&EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
EPA-454/R-01-005
NTISPB#2001-105814
Date: May 2001
Air
Evaluation of PM7 * Chemical
j-L • t/
Speciation Samplers for Use in
the EPA National PM2 5
^—' • *./
Chemical Speciation Network
-------
Evaluation of PM2.5 Chemical Speciation Samplers for Use in the EPA
National PM2.5 Chemical Speciation Network
Volume I - Introduction, Results, and Conclusions
Final Report
15 July 2000
Prepared By
Paul A. Solomon
William Mitchell
Michael Tolocka
Gary Norris
David Gemmill
Russell Wiener
US EPA
Office of Research and Development
National Exposure Research Laboratory
Research Triangle Park, NC 27711
Robert Vanderpool
Robert Murdoch
Sanjay Natarajan
Eva Hardison
Research Triangle Institute
Research Triangle Park, NC 27711
Prepared for
Richard Scheffe
James Homolya
Joann Rice
Office of Air Quality Planning and Standards
Research Triangle Park, NC
-------
DISCLAIMER
This work has been funded wholly or in part by the United States Environmental Protection Agency.
Portions of the work were performed under Contract No. 68-D5-0040 by Research Triangle Institute.
It has been subjected to Agency review and approved for publication. Mention of trade names or
commercial products does not constitute an endorsement or recommendation for use.
Part I, Page i
-------
ACKNOWLEDGMENTS
The authors would like to thank the many people who assisted in the design, implementation, data
analysis, and preparation of this final report. In particular, the authors would like to thank, Mel Zeldin
(SCAQMD) and Tom Moore (Arizona DEQ) for providing space, operational support, logistics, and
power at the Rubidoux and Phoenix air monitoring sites, respectively. We would also like to thank the
site operators at Rubidoux and Phoenix for their long hours and their dedication to the project. We are
also grateful to the RTI staff who operated the Philadelphia and RTF sites and performed the field
audits and those that spent many long hours changing filters and performing chemical analysis. We are
also appreciative of Judy Chow and her staff (DRI), Bob Gary and his staff (Sunset Labs), and Bob
Kellogg (Mantech) for analyzing filters with a very quick turn around time. The project would not have
been successful without the assistance of the manufacturer's representatives, Tom Merrifield (MetOne),
Wes Davis (Andersen), and Jon Stone (URG) and their willingness to help train and set up the
samplers, and their prompt response to problems encountered during sampling. Thanks is also given to
Lowell Ashbaugh of UC Davis who supplied the IMPROVE samplers for this study. We are indebted
to Jack Suggs (EPA, ORD) for his assistance with the statistical analysis effort described in this report.
The PM Expert Panel reviewed the program plan and provided valuable assistance in their first review
of the Speciation Guidance Document with the initial recommendations for having this evaluation study.
Finally, we thank our clients at the Office of Air Quality Planning and Standards, and in particular,
James Homolya, Joann Rice, Shelly Eberly, and Richard Scheffe. Their support and assistance
throughout the study was indispensable, as well as support from Russell Wiener, Branch Chief,
AMMB. This project was supported by funds from the OAQPS speciation program and from
ORD/NERL/HEASD/AMMB PM Methods Team.
Part I, Page ii
-------
EXECUTIVE SUMMARY
To develop improved source-receptor relationships and for better understanding the causes of high
PM2 5 concentrations in the atmosphere, it is necessary to not only determine concentrations of PM2 5
mass, the NAAQS indicator, but also the chemical components of PM2 5. A sampling program of this
type, which will consist of up to 300 sites nationwide has been initiated by EPA (Speciation Guidance
Document, 1999 at http://www.epa.gov/ttn/amtic/pmspec.htmll Since the PM2.5 Federal Reference
Method (FRM) using only Teflon filters is not suitable for determining the chemical composition of the
collected aerosol, since carbon can not be directly measured (Speciation Guidance Document, 1999),
EPA solicited innovative designs for speciation samplers, based on performance specifications. This
led to the development of three slightly different candidate samplers manufactured by Andersen
Instrument Inc., MetOne, Inc., and University Research Glassware (URG). These samplers are
designed to allow for a nearly complete mass balance of the collected aerosol, while minimizing
sampling artifacts for nitrate and allowing flexibility for minimizing organic carbon artifacts in the future.
