Grantee Research Project Results
Final Report: Improvements in Emissions Inventories using Semi-Continuous Monitoring Data and Concentrations Field Analysis
EPA Grant Number: R834557Title: Improvements in Emissions Inventories using Semi-Continuous Monitoring Data and Concentrations Field Analysis
Investigators: Schauer, James J. , Turner, Jay R. , deFoy, Benjamin
Institution: University of Wisconsin - Madison , Washington University , Saint Louis University - Main Campus
Current Institution: University of Wisconsin - Madison , Saint Louis University - Main Campus , Washington University
EPA Project Officer: Chung, Serena
Project Period: June 1, 2010 through May 30, 2013 (Extended to May 30, 2014)
Project Amount: $499,777
RFA: Novel Approaches to Improving Air Pollution Emissions Information (2009) RFA Text | Recipients Lists
Research Category: Air Quality and Air Toxics , Air
Objective:
Summary/Accomplishments (Outputs/Outcomes):
Estimation of mercury emissions from forest fires, lakes, regional and local sources using measurements in Milwaukee and an inverse method
Under an earlier EPA STAR grant, the project team took hourly measurements of speciated mercury at an urban site in Milwaukee from June 2004 to May 2005. Although the data had already been analyzed and published, this project sought to mine the data for more information about emissions by combining it with numerical simulations using an inverse method. The measurements included Gaseous Elemental Mercury (GEM), Reactive Gaseous Mercury (RGM), and Particulate Mercury (PHg).
We performed high-resolution WRF simulations of the winds around Milwaukee using three domains with resolutions of 27, 9, and 3 km. These simulations were used to perform hourly back-trajectories from the measurement site with the FLEXPART model. We also performed forward simulations of the dispersion of gaseous elemental mercury from known sources using the CAMx model. This was done using biomass burning emissions from the Fire INventory from NCAR (FINN) model. We also used parameterizations of GEM outgassing from water surfaces to estimate the concentrations of GEM due to outgassing from the Great Lakes.
We developed an inverse model that is able to estimate corrections to the estimates of emissions from the lake and from biomass burning, and at the same time, from the area within approximately 600 miles of the measurement site. Those emissions were estimated on a polar grid using the back-trajectories, and represent an estimate of all point and area sources within the domain that were not explicitly simulated using CAMx. We showed in the publication that we could represent a Bayesian inversion with a least squares approximation that could be solved in minutes on a regular workstation. This enabled us to characterize the uncertainties of the results by running the model multiple times. We use the bootstrap algorithm to select times to be included in each realization of the model at random. Simulation errors can be correlated in time; if a model simulates a particular weather episode incorrectly, it will be wrong for several hours. We therefore use block-bootstrapping where we select at random which days of the year to include in the analysis. In this way, if we assume that measurement errors and model errors are uncorrelated from day to day, the block-bootstrapping algorithm randomly samples days with different errors. The variation in the inverse model results therefore represents the impact of those errors on the uncertainty. We show that the results of the inversion are statistically significant and we provide 95% and 50% confidence intervals on the estimated emissions.
Estimation of direct emissions and atmospheric processing of reactive mercury using inverse modeling
Having used the inverse model to analyze emissions of Gaseous Elemental Mercury based on measurements in Milwaukee, we then analyzed the speciated measurements: Reactive Gaseous Mercury (RGM), Particulate Mercury (PHg) and the sum of the two: Reactive Mercury (RHg). The basic structure of the inverse model was the same: we used forward Eulerian simulations with CAMx to simulate time series of RGM at the measurement site, and we used backward trajectories with WRF-FLEXPART to estimate emissions on a polar grid in the domain within 1,000 km of the site. We also performed simulations with a tracer for transport from the free troposphere to the surface. These suggested that transport from the free troposphere could account for approximately 15% of RGM at the surface.
