DOC

Work Task 6

By Russell Green,2014-07-04 22:20
11 views 0
1 Work Task 6.5 How does deposition, owing to fog droplet growth, balance creation of secondary aerosol in fog droplets? What significance does the drop size-dependence of the fog chemistry have on aerosol formation and deposition in fogs? Are other parameters important? How does acidification of drops due to aqueous phase acid producti..

    Work Task 6.5

    How does deposition, owing to fog droplet growth, balance creation of secondary aerosol in fog droplets? What significance does the drop size-dependence of the fog chemistry have on aerosol formation and deposition in fogs? Are other parameters important? How does acidification of drops due to aqueous phase acid production limit aerosol formation in fog drops?

1. Task objectives

    The primary objectives of this work will be to evaluate how deposition of organic and inorganic material by fog droplets balances new aerosol formation in San Joaquin Valley (SJV) fogs and how the drop-size dependence of fog droplet composition affects new aerosol mass formation through S(IV) oxidation and deposition.

Specifically the objectives are to:

    ; Calculate sulfate formation rates in all collected samples for which sufficient

    information is available. Evaluate the predominant sulfur oxidation processes by the

    use of a single droplet model, developed at CSU and described by Reilly et al (2001).

    Use this model to examine if sulfate production, and hence aerosol mass formation, is

    limited by finite rates of mass transport into the fog drops.

    ; Examine distributions of individual chemical species as a function of drop size.

    Determine individual species deposition velocities in order to determine relative rates

    of removal for different solute species. Determine how strongly variations in

    composition across the drop size spectrum influence species deposition velocities.

    ; Compare measured fog solute deposition fluxes with calculated aqueous sulfate

    production rates to determine net effect of fog on airborne species’ concentrations.

    Unlike previous IMS95 efforts, this will include an analysis of the fog influence on

    removal of organic carbon.

    ; Use the Carnegie Mellon University (CMU) bulk fog chemistry model (Pandis 1989a,

    1989b) to predict concentrations of major species in fogwater and their temporal

    evolution for selected CRPAQS fog events. Compare these model predictions with

    the CSU time-resolved fog composition measurements. Evaluate how well the model

    simulates the observations. Use the model simulation to compare the relative

    importance of aerosol scavenging and removal vs. new aerosol mass production for

    simulated fog events.

    ; Use the size-resolved version of the CMU fog chemistry model to simulate at one key

    fog episode. Compare model predictions of drop size-dependent fog composition

    with CSU observations (2- and 5-stage fog collectors). Compare simulated

    deposition velocities for individual species with CSU observations. Use the model

     1

    simulation to evaluate the net influence of drop size-dependent composition on

    aerosol scavenging/removal vs. new aerosol mass creation.

    ; Evaluate scavenging of organic particulate matter by fog drops using molecular

    markers and elemental/organic carbon ratios. Conduct additional chemical analyses

    of archived CRPAQS fog samples to gain further information regarding organic

    carbon species scavenged by the fogs.

    ; Compare the results of the CRPAQS study with previous studies in the SJV (e.g. IMS

    95). Study the spatial variation of SJV fogwater composition using CRPAQS and

    IMS95 data. Evaluate if there has been any significant change over the years when

    fog studies have been conducted in the SJV.

    ; Present findings of study at one or more national/international meetings. Prepare

    findings for publication in a peer-reviewed journal.

2. Approach

Fogs have two important, competing effects on aerosol populations:

    ; New aerosol mass formation through gas scavenging and chemical reaction in the

     to sulfate) that droplets leading to non-volatile species (e.g. conversion of SO2

    remain in the particle phase after droplet evaporation

    ; Aerosol scavenging followed by deposition through droplet settling and/or

    impaction

    The relative importance of these two processes depends on the environment in which the fog forms: meteorological conditions, number and composition of aerosol particles, gas phase chemical composition, etc…. In addition, fogs are dynamic in nature; during the course of a fog event it is quite likely that droplet number concentrations and drop sizes change in addition to fog chemical composition. The net effect on atmospheric aerosol concentrations may change during a fog event; oxidation could be more important at the beginning of the fog event when reactant concentrations are higher, while deposition rates may increase over time with the growth of fog droplets.