Due to the need to have consistency across this national network, the Speciation Expert Panel
(Recommendations of the 1998 Expert Panel, 1998 at http://www.epa.gov/ttn/amtic/ pmspec. html)
recommended a methods comparison field study among the new speciation samplers, historically used
samplers, and the PM2 5 FRM. The program plan for EPA's Chemical Speciation Sampler Evaluation
Study (1999, http://www.epa. gov/ttn/amtic/casacinfhtmD details the approach and implementation of
the study. This report presents the approach and results from the 4-City intercomparison study; Phase
1, of the full evaluation of these samplers. Other Phases are described in Field Program Plan (1999)
and include evaluation of denuders and reactive post filters for sampling organic aerosols with minimal
artifacts (Phase n, Seattle, WA, J. Lewtas, PI), an evaluation of the chemical speciation samplers under
summertime conditions (Phase n, Atlanta, GA in conjunction with the Atlanta Supersites Program, P.
Solomon, PI), and an evaluation of the samplers under a variety of environmental conditions to test
operational performance and logistics with the National Chemical Speciation Laboratory (Phase IV, 15
Cities throughout the US (Mini-trends network, J. Homolya, PI).
Methods. Because of potential sampling artifacts when using filters and potential differences in inlet
cutpoints and sample fractionators, the chemical speciation samplers must be able to properly
determine the chemical components of PM2 5 under a variety of atmospheric and environmental
conditions. Four locations, with different atmospheric chemical and meteorological conditions were
chosen and included: Rubidoux, CA (high nitrate and carbon and low sulfate), Phoenix, AZ (high
crustal material and moderate carbon and nitrate), Philadelphia, PA (high sulfate, moderate carbon, and
low nitrate), and Research Triangle Park (RTF), NC (low PM2 5 concentrations). The latter site also
allowed for a more thorough evaluation of the samplers' in-field operational performance as it was
located near EPA offices in RTF. In addition to the three candidate samplers, a Versatile Air Pollution
Sampler (VAPS), an IMPROVE sampler, and an FRM were collocated at each site. Replicate
samplers were located at Rubidoux. Samples were collected for up to 20 days during January and
February, 1999 using state personnel (Rubidoux and Phoenix) or EPA contractors (Philadelphia and
RTF). All sampling periods were 24-hrs in duration. Mass and trace elements were determined on
Part I, Page iii
-------
Teflon filters; sulfate, nitrate, and ammonium were determined on either Teflon, pre-fired quartz-fiber,
or nylon filters depending on the sampler; and OC/EC were determined on pre-fired quartz-fiber filters.
To minimize variability, all filter preparation, filter changing, and chemical analyses for a particular
species were performed by one contractor. Quality assurance/quality control followed EPA guidelines
(QAPP for the Four-City PM2 5 Chemical Speciation Sampler Evaluation Study, January, 1999
Research Triangle Institute, Project Number 07263-030).
Results. All samplers encountered operational problems that increased variability in the results;
however, the Andersen and MetOne samplers collected over 90% of the attempted samples on a site-
by-site basis successfully, while the URG and Versatile Air Pollution Sampler (VAPS) collected greater
then 75% of the samples attempted on a site-by-site basis. Most manufacturers have resolved
operational issues. Other minor engineering changes were made to two of the samplers after the study,
to allow for easier operation in the field. A fundamental problem was noted early on with the MetOne
spiral inlet, which was allowing particles greater then 2.5 //m to penetrate the inlet. The spiral inlet has
been replaced with a sharp cut cyclone.
