Grantee Research Project Results
Final Report: Atmospheric Aerosols from Biogenic Hydrocarbon Oxidation
EPA Grant Number: R831079Title: Atmospheric Aerosols from Biogenic Hydrocarbon Oxidation
Investigators: Milford, Jana B. , Guenther, Alex , Wiedinmyer, Christine , Helmig, Detlev
Institution: University of Colorado at Boulder , National Center for Atmospheric Research
EPA Project Officer: Chung, Serena
Project Period: October 1, 2003 through September 30, 2006 (Extended to September 30, 2007)
Project Amount: $440,000
RFA: Measurement, Modeling, and Analysis Methods for Airborne Carbonaceous Fine Particulate Matter (PM2.5) (2003) RFA Text | Recipients Lists
Research Category: Air , Air Quality and Air Toxics , Particulate Matter
Objective:
The objective of this research was to improve estimates of the contributions of monoterpenes (MT) and sesquiterpenes (SQT) to secondary organic aerosol (SOA) production in the U.S. The project included measurements of SQT and MT emissions; incorporation of new biogenic emissions factors into the Model of Emissions of Gases and Aerosol from Nature (MEGAN); development of BVOC inventories for input to the Community Multiscale Air Quality Model (CMAQ); and regional-scale air quality modeling with CMAQ.
Starting with the emissions measurement portion of this project, Duhl et al. (2007) compiled a literature review on environmental controls for SQT emissions, which summarizes most of the SQT research done to date, both in terms of methods used and results obtained. The review informed the experimental methods development accomplished in the project, and provided a basis for comparing new emissions factor measurements with those reported in previous studies.
A major focus of our work was to improve the data base of SQT emissions measurements. Enclosure experiments were conducted at a number of sites in four different locations (Boulder, CO; Arcata, CA; Duke Forest, NC; Pellston, MI). Besides these outdoor experiments on naturally growing vegetation, experiments were also conducted in the National Center for Atmospheric Research (NCAR) greenhouse. Altogether, more than 50 enclosure experiments on more than 40 different plant species were performed. Each of these experiments typically lasted 2-4 days, with in most cases 10-30 samples collected and quantified. All emissions data from a given enclosure experiment were investigated for their dependency on temperature and light. As part of this experimental work, we developed improved techniques for measuring SQT emissions in branch enclosures and an in-situ GC-MS/FID method tailored for SQT measurement (Ortega et al., 2007a).
In collaboration with colleagues at Washington State University, procedures were developed to calculate biogenic emissions using a FORTRAN90 version of the MEGAN framework to produce emission files compatible with the SMOKE and CMAQ models. The emissions are driven by temperature and solar radiation values from the same MCIP-processed MM5 output used to drive CMAQ. Additional driving variables include plant functional type distributions, leaf area index, and emission factors, which are available in high resolution (~ 1 km) global files. Emissions are estimated for 138 individual chemical species, including many monoterpenes and sesquiterpenes. As part of this study, new lumping schemes were developed to assign each of these species to the categories used for SAPRC99, RADM2, RACM, and CBMZ (Sakulyanontvittaya, 2008). MT and SQT emission factors and emission algorithms (e.g., 2 response to light and temperature) were developed using new emissions data and were compared to estimates based on BEIS3.0 for MT, and to literature values and a simple parameterization of BEIS3.0 emissions for SQT (Sakulyanontvittaya et al., 2007).
The CMAQ model, version 4.5, was modified to estimate secondary organic aerosol formation from SQT, based on the existing gas-particle partitioning module for condensable products of MT and anthropogenic VOC oxidation. The treatment of isoprene in the model was not changed; thus formation of SOA from isoprene oxidation products is not included (although isoprene does participate in gas-phase chemistry). For this project, the gas-particle partitioning module was also modified to allow the option of treating a portion of SOA as polymerized aerosol that is not susceptible to evaporation. Additionally, data on the enthalpy of vaporization of condensable products of VOC oxidation were compiled from the literature and the results used to develop new estimates of this parameter for products of SQT, MT, and other SOA-forming VOCs.
Using newly developed inventories for BVOC, CMAQ runs were performed for July 2001 and January 2002, with and without SQT emissions. The sensitivity of modeled SOA concentrations to the assumed enthalpy of vaporization values, polymerization in the aerosol-phase, and the level of anthropogenic NOx emissions were examined in additional simulations. CMAQ results were compared to observations from the IMPROVE, SEARCH, AIRS and STN networks.