2.1 Single drop simulations of sulfate production

    There have been numerous studies of fog chemistry and sulfur oxidation in fog droplets in the San Joaquin Valley (e.g. Hoag et al., 1999; Collett et al., 1999). Some of our recent work has highlighted that in central valley fogs sulfate production is drop-size dependent (Hoag et al., 1999), and can be strongly suppressed by the formation of S(IV)-aldehyde complexes, including hydroxymethanesulfonate (HMS) (Reilly et al., 2001). This latter study, which examined Davis, California, fogs, also considered the influence on reaction rates of limits to the rate of mass transport of reactants into fog drops. One key finding from this work was that sulfate production rates were lower than expected from a simple

     2

    examination of drop composition, due to rapid formation of HMS and an inability of mass transport to supply reactants to the fog drop interior as fast as they were consumed by reaction. This effect was particularly important in large fog drops where mass transport is slower. In fact, Reilly et al. (2001) showed, both by use of the single drop model and by application of a Se tracer technique for quantifying aqueous sulfate production, that sulfate production was slower in large fog drops than in small drops, despite the fact that the high pH of the larger drops suggested their composition was more favorable to rapid oxidation of dissolved sulfur dioxide.

    We propose to once again make use of the single drop model used by Reilly et al. (2001) to study mass transport limitations for all events and drop-sizes collected during the CRPAQS study. The main goal of these calculations is to improve our evaluations of aerosol mass formation in radiation fogs. The single drop model can readily be run for many time CRPAQS time periods and drop sizes, so it will provide a fuller overview of sulfate formation than is economical with the more complex CMU fog model simulations.

2.2 Aerosol removal by fog drop deposition

    Deposition due to fog drop sedimentation or impaction has been known to be an important removal process for atmospheric pollutants for a long time (Waldman, 1986). Some studies have tried to assess the deposition fluxes by fog in the SJV by modeling (Lillis et al., 1999) or by measurements (Collett et al., 2001). Nevertheless, relatively few measurements exist regarding atmospheric removal of fog solutes by drop deposition or how drop-size dependent fog composition affects removal rates for various chemical species. For radiation fogs formed in low wind speed environments, drop sedimentation strongly dominates fog drop deposition. This is especially true for larger fog drops that have higher terminal settling velocities.

    CSU measured fog deposition during CRPAQS by the use of surrogate surfaces: large Teflon plates designed to permit deposited fog retrieval for analysis of collected fogwater mass and composition. For the first time, the solute deposition measurements were extended to include organic carbon (measured as total organic carbon (TOC) by thermal oxidation to carbon dioxide). These flux measurements will be analyzed for all available measurement periods to examine the total rate of solute deposition due to fog drop sedimentation, for each solute measured (e.g., ammonium, nitrate, sulfate, organic carbon). These removal rates will be compared to calculated sulfate production rates (see above) to determine the net effect of the fog episodes on atmospheric burdens of individual chemical species.

    Deposition velocities for individual solutes can be determined based on the deposition flux measurements, along with “airborne” measurements of fog liquid water content (LWC) and composition. These will be examined to determine how variations in solute concentrations across the fog drop size spectrum influence fog removal rates for individual solute species. These deposition velocities can also be used to bound simulations made using the CMU bulk and size-resolved fog models.

     3

2.3. Fog modeling

    Our understanding of atmospheric phenomena is reflected by our ability to model these phenomena. The prediction of fog formation is very difficult due to the large number of parameters influencing its formation and only a few studies modeling fog chemistry exist (e.g., Pandis and Seinfeld 1989a, 1989b; Bott and Carmichael, 1992; Lillis et al., 1999). During IMS 95, we obtained good agreement between the CMU fog model and measurements by CSU (Lillis et al., 1999). We propose to apply this model to selected CRPAQS fog episodes and compare to the measurements made by CSU. For some events during the winter intensive, the time resolution of CSU measurements is very high and will allow a better comparison with the modeled fog evolution. In addition to simulating the fog occurrence and chemical composition, the model will also simulate solute deposition and sulfate production. Both the bulk and size-resolved versions of the CMU fog model will be run in order to examine the significance of the influence of drop size-dependent fog composition on aerosol processing by the fogs.