Chemical composition of the aerosols at each site were within expectations with the exception of high
nitrate and OC in Philadelphia, where nitrate and sulfate both were about 20% of the total PM2 5 mass
and OC was about 50%. Results from most studies in the eastern US indicate that sulfate is the highest
species (-50% of the mass), followed by OC at about 30% of the mass, with nitrate accounting for less
than 5% or so of the mass. However, most previous studies have occurred during the summertime,
when temperatures are high and ammonium nitrate would be mostly in the gas phase. Finally, coarse
particle concentrations were highest in Phoenix and Rubidoux (about equal to the fine particle mass)
and only about 20% or less relative to the fine particle mass at Philadelphia and RTF, as expected.
Therefore, this study met its objective of testing the chemical speciation samplers under a fairly wide
range of chemical conditions.
Means, time series, and regression analyses were performed for all species measured, allowing
comparison among the samplers for a given variable at a given site. On the average, the major species
agreed within 10-15% among the FRM, Andersen, and Improve samplers. Sulfate had even better
agreement, which was observed across all samplers. The MetOne and VAPS samplers tended to be
high for species that normally have a coarse particle component (i.e., mass, Si, Fe, Ca, etc.). In general,
individual species from all samplers tracked each other, with the majority of correlation coefficients (r)
being greater then 0.85. A few exceptions were noted. More variability was observed for trace
elements (Si, K, Ca, Fe, Cu, Zn, Pb, and As).
Differences, on the order of up to 1 |ig/m3 on the average were observed among the samplers for
particle nitrate due to a possible positive artifact associated with determining nitrate on pre-fired quartz-
fiber filters, which usually is not observed with quartz-fiber filters that have not been pre-treated (Chow,
1995 JAWMA 45, 320). The quartz-fiber filter was used due to concerns regarding loss of nitrate
during vacuum XRF analysis (i.e., XRF has to be performed before the filter is extracted for ions
analysis). Tests comparing nitrate concentrations measured on Teflon filters, collected in parallel, with
Part I, Page iv
-------
and without having vacuum XRF analysis indicated loss of up to 40% of the nitrate, assumed to be
ammonium nitrate. An additional bias for collecting particulate nitrate was observed due to the method
of collecting particulate nitrate, where nitrate concentrations determined by the direct method (nitrate
measured directly on a filter behind a denuder) were up to 1.5 |ig/m3 lower than nitrate concentrations
measured by the indirect method (nitrate measured on a quartz-fiber filter behind a denuder and Teflon
filter plus nitrate measured on a quartz-fiber filter in parallel).
Differences also were observed among the samplers for organic carbon and appear to be due to filter
face velocity variations among the samplers. Lower flow rates appear to result in higher OC
concentrations; although EC is consistent among the samplers. A positive artifact was also noted for
OC and ranged from about 3.5 jig/m3 at Rubidoux to essentially zero at RTF. Based on the design of
the study, no information can be implied about OC negative artifacts, but the assumption has been made
in the above discussion that negative artifacts for OC are similar between Teflon and quartz-fiber filters
operating at the same face velocity.
Differences were observed between EC values reported the IMPROVE OC/EC protocol versus the
NIOSH protocol. The IMPROVE protocol reported EC values approximately 2 times higher then the
MOSH method. These differences are currently under investigation.
Ammonium ion as measured by the IMPROVE sampler was on average lower than on the other
samplers, even though a similar bias was not observed for nitrate or sulfate. It is postulated that
ammonium is being lost due to volatilization of the ammonium nitrate that is collected on the nylon filter in
the IMPROVE sampler. While nitric acid volatilized from the collected ammonium nitrate would be
collected by the basic (pH) nylon filter, ammonia would not be collected. It also is possible that the
basic filter is enhancing ammonium volatilization. More careful experiments need to be conducted to
establish if this potential bias is significant or not.
