Rice STNI mprove

Document Sample
Rice STNI mprove Powered By Docstoc
					STN/IMPROVE Comparison Study
Preliminary Results

            Paul Solomon, ORD
       Tracy Klamser-Williams, ORIA
            Peter Egeghy, ORD
         Dennis Crumpler, OAQPS
           Joann Rice, OAQPS
                  Others
         PM Model Performance Workshop
               February 10, 2004
       Discussion Topics

Study Background
STN and IMPROVE Protocol
Differences
Preliminary STN versus IMPROVE
Data Comparison Results and
Conclusions
Next Steps



                                 2
    STN/IMPROVE Intercomparison

Why did we do this study?
   To establish data comparison and
    relationships for historical data
Collecting data since fall ’01
   3 areas (1 urban/rural pair)
Presenting Analysis of Oct ’01 to
Sept ’02 data
Finishing QA of first year data
Final results early ’04

                                        3
            STN/IMPROVE Intercomparison
  Operated According to Each Network’s Protocols
                            STN/IMPROVE Monitoring Intercomparison
                                   Sites:Oct. 2001 – Oct. 2002
Seattle
          Beacon Hill
           Mt. Rainier NPS
    URG MASS STN
      Samplers
                                                                      Haines
                                          USDA FS Dolly Sods          Point, NPS
                                                Wildersness
                                                                       Wash. DC
                    Tonto
                    National                             Anderson RAAS 401
                    Monument                                STN Samplers
    Phoenix

     Met One SASS STN
          Samplers



           Official or designated STN site, host to IMPROVE sampler
           Official IMPROVE site, host to STN sampler

                                                                                   4
           Comparative Protocol Analysis

   It is Not Just the Analysis Methods
   From Monitor Inlet* to Data Mgmt
       Sample Collection
       Handling, Shipping, and Storage (after collection)
       Chemical Analysis
         Extraction
         Analysis Methods
       Standards
         Or Lack Thereof for Ambient Field PM Measurements
       Data Manipulation
         Blank Subtraction and Artifacts

* Begins w/ Filter Purchase, Acceptance Testing, Handling & Storage

                                                                      5
         How Protocols Might Affect Results

    Between Networks *
        Inlets
           Effect on Slope of Efficiency Curve and
            Cutpoint
        Flow Rate Differences
           Effect of Pressure Drop, Face Velocity and
            Residence Time
                Influences Collection of Semi-Volatiles
                   Negative vs Positive Artifacts

                   Blank Values Likely Different




* Not an Exhaustive List
                                                           6
         How Protocols Might Affect Results
    Between Networks (cont) *
        Shipping and Storage
           STN at Reduced Temperatures
           IMPROVE at Ambient Temperatures
                 Influences Collection of Semi-Volatiles
        Use of IMPROVE in Urban Areas
           Higher Flow Rate, Smaller Filters
                 Filter Clogging Potential in urban areas
                 Effect on Semi-volatiles
           Denuder Capacity and Efficiency
                 Na2CO3 vs MgO
                 Refurbishing Frequency of Na2CO3
* Not an Exhaustive List
                                                             7
            Bottom Line

Given All these Differences, Do the
Networks Provide Similar Results for
     Mass and the Components




               ?
                                       8
     Median Ratio of STN/IMPROVE
                    Urban (E to W)                  Rural (E to W)
                                 Beacon
           Haines Pt. Phoenix               DOSO         TONT        MORA
                                   Hill

 IMP/STN                        Median Ratio (IMP/STN)

PM2.5        0.91        1.05        0.96    0.89        0.91        0.90
EC           1.09        1.01        1.04    1.29        0.68        1.39
OC           0.54        0.48        0.90    0.53        0.36        0.58
Nitrate      0.89        0.87        1.31    0.87        1.22        0.86
Sulfate      1.00        1.31        1.06    1.01        0.90        0.86
As           0.62        0.49        0.67    0.46        0.46        0.63
Ca           0.87        1.61        1.37    0.95        1.46        1.16
Cr           0.55        0.27        0.75    2.22        0.11        0.28
Cu           0.93        1.28        0.63    0.78        1.44        0.16
Fe           0.93        1.59        1.34    0.91        1.35        1.11
Mn           0.70        1.11        1.19    0.84        0.90        0.85
Pb           0.83        0.84        1.01    0.88        0.69        0.72
S            0.98        0.97        1.03    1.03        0.95        0.84
Si           1.42        1.82        1.57    1.48        1.45        1.43
Zn           1.03        1.55        1.18    1.01        1.20        1.00

                                                                            9
 Urban – Rural Temporal Analysis: PM2.5 Mass


Haines Point              Dolly Sods




               Phoenix     Tonto




                                               10
 Urban – Rural Comparison of Means: PM2.5 Mass
                                                                        * Relative Percent Diff.
                                                                     (STN-IMP)/((STN+IMP)/2)*100
   9.0%* Urban (EW) Rural (EW)