Summary/Accomplishments (Outputs/Outcomes):
Our review of published research on SQT emissions revealed a high variability of methods used by different researchers. Given that SQT have rather low vapor pressures, they tend to easily adhere (‘stick’) to surfaces used in experimental systems. Also, due to their high chemical reactivity, SQT can easily be lost during air sampling unless atmospheric reactants are removed from the sampling stream. Another challenge is the calibration of SQT measurements. As SQT gas standards are not stable in cylinders, standards must either be generated in-situ or surrogate compounds must be used to determine instrument response factors. With these challenges, it is important to follow well tested methods that are tailored towards the specific requirements of SQT analysis. Unfortunately, the literature on previous SQT research did not emphasize these requirements, and many of the published data lack information needed to gauge how quantitative the measurements are. These questions motivated us to continue our thorough investigations on SQT emission experiments and to develop guidance for other colleagues working in this field. Ortega and Helmig (2007) review experimental conditions for SQT enclosure experiments and provide a list of recommendations for SQT research. In a companion paper, Ortega et al. (2007a) present a field sample collection and in-situ GC-MS/FID analysis that was developed particularly for SQT emission studies.
The experimental portion of this project yielded the largest body of data on SQT emissions available to date, and additionally provided many new data on MT emissions. Compound identifications, calculated emission rates, and parameterization of the temperature dependency from these studies have been included in four research manuscripts that have resulted from this research (Helmig et al., 2006, 2007; Ortega et al., 2007a,b).
The new emissions factor data were used in the MEGAN biogenic emissions model to estimate SQT emissions for the contiguous U.S. domain. Resulting monthly average emissions rates of SQTs estimated for the domain were about 300 tons hr-1 for July and 10 tons hr-1 for January. The SQT emission factors developed for this project result in average U.S. SQT emissions that are more than a factor of two lower than those based on previously reported values. The MT emission factors developed for this project are about 30% lower than those used in the BEIS3.0 model (Sakulyanontvittaya et al., 2007). In the current version of MEGAN, a significant fraction of SQT emissions over the U.S. is associated with the grass-crop plant function type, for which only a limited number of measurements are available.
A study of the SQT emission response to temperature demonstrated that the model is very sensitive to this parameter. The U.S. average emission varies by about 25% in July when the temperature-dependence coefficient is varied from 0.07 K-1 to 0.25 K-1, because emissions increase in some regions and decrease in others; however, the maximum emission varies by a factor of four. Average emissions in January change by a factor of three since emissions throughout most of the U.S. increase with decreasing temperature sensitivity (Sakulyanontvittaya et al., 2007).
When the modified version of CMAQ was used with SQT emissions estimates from MEGAN, with updated enthalpy of vaporization values, and without polymerization, the results indicated that over much of the eastern and northwestern U.S., SQT contribute monthly average surface level SOA concentrations ranging from 0.1 to 0.6 μg m-3 in July and less than 0.1 μg m-3 in January. The amount of SOA produced from SQT is comparable to the amount produced from MT emissions, even though SQT emissions are lower than MT emissions by a factor of six in July and a factor of ten in January.
Our sensitivity tests show that use of the CMAQ v4.5 default value of ΔHvap = 156 kJ mol-1 results in SOA concentrations that are three to six times higher than concentrations produced with the values we estimated from the recent literature. The difference is mostly due to SOA from monoterpenes, which contribute significantly more SOA when the default value of ΔHvap = 156 kJ mol-1 than when our updated value of 83.1 kJ mol-1 is used for their condensable products. Polymerization of SOA increases the SOA mass in the domain by about a factor of two. The temporal and spatial distribution of the polymerized SOA differs significantly from that of non-polymerized SOA, with the former displaying relatively constant and uniform concentrations (Sakulyanontvittaya, 2008).
The results from the modified CMAQ simulations show the model significantly underpredicts the organic matter fraction of PM2.5 in all cases studied. The highest SOA concentrations were produced in the case with SQT emissions and the default enthalpy of vaporization value, and in the case with SQT emissions and polymerization. Even in these cases the mean normalized bias in surface concentrations of organic matter was about 50% in July and about 40% in January.
Conclusions:
This study represents a first attempt to explicitly include SQT in regional air quality modeling efforts, beginning with development of an emissions inventory from SQT emissions measurements. The results demonstrate the important contributions of SQT emissions to SOA production in the U.S. and indicate that additional efforts are warranted to reduce the large uncertainties associated with these emissions. Our analysis of SQT emission rate measurements revealed large variations in observations for individual plant species and genera. As a result, we were unable to assign emission factors to individual plant species and could only justify estimating emission factors for broad plant categories, or plant functional types. It is likely that there are genetic controls over sesquiterpene emissions but that these are obscured by uncertainties associated with measurement techniques and by variability associated with plant stress conditions. Improved emission factor estimates will require improved measurement techniques and better characterization of plant stress conditions.
The SQT temperature response sensitivity study demonstrated the importance of accurately describing this relationship. Some of the temperature response variability observed in field measurement data may be associated with the light response, since light and temperature tend to be highly correlated in field observations. A better understanding of the light and temperature controls over sesquiterpene emissions is needed and will require measurements under controlled light and temperature conditions.