Conclusions. In general, the performance of the candidate samplers is reasonable for their first use in
the field. All samplers had operational problems that increased their variability, most of which have been
addressed by the manufactures. Tradeoffs exist among the samplers for ease of use, flexibility for
sampling, and cost. Performance of the samplers was excellent for sulfate and reasonable for other
stable species. However, real differences among the samplers exist for nitrate and organic carbon and
possibly ammonium as collected in the IMPROVE sampler. These differences are significant and can
possibly affect design of compliance strategies for controlling PM2 5 mass concentrations in air, as total
differences as high as 3-5 |ig/m3 are observed among the samplers for these two species. Results from
this study yield the following recommendations for the collection of nitrate and organic carbon:
The Teflon filter used for mass and XRF analysis should not be used for ions analysis,
particularly nitrate and ammonium ions, as these species are lost during XRF analysis.
Part I, Page v
-------
To minimize artifacts for the collection of aerosol nitrate, it should be measured using a denuder
(coated with MgO or Na2CO3) followed by a single filter (Nylasorb or Na2CO3). Measuring
nitrate on a quartz-fiber filter prepared for carbon analysis can results in a significant (1-3 |ig/m3)
positive artifact for aerosol nitrate, after accounting for volatilized nitrate measured on a nylon
filter behind a denuder and Teflon filter.
Organic carbon should be measured at the same face velocity as the Federal Reference
Method. This will result in similar negative biases between OC measured on a quartz-fiber filter
and that of a Teflon filter. Positive biases were observed on the quartz-fiber filter collecting
aerosol directly behind a PM2 5 inlet relative OC measured behind the same inlet that is followed
by an XAD-4 coated annular denuder. It is recommended that the speciation network
eventually consider use of an XAD-4 denuder or similar denuder for removing potential gas
phase artifacts followed by a quartz-fiber filter and a reactive backup filter to obtain OC with
minimal bias.
Part I, Page vi
-------
TABLE OF CONTENTS - Volume 1: PARTS I AND II
VOLUME I: INTRODUCTION. RESULTS. AND CONCLUSIONS
DISCLAIMER Part I Page i
ACKNOWLEDGMENTS Part I Page ii
EXECUTIVE SUMMARY Part I Page iii
TABLE OF CONTENTS - PARTS I AND II Part I Page vii
List of Tables - Parts I and II Part I. Page xi
List of Figures - Parts I and n Part I. Page xiii
List of Appendices - Parts I and II Part I. Page xiv
Part I: Introduction and Experimental Design
INTRODUCTION Part I Page 2
About this Report Part I. Page 2
Study Objectives Part I. Page 3
Overview of the Intercomparison Part I. Page 3
Study Design Part I. Page 6
Statistical Design Part I. Page 6
EXPERIMENTAL Part I. Page 8
Sampler Types and Rationale Part I. Page 8
Sampler Descriptions - The Chemical Speciation Samplers Part I. Page 9
Reference Ambient Air Sampler (RAAS) developed by
Andersen Instruments Part I. Page 9
Spiral Ambient Speciation Sampler (SASS) developed by
MetOne Part I. Page 10
Mass Aerosol Speciation Sampler (MASS) developed by
University Research Glassware Part I. Page 11
Sampler Descriptions - Historical Samplers Part I. Page 11
IMPROVE Sampler Part I. Page 11
Versatile Air Pollution Sampler Part I. Page 12
Sampler Descriptions - Federal Reference Method Part I. Page 12
SCAQMD PTEMP Sampler Part I Page 12
Part I, Page vii
-------
Chemical Speciation and Chemical Analysis Part I. Page 13
PM2 5 Mass Part I. Page 14
Trace Elements (Na-Pb) Part I Page 14