2.4 Fog interaction with carbonaceous aerosol

    While much is now known about the inorganic composition of SJV fogwater, little is known about the scavenging and removal of carbonaceous aerosol by these fogs. Despite measurement of high total organic carbon (TOC) concentrations (e.g., during IMS95), the composition of the organic species making up this TOC is largely unknown in the SJV and elsewhere. A few studies have focused on very specific compounds, like pesticides (Schomburg et al, 1991), organic nitrogen compounds (Anastasio et al., 2001) or methoxyphenols (Sagebiel and Seiber, 1993). Although the net effect of SJV fog episodes is expected to be to reduce atmospheric loadings of carbonaceous aerosol, the magnitude of this removal is unknown.

    We propose to use CSU data collected during CRPAQS to study the scavenging of carbonaceous aerosol by fog drops and the deposition of organic carbon to the surface via drop sedimentation. Molecular tracer concentrations (analyzed by GC/MS) will be followed from pre-fog aerosol to fogwater and interstitial aerosol during fog events to post-fog aerosol. Comparison of fog concentrations of these markers with concentrations in either pre-fog aerosol or interstitial aerosol will allow their scavenging efficiencies to be determined. If appropriate source markers (e.g., levoglucosan for wood smoke) are present at concentration levels sufficient for accurate quantitation, the relative scavenging efficiency of carbonaceous aerosol from different source types (e.g., wood smoke vs. vehicle exhaust) can be determined. Besides tracking individual organic compounds, EC/OC analysis will be performed on archived samples from the CRPAQS campaign to see how this ratio evolves during a fog-event. Preliminary analysis of one period suggests that the aerosol EC/OC ratio climbs during a fog event, consistent with preferential fog scavenging of carbonaceous aerosol particles low in EC content.

    The comparison of pre and post-fog aerosol concentration will give an indication of the net effect of fog on the aerosol and the analysis of fogwater concentrations and interstitial aerosol samples will give an indication of the partitioning of these organic compounds

     4

    between phases during a fog event. For total organic matter, deposition fluxes can be calculated based on the fog concentrations and compared to measurements made by CSU.

    We will also use CSU CRPAQS measurements of fog TOC concentrations and fog TOC deposition fluxes to examine the removal rates (and deposition velocities) of organic material during CRPAQS fog episodes (as discussed above). Measurements of the drop size-dependence of fog TOC concentrations will be used to interpret any differences in deposition velocities for TOC vs. other inorganic solutes.

    Because our understanding of the effect fogs have on atmospheric concentrations of organic species is severely limited by the lack of understanding of the organic composition of SJV fogwater, we also propose some additional chemical analyses of archived CRPAQS fog samples to try and increase the fraction of identified organic carbon (at present we know that formate, acetate, and formaldehyde typically comprise about 10-25% of SJV fog TOC while many additional compounds identified by GC/MS (e.g., n-alkanes, n-alkanoic acids, PAH, etc…) add a few percent more). We propose additional analyses of these archived CRPAQS fog samples by High Performance Liquid Chromatography (HPLC) to further speciate the fog TOC. In particular we plan to examine fog concentrations of carbonyl compounds and phenolic compounds. These compound families have been shown to be important constituents of fog TOC in some other regions.

3. Expected results

    Completion of the modeling and data analysis taks outlined above is expected to permit achievement of the previously stated project objectives. As a result of this work we will have an overview of the role CRPAQS fog episodes played in influencing airborne concentrations of key aerosol components and their precursors. We will also have an opportunity to further test the ability of the CMU fog model to successfully simulate SJV fog episodes. Although the model performed well in IMS95 simulations, further evaluation opportunities are needed to examine the utility of the model for future use. Analysis of this CRPAQS fog data set will also provide the first quantitative insight into the effect of fogs on carbonaceous aerosol concentrations, a subject which has not received significant prior study in the SJV or elsewhere.

