Grantee Research Project Results
Final Report: Constraining ammonia emissions and PM2.5 control efficiencies with a new combination of satellite data, surface observations and adjoint modeling techniques
EPA Grant Number: R834559Title: Constraining ammonia emissions and PM2.5 control efficiencies with a new combination of satellite data, surface observations and adjoint modeling techniques
Investigators: Henze, Daven K
Institution: University of Colorado at Boulder
EPA Project Officer: Chung, Serena
Project Period: May 1, 2010 through April 30, 2013 (Extended to April 30, 2014)
Project Amount: $249,942
RFA: Novel Approaches to Improving Air Pollution Emissions Information (2009) RFA Text | Recipients Lists
Research Category: Air Quality and Air Toxics , Air
Objective:
The project goal is to reduce existing uncertainties in current estimates of NH3 emissions in order to better characterize and control distributions of fine particulate matter (PM2.5) and reactive nitrogen. The specific research objectives are to:
- quantify the magnitude and variability, both geographical and seasonal, of US NH3 emissions at a high spatial resolution.
- provide detailed estimates of PM2.5 control efficiencies and how they will evolve owing to regulations that alter key balances among inorganic particulate species.
Summary/Accomplishments (Outputs/Outcomes):
In the first year of this project, progress was made in utilizing remote sensing observations to understand distributions of NH3. Initial efforts focused on characterization of retrievals from the TES satellite instrument in terms of instrument sensitivity, accuracy and detection limits (Shephard et al., 2011). As part of this effort, GEOS-Chem model profiles of NH3 were used to generate a priori profiles and constraint matrices, which are necessary to retrieve NH3 profiles from the TES instrument. Additional model simulations were performed to determine the detection limit (∼1 ppbv) and bias (also ∼ +0.5 ppbv) of the retrieval. TES observations were then compared to measurements of NH3 from CAMNet surface monitoring sites, demonstrating the ability of TES to capture both spatial and temporal variability in NH3 concentrations (Pinder et al., in 2011). Global retrievals of NH3 were processed from TES for four months of the year over four years, identifying interesting trends and regions for future study through comparison to equivalent values from GEOS-Chem (Shephard et al., 2011).
The main goal of the second year of this project was to constrain NH3 emissions through assimilation of remote sensing observations from TES. Assimilation of TES NH3 retrievals into the GEOS-Chem model was performed for April, July and October over North America, using observations from the years 2006-2009 and model simulations for the year 2008 run at 2o × 2.5o resolution (Zhu et al., 2013). Overall, emissions appear to be underestimated in many regions, by up to a factor of three or more, although adjustments vary strongly by location and season. The NH3 simulations after emissions optimization were compared to surface observations from several sources: the AMoN network of NH3 monitoring, NTN wet deposition data, and IMPROVE aerosol measurements.
The main goal of the third year of this project was to evaluate the top-down constraints on emissions derived from assimilation of remote sensing NH3 data from Zhu et al. (2013) in terms of impacts on aerosol concentrations. Below we describe activities in this area in addition to work that continues to assess the validity of the top-down constraints through assimilation of additional measurements (e.g., NHx deposition) and improved process-based treatment of NH3 sources in air quality models. Possible explanations for the large adjustments are explored, such as the impacts of the diurnal variability of livestock sources making it appear that NH3 concentrations are underestimated, when in fact the timing of the emissions was contributing to a perceived underestimation by the satellite (Zhu et al., 2013). We continued to address the impacts of diurnal variability and bidirectional exchange on the top-down emissions constraints. An active soil NH3 pool has been implemented in GEOS-Chem to see if late summer emissions may in fact be a result of evaded fertilizer earlier in the year, rather than fresh livestock emissions.