Sulfate, Nitrate, and Ammonium Ions Parti. Page 15
Organic and Elemental Carbon Part I. Page 15
Special Studies: XRF and Thermal Analysis for OC/EC Part I. Page 15
Loss of Nitrate During XRF Analysis Parti. Page 15
TOR vs TOT Analysis for OC and EC Part I. Page 16
Splitting Filters for Multiple Analyses Part I. Page 17
Filter and Denuder Preparation Parti. Page 17
Sampling Locations and Rationale Parti. Page 17
Program Schedule Parti. Page 18
Overall Program Schedule Part I. Page 18
Sampling Schedule Parti. Page 19
Preliminary Evaluation and Training Parti. Page 19
Site Installation and Decommissioning Parti. Page 19
Sampler Operation and Filter Shipping and Storage Part I. Page 19
Quality Assurance/Quality Control (QA/QC) Part I. Page 20
Field QA/QC Part I Page 20
Laboratory QA/QC Part I. Page 21
STATISTICAL ANALYSIS Part I. Page 21
Questions or Statistical Hypotheses Being Tested Parti. Page 21
Data Reporting Format Part I. Page 27
REFERENCES Part I. Page 28
Tables - Part I Part I Tables, Page 1
Figures - Part I Part I Figures, Page 1
Part II: Results and Conclusions
TABLE OF CONTENTS - Part H Part II, Page ii
List of Tables - Part II Part II, Page v
List of Figures - Part II Part II, Page vi
RESULTS Part II, Page 1
Quality Assurance Results Part II, Page 1
Part I, Page viii
-------
Performance Audit Methods and Verification Part II, Page 1
Flow Rate Audit Equipment Part II, Page 1
Barometric Pressure Audit Equipment Part II, Page 1
Temperature Audit Equipment Part II, Page 1
Field Audit Results Part II, Page 1
Limits of Detection Part II, Page 3
Field Blanks Part II, Page 3
Precision Part II, Page 4
Data Validation Part II, Page 4
Level la Data Validation Part II, Page 4
Level Ib Data Validation Part II, Page 5
Data Capture Part II, Page 5
Chemical Characteristics of PM2 5 Part II, Page 7
Chemical Components Reported in the 4 City Study Database Part II, Page 7
PM2 5 and Chemical Components Collected by the FRM Part II, Page 7
Volatilization of Aerosol Nitrate Part II, Page 9
Statistical Analysis Part II, Page 11
Means of PM2 5 Mass and Major Chemical Components Part II, Page 11
Explanations of Anomalous Data Part II, Page 12
Comparison of Means Among Samplers Part II, Page 13
Comparison Performance Criteria Part II, Page 14
Time Series and Analysis Part II, Page 15
Regression Analysis Part II, Page 16
Mass Part II, Page 17
Sulfate Part II, Page 17
Nitrate Part II, Page 17
Ammonium Part II, Page 18
Organic and Elemental Carbon Part II, Page 18
Trace Elements Part II, Page 19
T-TEST Results Part II, Page 20
Results from the Paired T-Tests Part II, Page 21
Paired T-Test Results for FRM and Andersen Samplers Part II, Page 22
Paired T-Test Results for FRM and MetOne Samplers Part II, Page 22
Paired T-Test Results for FRM and IMPROVE Samplers . . . Part II, Page 22
Paired T-Test Results for FRM and URG Samplers Part II, Page 23
Paired T-Test Results for the FRM and VAPS Samplers Part II, Page 23
Results from the ANOVA on Particle Nitrate Data Part II, Page 23
Aerosol Nitrate Part II, Page 24
Organic and Elemental Carbon Part II, Page 25
Part I, Page ix
-------
Denuded vs Non-Denuded Organic Carbon Results Part II, Page 25
Comparison between TOR and TOT for OC and EC in PM2 5 Part II, Page 26
Loss of Nitrate During Vacuum XRF Analysis Part II, Page 28
Field Experience with the Speciation Samplers Part II, Page 28
Andersen RAAS Part II, Page 29
Met One SASS Part II, Page 30
URGMASS Part II, Page 30
IMPROVE Part II, Page 30
VAPS Part II, Page 30
FRM Part II, Page 31
Sampler Costs and Related Spare Parts Part II, Page 31
DISCUSSION Part II, Page 32
Expectations Part II, Page 32
Major Questions Addressed Part II, Page 32