         -5.1%                                 3.3%
                        11.0%                                                STN-IMP (ug/m3)
                                                              5.9%
                 4.3%                                                                    1.3
                                                      10.8%               Haines Point
                                11.3%
                                                                          Phoenix        -0.6
                                        9.5%
                                                                                         0.3
                                                                          Beacon Hill
                                                                          Dolly
                                                                                         1.1
                                                                          Sods

                                                                          Tonto          0.7

  Annual Average Results                                                  Mt Rainier     0.3
 East Coast Sites Have Higher Conc. than West Coast                      Urban          0.36
    Sites                                                                 Rural          0.71
 Urban Site Levels Exceed Rural Sites by 20-100%
                                                                          All Sites      0.52
 Similar Agreement, within ~10%, is Observed at Both
    Urban and Rural Sites, with Slightly Better
    Average Agreement at Urban Sites
                                                                                                11
         Urban – Rural Temporal Analysis: Sulfate



Haines Point                  Dolly Sods




               Beacon Hill
                              Mt. Rainier




                                                    12
       Urban – Rural Comparison of Means: Sulfate
                                                                    * Relative Percent Diff.
                                                                 (STN-IMP)/((STN+IMP)/2)*100
            Urban (EW)    Rural (EW)
    -0.9%

                          -24.2%
                                                                          STN-IMP (ug/m3)
                                                                       Haines Pt.    -0.05
                                                -6.0%       -9.0%
                                                                       Phoenix       -0.36
            -23.4%                                      -13.7%
                                                                       Beacon Hill   -0.09
                  -7.5%         12.0%
                                         9.5%                          Dolly Sods    -0.89
                                                                       Tonto         0.12
                                                                       Mt. Rainier   0.06
     Annual Average Results                                            Urban         -0.15
 East Coast Sites Have Higher Conc. than West Coast Sites
                                                                       Rural         -0.25
 Urban Site Levels Exceed Rural Sites By 50 to 70% (Avg 48%)
    Based on STN Data and By 22-100% (Avg 37%) Based on                All Sites     -0.20
    IMP Data
 There Is Better Agreement On Average at Urban Sites
 The Dolly Sods STN Sulfate Data For June & July 2001 Will Be
    Flagged as Invalid Due to a Flow Controller Problem – These
    Data Contribute Significantly to the Larger Difference
    Between the Two Networks at Dolly Sods                                                   13
Urban – Rural Temporal Analysis: Nitrate


         Haines Point                 Dolly Sods




             Beacon                     Mt.
               Hill                    Rainier




                                                   14
         Urban – Rural Comparison of Means: Nitrate
                                                                          *Relative Percent Diff.
                                                                      (STN-IMP)/((STN+IMP)/2)*100
       11.7% Urban (EW) Rural (EW)

                                                                               STN-IMP (ug/m3)
             13.1%                             0.5%
                     -31.9%                                                Haines Pt.     0.15

                                                              -0.2%        Phoenix        0.12
                         13.7%
                                 -20.8%                                    Beacon Hill    -0.24
                                                      -2.8%
                                       -3.3%                               Dolly Sods     0.060

                                                                           Tonto         -0.085

                                                                           Mt. Rainier   -0.005

                                                                           Urban          0.005

    Annual Average Results                                                 Rural         -0.009
   On Average, Nitrate Conc Are Low, < 1.4 µg/m3
                                                                 All Sites               -0.002
   East Coast Sites Have Higher Conc. than West Coast Sites
   Paired Urban Site Levels Exceed Rural Sites by ~ Factor of 3
   By Site, Samplers Agree Within About 30% on Average
   At Beacon Hill the Difference is Largest at Higher Conc as
      Seen in the Temporal Distribution
                                                                                                  15
Urban – Rural Temporal Analysis: OC

   Haines Point
                                            Dolly Sods




                  In General IMP < STN


Beacon Hill                              Mt. Rainier




                                                         16
        Urban – Rural Temporal Analysis: OC

                                                                    STN Values
                                                                       Blank
                 Haines Point                                        Corrected




                                                                    IMP/STN OC   IMP/STN-Blk OC

                                                      Haines Pt.       0.59           0.83

                                                      Phoenix          0.56           0.77

        After STN Blank Correction                    Beacon Hill      0.95           1.06

               IMP ~ STN*                             Dolly Sods       0.57           1.07

       Mostly On Average Within 15%                   Tonto            0.41           0.87

                                                      Mt. Rainier      0.67           0.84

*Note Negative Values in STN After Blank Correction
                                                      Urban            0.66           0.88