Building on the source attribution study of Paulot et al. (2013), additional top-down constraints on NH3 sources are derived using the GEOS-Chem adjoint model and wet deposition measurements. These emission constraints were also explained with a new bottom-up NH3 inventory (MASAGE NH3). This approach finds that emissions in current inventories may be too high in parts of the eastern United States, but are likely underestimated in many midwestern regions during spring and fall. The overall results agree well with the satellite constraints in spring and fall, but there is still significant discrepancy in July. Changing our estimates of NH3 sources impacts model estimates of ammonium nitrate formation, and also the amount by which NH3 governs PM levels relative to other precursors such as SO2 and NOx. Despite high NH3 concentrations in some seasons, and constraints on these levels developed through satellite observations from TES (Shephard et al., 2011) and IASI (Kharol et al., 2013) or deposition (Paulot et al., 2014), PM2.5 concentrations in Asia appear to be most persistently influenced year round by NOx sources (Kharol et al., 2013). In the U.S., ongoing work focuses on the impacts of the diurnal variability of NH3 on PM2.5 via ammonium nitrate aerosol. We have also examined the impacts of changing NH3 and NOx emissions on reactive nitrogen deposition following RCP emissions scenarios (Paulot et al., 2013).
In the final year of this project, we completed our updates to mechanistic treatment of NH3 sources. From comparisons of weekly in situ measurements, hourly observations, and TES NH3 retrievals, we realized that a significant bias in modeled NH3 and, to a lesser extent, aerosol nitrate, may be owing to the assumed diurnal variability of NH3 emissions. In GEOS-Chem, these were previously treated as being constant throughout the day, and constant every day of the month, regardless of meteorological conditions. An improved treatment of the diurnal variability of livestock emissions (Bash et al., in prep) has been added to the GEOS-Chem model (Zhu et al., in prep.), and we implemented the bi-directional exchange of NH3 for the first time in a global model. The adjoint of bi-directional exchange has also been developed in the GEOS-Chem model and is used to investigate the sensitivity of NH3 concentrations with respect to soil pH and fertilizer application rate. This study thus lays the ground work for future inverse modeling studies to more directly constrain these physical processes rather than turning bulk uni-directional NH3 emissions.
Conclusions:
TES NH3 retrievals are found to have correlations in time and space with in situ NH3 observations (Pinder et al., 2011). The retrieval detection limit has been determined to be 1.0 ppb and the retrieval bias +0.5 ppb. TES NH3 retrievals are generally higher than model values in most regions of the world, suggesting model treatment of NH3 warrants revision. The overall finding is that NH3 emissions appear to be broadly underestimated, though significant spatial and seasonal variability in this feature is evident. Multiple top-down and bottom-up constraints are beginning to converge on a consistent set of emissions in the spring and fall, while significant discrepancies persist in the summer. Increasing NH3 emissions can do much to reduce the discrepancy between simulated and observed values, but adjustment to primary unidirectional emissions alone cannot reconcile all differences, particularly in July. Such discrepancies may in part be explained through improvements in process-based treatment of NH3 emissions, such as accounting for the diurnal variability of livestock emissions and the bidirectional exchange of NH3. The former can enhance NH3 concentrations aloft in over NH3 hotspots driven by livestock emissions, and the latter can enhance the lifetime (and hence concentrations) of NH3 in the summer without large increases to emissions. Constraints from deposition also provide another valuable means of assessing NH3 sources (Paulot et al., 2014). Projected growth in NH3 emissions will lead to an increased role of NH3 in governing reactive nitrogen deposition in future years (Paulot et al., 2013).
Although the GEOS-Chem simulations of NH3 atmospheric mixing ratio values are lower than the observed TES satellite observations, the global distribution pattern of the model NH3:CO ratio agrees well with the observations, indicating that the model represents the general location of the source regions and the seasonal enhancements of NH3 and CO globally over large regional scales. The outcomes of this project also lay the groundwork for the use of remote sensing data from satellite observations to constrain uncertainties in NH3. While this work was based primarily on the TES instrument aboard the NASA Aura satellite, which is nearing retirement, the work is being carried on through new analysis of NH3 retrievals from the CrIS instrument aboard the Suomi NPP satellite launched in 2011 (Shephard and Cady-Pereira, submitted). The broader impacts of this project are thus improvements in air quality model simulations of NH3, estimates of the transport and deposition of reactive nitrogen, and more reliable estimates of the atmospheric response to existing PM2.5 control measures.