Whereas GEM is a long-lived species that behaves as a passive tracer on the time scales of regional transport, RHg is both formed and deposited in the atmosphere. We therefore needed to expand the model to represent these processes. We did this by using a simple box model over Milwaukee to evaluate the formation of RGM from three different pathways: oxidation due to ozone, due to the hydroxyl radical, and due to bromine. We used measured or estimated concentrations of the primary species and current estimates of the reaction rates to estimate the production of RGM. The time series produced by the box model were included in the inverse model. In this way, the inverse model was able to estimate a correction factor on these three factors and hence estimate their importance in RGM formation based on the measurements.
Based on our model, we estimate that 50 to 70% of the reactive mercury at the measurement site was due to direct emissions. This underlines the importance of developing emission inventories for reactive mercury species. We detected direct emissions from both area sources and from forest fires. In terms of the chemical oxidation pathways, our results suggest that the estimates of the ozone oxidation pathway are adequate in current parameterizations, but that oxidation by the hydroxyl radical and by bromine may be underestimated.
Estimating sources of elemental and organic carbon and their temporal emission patterns using a Least Squares Inverse model and hourly measurements from the St. Louis-Midwest Supersite
For this project, we were interested in quantifying emissions from different source groups and evaluating the temporal pattern of the emissions on the diurnal and annual time scale. We obtained the emissions inventory for 2007 from the Lake Michigan Air Directors Consortium (LADCO) separated into On-Road, Non-Road, Marine/Aircraft/Railroad (MAR), “Other” and Point source emissions. We performed independent forward Eulerian simulations for each group using the CAMx model. In addition, we used the Fire INventory from NCAR (FINN) emissions of open burning and performed simulations with CAMx. These formed the bulk of the inputs to the inverse model. We then performed backward trajectories with WRF-FLEXPART to estimate emissions that may be entirely missing from the LADCO inventory.
The inverse model was refined to be able to estimate the diurnal and annual variations in the emissions for each source category. The model provides information on the weekday (WD) and weekend (SSH: Saturday, Sunday, Holiday) emissions by time of day and by month of the year. The reduction on the weekends can be clearly seen in both the LADCO inventory (solid lines with markers) and the inverse model results, although the inversion suggests that the weekend emissions should be further lowered in the inventory. The LADCO inventory was improved to represent the spring and fall increase in emissions due to agricultural equipment, and this was corroborated by the model. For OC, we see large increases during the summer and around midday. These represent the OC at the measurement site that is due to atmospheric formation, giving an estimate of the importance of secondary formation for this species.
Particle Nucleation in a Sulfur Dioxide Rich Plume Reaching the St. Louis-Midwest Supersite
Particle numbers were measured at the St. Louis-Midwest supersite as described in Qian et al., 2007. We are in the process of reanalyzing the data in combination with detailed meteorological measurements. There is a very clear southwest signature that was found to correspond to plume impacts from a local smelter. The color scheme of the wind rose shows that these events happen during the daylight hours. The wind rose in the top right shows the events with high SO2 but low particle numbers. These correspond to plume impacts from the smelter at night, suggesting that the daytime particle number peaks are indeed due to particle formation in the plume. On the bottom left are the events with high particle numbers and low SO2. These have a bimodal distribution showing events from both the northwest and from the south which are thought to be related to regional particle nucleation events.
Impact of Regional Transport on the Anthropogenic and Biogenic Secondary Organic Aerosols in the Los Angeles Basin
Daily PM2.5 OC, EC, water soluble organic carbon (WSOC), and organic molecular markers measurements from May 2009 through April 2010 at the University of Southern California (34º1’19.12”N, 118º16’38.41”W) in downtown LA, and results from a positive matrix factorization (PMF) analysis of these data, were used understand the role of regional transport on SOC concentrations (Heo et al. 2013a). A backward-trajectory analysis, coupled with the measurements and estimated source contributions, were used to evaluate the origins of SOC aerosols. The Hybrid Single-Particle Lagrangian Integrated Trajectory model (HYSPLIT 4.9 version) with the 12 km resolution North American Mesoscale Model (NAM-12) meteorological data was applied to calculate the back trajectories of air parcels. Five-day backward trajectories arriving at heights of 500 m above ground level at the monitoring site with an hourly interval were computed for each of the 24-hour integrated samples. The daily concentrations of organic molecular markers and source contributions were assigned to grid cells of 0.1º × 0.1º geographical coordinates (latitude and longitude) along the corresponding trajectories. In order to estimate the potential source regions associated with an increase in the organic molecular markers concentrations and source contributions, a potential source contribution function (PSCF) was applied.
