R830962 (Final) |
Exit Exit |
|
Chen J, Avise J, Lamb B, Salathe E, Mass C, Guenther A, Wiedinmyer C, Lamarque J-F, O'Neill S, McKenzie D, Larkin N. The effects of global changes upon regional ozone pollution in the United States. Atmospheric Chemistry and Physics 2009;9(4):1125-1141. |
R833369 (2008) R833369 (2009) R833369 (2010) R833369 (Final) R830962 (Final) |
Exit Exit |
|
Chen J, Avise J, Guenther A, Wiedinmyer C, Salathe E, Jackson RB, Lamb B. Future land use and land cover influences on regional biogenic emissions and air quality in the United States. Atmospheric Environment 2009;43(36):5771-5780. |
R833369 (2009) R833369 (2010) R833369 (Final) R830962 (Final) |
Exit Exit Exit |
|
Duliere V, Zhang Y, Salathe Jr. EP. Extreme precipitation and temperature over the U.S. Pacific Northwest:a comparison between observations, reanalysis data, and regional models. Journal of Climate 2011;24(7):1950-1964. |
R833369 (2009) R833369 (2010) R833369 (Final) |
Exit Exit Exit |
|
Salathe Jr. EP, Leung LR, Qian Y, Zhang Y. Regional climate model projections for the State of Washington. Climatic Change 2010;102(1-2):51-75. |
R833369 (2009) R833369 (2010) R833369 (Final) |
Exit Exit Exit |
|
Salathe Jr. EP, Steed R, Mass CF, Zahn PH. A high-resolution climate model for the U.S. Pacific Northwest: mesoscale feedbacks and local responses to climate change. Journal of Climate 2008;21(21):5708-5726. |
R833369 (2009) R833369 (2010) R833369 (Final) |
Exit Exit Exit |
|
Zhang Y, Duliere V, Mote PW, Salathe Jr. EP. Evaluation of WRF and HadRM mesoscale climate simulations over the U.S. Pacific Northwest. Journal of Climate 2009;22(20):5511-5526. |
R833369 (2009) R833369 (2010) R833369 (Final) |
Exit Exit Exit |
|
Zhang Y, Qian Y, Duliere V, Salathe Jr. EP, Leung LR. ENSO anomalies over the Western United States: present and future patterns in regional climate simulations. Climatic Change 2012;110(1-2):315-346. |
R833369 (2009) R833369 (2010) R833369 (Final) |
Exit Exit |
Supplemental Keywords:
RFA, Scientific Discipline, Air, climate change, Air Pollution Effects, Environmental Monitoring, Ecological Risk Assessment, Atmosphere, air quality modeling, global change, Baysian analysis, climate models, atmospheric modelsProgress 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.