Q1. How well do PM2 5 mass and the chemical components
of mass agree between the FRM and the chemical speciation
samplers tested in this study? Part II, Page 33
Q2. How well can the FRM mass be reconstructed by summing
the chemical components measured by the speciation samplers Part II, Page 34
Specific Hypotheses Related to Questions Ql andQ2 Part II, Page 35
Denuded vs Non-Denuded Organic Carbon Results Part II, Page 38
Q3. How well do the measured concentrations from the
various speciation samplers agree? Part II, Page 39
Q4. What are the causes of the differences among the speciation
samplers for measured concentrations of mass and the
components of mass if they exist Part II, Page 39
Specific Hypotheses Related to Questions Q3 andQ4 Part II, Page 39
SUMMARY AND CONCLUSIONS Part II, Page 44
Site Characteristics Part II, Page 44
Operations Part II, Page 44
Measurement Performance Part II, Page 44
Implications Part II, Page 45
Nitrate Part II, Page 45
Carbon Part II, Page 46
REFERENCES Part II, Page 47
Part I, Page x
-------
Tables - Part I Part H Tables, Page 1
Figures - Part I Part n Figures, Page 1
VOLUME II. APPENDICES
Part I, Page xi
-------
List of Tables - Parts I and II
Parti
Table 1-1. Analyte Listing for Speciation Sampler Intercomparison
Table 1-2. List of Standard Operating Procedures for Field and Laboratory Efforts During the
Chemical Speciation Monitor Evaluation Study.
Table 1-3. Experimental Design Including Filter and Denuder Preparation.
Table 1-4. Measurements Made at Rubidoux, CA in Conjunction with the Chemical Speciation
Sampler Evaluation Study.
Table 1-5. Measurements Made at Phoenix, AZ in Conjunction with the Chemical Speciation
Sampler Evaluation Study.
Table 1-6. Overall Planned Study Schedule.
Table 1-7. Field Operations Sampling Schedule.
PartH
Table II-1. Summary of Flow Audit Results.
Table II-2a. Limits of Detection in ng m"3.
Table II-2b. Average Field Blank Data for All Species and Samplers Averaged Across All Sites In
Atmospheric Concentrations.
Table II-2c. Precision (as % CV) Achieved by FRM and Speciation Samplers Based on Results
from the Collocated Samplers at Rubidoux.
Table II-3. Valid Data Capture in Percent by Sampler and Major Species.
Table II-4. Summary of Problems Encountered In the Field During Operations of Sampler
Evaluated in this Study.
Table II-5. Species Concentration Data for the FRM at Each Location of the 4-City Study.
Table II-6. Estimated PM2.5 Mass Balance of Species versus Measured PM2.5 Mass (ug/m3) for
the FRM at Each Site.
Table II-7. Average Volatilized Nitrate (NO3 V) in ug/m3 Observed for Each Sampler at Each
City.
Table II-8a. Mean Analyte Concentrations for Each Sampler at All Sites.
Table II-8b. Ratio of Speciation Sampler to FRM for Chemical Components by Site.
Table II-9. Regression Statistics of FRM (x-axis) versus Speciation Samplers (y-axis) for All Sites,
Samplers, and Major Species.
Table 11-10. Results from the Paired t-Tests Between the Speciation Samplers and FRM Samplers
for Each Analyte/Site.
Table II-11. Results from the ANOVA for Examining Equivalency Among the Samplers for
Paniculate Nitrate.
Table II-12a. Nitrate Concentrations (ug/m3) Measured on Teflon (T) or Quartz-Fiber (Q) Filters by
Sampler Type Averaged Over the Study Period.
Table II-12b. Total Particle Nitrate Concentrations (ug/m3) Measured by Each Sampler Averaged
Over the Study Period.
Part I, Page xii
-------
Table II-12c. Volatilized Nitrate Concentrations (ug/m3) Measured by Each Sampler Averaged Over
the Study Period.
Table II-12d. Sulfate Concentrations (ug/m3) Measured on Teflon (T) or Quartz-Fiber (Q) Filters
Averaged Over the Study Period.