The contributions of anthropogenic SOC from the Central Valley and the South Coast areas in California are clearly visible in the PSCF analysis. The results emphasize the effect of emissions outside of the LA Basin on SOC concentrations, and the fact that SOC in the LA Basin cannot be explained by local emissions alone.
Impact of topographic conditions on mobile source contributions to PM2.5 OC in the Los Angeles Basin
PM2.5 levels in the LA basin are among the highest in the USA. This area must create emission control plans for compliance with the PM2.5 air quality standard. However, it is very challenging to develop appropriate emission control strategies and to determine the major causal factors of an increase in PM2.5 levels in the region, because the elevated air quality episodes in the LA basin are typically associated with topographic conditions and complex ocean-land circulation. This study shows that local stable drainage flows and/or recirculation of air masses by mountains surrounding the LA basin can greatly increase the impact of mobile source emissions on levels of PM2.5, and further, that considering topographical and meteorological conditions can be critical in developing PM2.5 control plans in the region.
Daily mobile source contributions used in this study were decribed in detail in Heo et al. (2013). Meteorological simulations were performed with the Weather Research and Forecast (WRF) model and were initialized with the North American Regional Reanalysis. For this preliminary modeling study, the model was run on a domain with 9 km horizontal resolution and 40 vertical levels using version 3.3.1. Particle back-trajectories were calculated with FLEXPART (Stohl et al., 2005), using WRF-FLEXPART (Fast and Easter, 2006) for 2 days duration starting every hour of the daily mobile source contributions. One thousand particles were released per hour at 50 m above the ground, and were allowed to disperse in the domain using the WRF mixing heights and surface friction velocity. The particle positions were presented on retangular grids to provide a Residence Time Analysis (RTA, Ashbaugh et al., 1985) which represents the amount of time that an air mass has spent in different grid cells before arriving at the monitoring site. RTA grids can also be rescaled to yield the impact that a source in each grid cell would have at the receptor site (Seibert and Frank, 2004). A Concentration Field Analysis (CFA, Seibert et al., 1994) was applied to identify potential source regions using daily mobile soruce contributions and the RTA results. A column-CFA method, calculated by multiplying the time series of source contributions by the planetary boudary layer heights at the sampling site taken from WRF, was also tested in order to take a look at an effect of very stable atmospheric conditions on highlighting the source locations.
Journal Articles on this Report : 15 Displayed | Download in RIS Format
Other project views: | All 17 publications | 15 publications in selected types | All 15 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Croteau MN, Dybowska AD, Luoma SN, Misra SK, Valsami-Jones E. Isotopically modified silver nanoparticles to assess nanosilver bioavailability and toxicity at environmentally relevant exposures. Environmental Chemistry 2014;11(3):247-256. |
R834557 (Final) R834575 (Final) |
Exit |
|
Croteau MN, Misra SK, Luoma SN, Valsami-Jones E. Bioaccumulation and toxicity of CuO nanoparticles by a freshwater invertebrate after waterborne and dietborne exposures. Environmental Science & Technology 2014;48(18):10929-10937. |
R834557 (Final) R834575 (Final) |
Exit Exit Exit |
|