Journal Articles on this Report : 10 Displayed | Download in RIS Format
Other project views: | All 49 publications | 10 publications in selected types | All 10 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Kharol SK, Martin RV, Philip S, Vogel S, Henze DK, Chen D, Wang Y, Zhang Q, Heald CL. Persistent sensitivity of Asian aerosol to emissions of nitrogen oxides. Geophysical Research Letters 2013;40(5):1021-1026. |
R834559 (2012) R834559 (Final) |
Exit Exit |
|
Luo M, Shephard MW, Cady-Pereira KE, Henze DK, Zhu L, Bash JO, Pinder RW, Capps SL, Walker JT, Jones MR. Satellite observations of tropospheric ammonia and carbon monoxide:global distributions, correlations and comparisons to model simulations. Atmospheric Environment 2015;106:262-277. |
R834559 (Final) |
Exit Exit Exit |
|
Paulot F, Jacob DJ, Henze DK. Sources and processes contributing to nitrogen deposition: an adjoint model analysis applied to biodiversity hotspots worldwide. Environmental Science & Technology 2013;47(7):3226-3233. |
R834559 (2011) R834559 (2012) R834559 (Final) |
Exit Exit Exit |
|
Paulot F, Jacob DJ, Pinder RW, Bash JO, Travis K, Henze DK. Ammonia emissions in the United States, European Union, and China derived by high-resolution inversion of ammonium wet deposition data: interpretation with a new agricultural emissions inventory (MASAGE_NH3). Journal of Geophysical Research–Atmospheres 2014;119(7):4343-4364. |
R834559 (2012) R834559 (Final) |
Exit Exit Exit |
|
Pinder RW, Walker JT, Bash JO, Cady-Pereira KE, Henze DK, Luo M, Osterman GB, Shephard MW. Quantifying spatial and seasonal variability in atmospheric ammonia with in situ and space-based observations. Geophysical Research Letters 2011;38(4):L04802 (5 pp.). |
R834559 (2010) R834559 (2011) R834559 (2012) R834559 (Final) |
Exit Exit Exit |
|
Shephard MW, Cady-Pereira KE, Luo M, Henze DK, Pinder RW, Walker JT, Rinsland CP, Bash JO, Zhu L, Payne VH, Clarisse L. TES ammonia retrieval strategy and global observations of the spatial and seasonal variabiity of ammonia. Atmospheric Chemistry and Physics 2011;11(20):10743-10763. |
R834559 (2010) R834559 (2011) R834559 (2012) R834559 (Final) |
Exit Exit |
|
Shephard MW, Cady-Pereira KE. Cross-track Infrared Sounder (CrIS) satellite observations of tropospheric ammonia. Atmospheric Measurement Techniques 2015;8(3):1323-1336. |
R834559 (Final) |
Exit Exit |
|
Wells KC, Millet DB, Cady-Pereira KE, Shephard MW, Henze DK, Bousserez N, Apel EC, de Gouw J, Warneke C, Singh HB. Quantifying global terrestrial methanol emissions using observations from the TES satellite sensor. Atmospheric Chemistry and Physics 2014;14(5):2555-2570. |
R834559 (2012) R834559 (Final) |
Exit Exit |
|
Zhu L, Henze DK, Cady-Pereira KE, Shephard MW, Luo M, Pinder RW, Bash JO, Jeong G-R. Constraining U.S. ammonia emissions using TES remote sensing observations and the GEOS-Chem adjoint model. Journal of Geophysical Research–Atmospheres 2013;118(8):3355-3368. |
R834559 (2011) R834559 (2012) R834559 (Final) |
Exit Exit |
|
Zhu L, Henze D, Bash J, Jeong G-R, Cady-Pereira K, Shephard M, Luo M, Paulot F, Capps S. Global evaluation of ammonia bi-directional exchange and livestock diunrnal variation schemes. Atmospheric Chemistry and Physics 2015;15(22):12823-12843. |
R834559 (Final) |
Exit Exit |
Supplemental Keywords:
ammonia, emissions, inverse modeling, adjoint sensitivity, environmental policy, air quality regulations, fine particulate matter, aerosols, remote sensing, data assimilation, 4D-Var, public health, eutrophicationRelevant Websites:
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.