New emissions estimates for SQT were used in CMAQ to estimate the contribution of SQT to SOA formation. With the updated BVOC emissions inventories and enthalpy of vaporization values for MT and SQT oxidation products estimated from recent studies, the amount of SOA produced from SQT was similar to that produced from MT. SOA formation in the gas-particle partitioning module is very sensitivity to the value used for ΔHvap. Although the recent literature suggests the default enthalpy of vaporization value used in CMAQ is too high, for SOA precursors other than SQT, reducing the values to better match the literature degrades the model’s performance in simulating fine particle organic matter concentrations.
Comparisons to observations show that inclusion of SQT emissions improves CMAQ predictions for SOA, but the model still underpredicts fine particle organic matter concentrations. This underestimation suggests the emissions inventories and SOA process descriptions still need to be improved. SOA formation from isoprene should be considered as one potentially important source that is currently missing from CMAQ. Its inclusion could enhance formation of SOA from other precursors, including anthropogenic VOC, monoterpenes and sesquiterpenes.
Journal Articles on this Report : 10 Displayed | Download in RIS Format
Other project views: | All 21 publications | 10 publications in selected types | All 10 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Duhl TR, Helmig D, Guenther A. Sesquiterpene emissions from vegetation:a review. Biogeosciences 2008;5(3):761-777. |
R831079 (Final) |
Exit Exit |
|
Duhl TR, Guenther A, Helmig D. Estimating urban vegetation cover fraction using Google Earth® images. Journal of Land Use Science 2012;7(3):311-329. |
R831079 (Final) |
Exit Exit |
|
Helmig D, Ortega J, Guenther A, Herrick JD, Geron C. Sesquiterpene emissions from loblolly pine and their potential contribution to biogenic aerosol formation in the Southeastern US. Atmospheric Environment 2006;40(22):4150-4157. |
R831079 (2006) R831079 (Final) |
Exit Exit Exit |
|
Helmig D, Ortega J, Duhl T, Tanner D, Guenther A, Harley P, Wiedinmyer C, Milford J, Sakulyanontvittaya T. Sesquiterpene emissions from pine trees—identifications, emission rates and flux estimates for the contiguous United States. Environmental Science & Technology 2007;41(5):1545-1553. |
R831079 (2006) R831079 (Final) |
Exit Exit Exit |
|
Ortega J, Helmig D, Guenther A, Harley P, Pressley S, Vogel C. Flux estimates and OH reaction potential of reactive biogenic volatile organic compounds (BVOCs) from a mixed northern hardwood forest. Atmospheric Environment 2007;41(26):5479-5495. |
R831079 (Final) |
Exit Exit Exit |
|
Ortega J, Helmig D. Approaches for quantifying reactive and low-volatility biogenic organic compound emissions by vegetation enclosure techniques--Part A. Chemosphere 2008;72(3):343-364. |
R831079 (Final) |
Exit Exit Exit |
|
Ortega J, Helmig D, Daly RW, Tanner DM, Guenther AB, Herrick JD. Approaches for quantifying reactive and low-volatility biogenic organic compound emissions by vegetation enclosure techniques-Part B: Applications. Chemosphere 2008;72(3):365-380. |
R831079 (Final) |
Exit Exit Exit |
|
Pollmann J, Ortega J, Helmig D. Analysis of atmospheric sesquiterpenes: sampling losses and mitigation of ozone interferences. Environmental Science & Technology 2005;39(24):9620-9629. |
R831079 (Final) |
Exit Exit Exit |
|
Sakulyanontvittaya T, Guenther A, Helmig D, Milford J, Wiedinmyer C. Secondary organic aerosol from sesquiterpene and monoterpene emissions in the United States. Environmental Science & Technology 2008;42(23):8784-8790. |
R831079 (Final) |
Exit Exit Exit |
|
Sakulyanontvittaya T, Duhl T, Wiedinmyer C, Helmig D, Matsunaga S, Potosnak M, Milford J, Guenther A. Monoterpene and sesquiterpene emission estimates for the United States. Environmental Science & Technology 2008;42(5):1623-1629. |
R831079 (Final) |
Exit Exit Exit |
Supplemental Keywords:
Biogenic hydrocarbons, secondary organic aerosols, sesquiterpenes,, RFA, Scientific Discipline, Air, Ecosystem Protection/Environmental Exposure & Risk, particulate matter, air toxics, Environmental Chemistry, Air Pollution Effects, Monitoring/Modeling, Environmental Monitoring, Environmental Engineering, atmospheric particulate matter, health effects, aerosol particles, mass spectrometry, air quality models, air sampling, carbon particles, air quality model, emissions, thermal desorption, particulate matter mass, Volatile Organic Compounds (VOCs), biogenic hydrocarbon oxidation, aerosol analyzersRelevant Websites:
Information on the MEGAN model is available at
http://acd.ucar.edu/~guenther/MEGAN/MEGAN.htm Exit MEGAN can be downloaded for use from the NCAR Community Data Portal at http://cdp.ucar.edu/home/home.htm 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.