                                                      Rural            0.53           0.94
                                                      All Sites        0.62           0.89
                                                                                                  17
        Urban – Rural Comparison of Means: OC
                                                           STN Blank Values Are Based on
      Urban (EW)   Rural (EW)
                                                           Trip and Field Blanks and
                                                           Averaged Over the Time Period
                                                           of the Study

                                                           Anderson = 1.3 ug/m3
                                                                Haines Pt, Dolly Sods
                                                           MetOne = 1.4 ug/m3
                                                                Phoenix, Tonto
                                                           URG = 0.3 ug/m3
                                                                Beacon Hill, Mt Rainier




                                  Annual Average Results
Blank Correcting STN        East and West Coast Sites Can Have Similar
 OC Values Improves            Concentrations
   the Comparison           STN Urban Site Levels Exceed Rural Sites by 200-
                               500%
  Between STN and
                            Rural Sites Tend to Agree Better Than Urban Sites Before
  IMPROVE at Most              Blank Correction
      Locations             Values Agree Better, but Blank Correcting STN Results in
                               Negative Numbers That Seem to Be More Prevalent in the
                               Fall & Winter
                                                                                           18
   Urban – Rural Temporal Analysis: EC

Haines Point                        Dolly Sods




       Beacon Hill     Mt Rainier




                                                 19
             Urban – Rural Comparison of Means: EC
                                                                      *Relative Percent Diff.
                                                                  (STN-IMP)/((STN+IMP)/2)*100
            Urban (EW)   Rural (EW)
   -9.6%*                                      -3.7%

                  0.4%

         -1.4%                                                -6.2%

                          -28%                                                      IMP/STN EC
                                 23%                   -15%
                                        -41%                          Haines Pt.       1.10

                                                                      Phoenix          1.01

                                                                      Beacon Hill      1.00

                                                                      Dolly Sods       1.32

                                                                      Tonto            0.79

      Annual Average Results                                          Mt. Rainier      1.52

 EC Data Did Not Require Blank Correction                            Urban            1.04
 Urban Conc Are ~ 2X Rural Conc                                      Rural            1.16
 Factor of 2 Not Observed Between STN and IMPROVE At                 All Sites        1.06
    Urban or Rural Sites
 Better Agreement at Urban Sites (<10%) than at Rural
    Sites (± ~ 30%)


                                                                                                 20
                     Conclusions

Concentrations at Rural Sites Lower for Most
Species at Most Sites
Less Consistency (Greater Scatter) Observed at
Rural Sites
Higher Data Capture Observed at Urban Sites
Mass and Sulfate Agreed Well
   Typically within 10% for Mass and 20% for Sulfate

Organic Carbon Agreed Better After STN Data
Blank Corrected


                                                        21
          Conclusions (cont’d)

Species at Higher Concentrations Agree
Better
   MDL & Blanks Potential Issue Between Network
    Agreement

Site-to-Site Variations Observed for All
Species Even After a Thorough Review of the
Data and Outliers
   Questionable Data Still Remain

Factor of 2 Difference not Observed Between
STN and IMPROVE EC


                                                   22
           Next Steps

Finalize report on first year of data
from 6 sites
Conduct study to determine the
effects of shipping conditions
Begin analyzing quartz filters from
STN sites using IMPROVE protocol
Sponsoring projects to specifically
look at the effect of carbon meas.
protocols on concentration data

                                        23
            Next Steps

Expanding IMPROVE collocation
to nine additional “STN” sites in
2003/2004
   New York City
   Atlanta
   Pittsburgh
   Birmingham
   Detroit
   Chicago
   Houston (SS)
   Riverside-Rubidoux
   Fresno (SS)
                                    24
         Disclaimer & Acknowledgements
This work has been funded wholly by the United States Environmental
   Protection Agency. It has been subjected to Agency review and
   approved for publication. Mention of trade names or commercial
   products do not constitute endorsement or recommendation for
   use.

Acknowledgements:
This work was supported by the National Park Service and EPA. Staff
   from UC Davis, DRI and NPS personnel have played a significant
   role in collection and analysis of samples in the IMPROVE Network.
   Research Triangle Institute has played a significant role in
   preparing and analyzing samples for the STN network, as well as
   state site operators who have meticulously been collecting samples
   since October 2001, which continues to date.

Complete Authors List:
Solomon, Paul, A. (EPA, ORD, Las Vegas), Crumpler, Dennis (EPA,
   OAQPS, RTP), Klamser-Williams, Tracy (EPA, ORIA, Las Vegas),
   Egeghy, Peter, Homolya, James, Pitchford, Marc (EPA, OAQPS,
   Las Vegas), Rice, Joann, Ashbaugh, Lowell (UC Davis, Sacramento),
   Orourke, James (RTI, RTP), Frank, Neil, McDade, Charles,
   Flanagan, James, and Rickman, Edward
                                                                        25

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:6
posted:6/23/2012
language:
pages:25