de Foy B, Schauer JJ. Origin of high particle number concentrations reaching the St. Louis, Midwest Supersite. Journal of Environmental Sciences 2015;34:219-231. |
R834557 (Final) |
Exit Exit Exit |
|
de Foy B, Wiedinmyer C, Schauer JJ. Estimation of mercury emissions from forest fires, lakes, regional and local sources using measurements in Milwaukee and an inverse method. Atmospheric Chemistry and Physics 2012;12(19):8993-9011. |
R834557 (2011) R834557 (2012) R834557 (Final) |
Exit Exit Exit |
|
de Foy B, Heo J, Schauer JJ. Estimation of direct emissions and atmospheric processing of reactive mercury using inverse modeling. Atmospheric Environment 2014;85:73-82. |
R834557 (2012) R834557 (Final) R829798 (2005) |
Exit Exit Exit |
|
de Foy B, Cui YY, Schauer JJ, Janssen M, Turner JR, Wiedinmyer C. Estimating sources of elemental and organic carbon and their temporal emission patterns using a least squares inverse model and hourly measurements from the St. Louis-Midwest supersite. Atmospheric Chemistry and Physics 2015;15:2405-2427. |
R834557 (Final) |
Exit Exit |
|
Heo J, de Foy B, Olson MR, Pakbin P, Sioutas C, Schauer JJ. Impact of regional transport on the anthropogenic and biogenic secondary organic aerosols in the Los Angeles Basin. Atmospheric Environment 2015;103:171-179. |
R834557 (Final) |
Exit Exit Exit |
|
Laycock A, Stolpe B, Dybowska A, Valsami-Jones E, Lead J, Rehkaemper M. Synthesis and characterization of isotopically labeled silver nanoparticles for tracing studies. ENVIRONMENTAL SCIENCE:NANO 2014;1(3):271-283. |
R834557 (Final) |
Exit Exit |
|
Laycock A, Diez-Ortiz M, Lamer F, Dybwoska A, Spurgeon D, Valsami-Jones E, Rehkaemper M, Svendsen C. Earthworm Uptake Routes and Rates of Ionic Zn and ZnO Nanoparticles at Realistic Concentrations, Traced Using Stable Isotope Labeling. ENVIRONMENTAL SCIENCE AND TECHNOLOGY 2016;50(1):412-419. |
R834557 (Final) |
Exit Exit |
|
Levy JI, Fabian MP, Peters JL. Meta-analytic approaches for multistressor dose-response function development: strengths, limitations, and case studies. Risk Analysis 2015;35(6):1040-1049. |
R834557 (Final) R834577 (Final) R834798 (Final) |
Exit Exit Exit |
|
Misra SK, Dybowska A, Berhanu D, Luoma SN, Valsami-Jones E. The complexity of nanoparticle dissolution and its importance in nanotoxicological studies. The Science of the Total Environment 2012;438:225-232. |
R834557 (Final) R834575 (2013) R834575 (Final) |
Exit Exit Exit |
|
Xiu Z-M, Zhang Q-B, Puppala HL, Colvin VL, Alvarez PJJ. Negligible particle-specific antibacterial activity of silver nanoparticles. Nano Letters 2012;12(8):4271-4275. |
R834557 (Final) R834575 (2013) R834575 (Final) |
Exit Exit Exit |
|
Yang Y, Mathieu JM, Chattopadhyay S, Miller JT, Wu T, Shibata T, Guo W, Alvarez PJJ. Defense mechanisms of Pseudomonas aeruginosa PAO1 against quantum dots and their released heavy metals. ACS Nano 2012;6(7):6091-6098. |
R834557 (Final) R834575 (2013) R834575 (Final) |
Exit Exit Exit |
|
Yang Y, Wang J, Zhu H, Colvin VL, Alvarez PJJ. Relative susceptibility and transcriptional response of nitrogen cycling bacteria to quantum dots. Environmental Science & Technology 2012;46(6):3433-3441. |
R834557 (Final) R834575 (2013) R834575 (Final) |
Exit Exit Exit |
|
Zhang Q, Hu Y, Casterson C, Jang W, Xiao Z, Bohloul A, Garcia-Rojas D, Puppala H, Bennett G, Calvin V. When function is biological:Discerning how silver nanoparticle structure dictates antimicrobial activity. ISCIENCE 2022;25(7):104475. |
R834557 (Final) |
Exit Exit |
Supplemental Keywords:
National Emissions Inventory, Toxic Release Inventory, emissions, air pollutantsProgress 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.