    Results of the project will include a final report of activities and findings and are expected to result in at least two peer-reviewed journal publications: one concerning carbonaceous aerosol processing by the fogs and one concerning the overall effect of CRPAQS fog episodes on PM concentrations.

4. Detailed tasks

     5

    Colorado State University and Carnegie Mellon University will cooperate to achieve the

    goals previously stated.

    Colorado State University will perform the following tasks:

    ; calculate sulfate production rates for all CRPAQS samples using a single

    droplet model

    ; calculate deposition fluxes of major ionic species for all available

    measurement periods

    ; evaluate net effect of fog on aerosol mass

    ; calculate deposition fluxes of organic matter

    ; study scavenging of carbonaceous aerosol using selected organic tracer

    molecules

    ; reanalyze archived samples by HPLC to increase knowledge of molecular

    composition of organic matter

    ; compare CRPAQS fog composition observations with past SJV fog study

    results

    ; oversee CMU completion of the fog model simulations

    ; prepare progress and final reports

    ; attend Sacramento data analysis meetings

    ; prepare manuscripts for journal publication

    ; present project findings at San Diego national conference

    Carnegie Mellon University will perform the following tasks:

    ; model selected CRPAQS fog events using the CMU bulk and size-resolved

    fog models

    ; assist in preparation of project reports and publications

    Likely limitations to the analysis include:

    ; No information is available about the chemical form of iron present in

    most samples (only certain forms of iron are effective as catalysts for

    S(IV) oxidation. Past sensitivity studies have shown this is not a

    major issues as metal-catalyzed S(IV) autooxidation is generally too

    slow in this environment to be important.

    ; There may be significant amount of entrainment in fog as well as

    horizontal motion of the fog, making predictions of the concentrations

    and balancing production versus deposition difficult

    ; Samples may be altered during storage and hence the reanalysis by

    HPLC may yield an underestimation of the concentrations. Sample

     6

    aging will nevertheless be simulated during the study to quantify this

    effect.

5. Data Needs

    Many of the data needed for this project have been collected by CSU during the field study in winter 2000/2001 and, therefore, are readily available. These include fog composition data, fog liquid water content, fog deposition rates and composition of deposited fog water and gaseous peroxide data

Data needed from other investigators include:

     and PM mass and species concentrations at the Angiola site - PM102.5

    - Meterological data from the Angiola site

    - Vertical sounding data for model initiation

    - SO, NH, HNO, and O concentrations from the Angiola site 2333

    - MOUDI composition data from the Angiola site (can be supplemented if

    needed by analysis of MOUDI samples collected by CSU).

6. Time table

April 2002 Initiation of the contract

     Attend initial data analysis meeting

May-July 2002 Prepare draft workplan for review

    Collection and review of necessary data

     Selection of fog events for bulk and size-resolved modeling

     Submit additional pre-fog, interstitial, and post-fog filters for

     OC and EC analysis

     Analyze archived fog samples by HPLC

    Aug-Oct 2002 Use single drop model to analyze sulfate formation

     Analyze deposition data to obtain species removal fluxes and

     deposition velocities; compare with species drop size-dependence

     Conduct bulk fog simulations

     Examine carbonaceous aerosol scavenging by CRPAQS fogs

    Sept-Dec 2002 Compare model composition simulations with field observations

     Compare model deposition simulations with field observations

     Conduct size-resolved fog modeling

    Prepare journal manuscript regarding organic aerosol processing

    by fog

     7

    Attend second data analysis meeting

    Jan-March 2003 Compare size-resolved model simulations of composition and

    deposition with field observations

April-May 2003 Prepare and submit final report

    Prepare journal manuscript concerning net effect of CRPAQS fogs

    on PM

June 2003 Present findings at national conference

    Revise final report as needed in response to reviewer comments

     8

Report this document

For any questions or suggestions please email
cust-service@docsford.com