Table II-13a. Nitrate Concentrations (ug/m3) Measured on Teflon (T) or Quartz-Fiber (Q) Filters
Averaged Over the Study Period.
Table II-13b. Total Particle Nitrate Concentrations (ug/m3) Measured by Different Denuder-Filter
Pack Methods Averaged Over the Study Period.
Table II-13 c. Volatilized Nitrate Concentrations (ug/m3) Measured by Different Denuder-Filter Pack
Methods Averaged Over the Study Period.
Table II-13d. Sulfate Concentrations (ug/m3) Measured on Teflon and Quartz Filters Averaged Over
the Study Period.
Table 11-14. Loss of Nitrate Resulting from Analysis of Teflon Filter by Vacuum XRF.
Table 11-15. Summary of Site Operators Surveys Regarding Speciation Sampler Setup and
operation.
Table 11-16. Recommended Spare Parts and Supplies for Use of Chemical Speciation Samplers and
FRM Used in the Chemical Speciation Evaluation Study.
Part I, Page xiii
-------
List of Figures - Parts I and n
Parti
Figure I-la. Schematic of the Andersen RAAS Sampler.
Figure I-lb. Picture of the Andersen RAAS Sampler Deployed in the Field at RTF.
Figure I-2a. Schematic of the MetOne SASS Sampler.
Figure I-2b. Picture of MetOne Sampler Deployed in the Field at RTF.
Figure I-3a. Schematic of the URG MASS Sampler.
Figure I-3b. Picture of the URG MASS Sampler Deployed in the Field at RTF.
Figure I-4a. Schematic of the IMPROVE Sampler.
Figure I-4b. Picture of the IMPROVE Sampler Deployed in the Field at RTF
Figure I-5a. Schematic of the VAPS Sampler.
Figure I-5b. Picture of the VAPS Sampler Deployed in the Field at RTF.
Figure I-6a. Schematic of the Federal Reference Method Samplers.
Figure I-6b. Picture of FRM Samplers Deployed in the Field at RTF.
Figure 1-7. Schematic of the SCAQMD Multi-Channel Fine Paniculate Sampler.
Figure 1-8. Top - Samplers on the Platform at Rubidoux, CA.
Figure 1-9. Sampling Platform at Phoenix, AZ.
Figure 1-10. Philadelphia Sampling Site. Top - Roof View.
Figure 1-11. Research Triangle Park Sampling Site.
PartH
Figure II-l. Frequency Distributions, Given as Box and Whisker Plots of PM2.5 Species at Each of
the Four Cities Studies
Figure n-2. Chemical Mass Balance of PM2.5 at Each City.
Figure n-3. Time Series Plots.
Figure n-4. Regression Analysis Plots.
Figure D-5. Denuded and Non-Denuded OC Samples as a Function of the TOR Carbon Fractions.
Figure D-6. Organic Carbon concentrations from FRM and VAPS versus Total FRM PM2.5 Mass.
Figure n-7. Organic and Elemental Carbon as a function of Face Velocity.
Figure n-8. Loss of Aerosol Nitrate from Teflon filters Due to Vacuum XRF Analysis
Part I, Page xiv
-------
Volume n
List of Appendices - Parts I and II
Appendix A: Sample Analysis Methods for Chemical Speciation
Appendix B. Standard Operating Procedures
Appendix C. Summary of Comments from Field and Laboratory Analysis Logbooks
Appendix D: Final Data Archive, By Species, Site, Date, and Sampler for the 4-City Study
Appendix Dl: Final Data Archive, By Species, Site, Date, and Sampler for Replicate No.2
Measurements at Rubidoux
Appendix E: Time Series Plots for All Species Measured in the 4-City Study
Appendix F: Regression Analysis Plots for All Species Measured in the 4-City Study
Appendix G: Absolute Differences Between the FRM, (Reference Sampler), and the Speciation
Samplers by Site and By Sampling Period
Appendix Gl: Percent Differences Between the FRM, (Reference Sampler), and the Speciation
Samplers by Site and By Sampling Period
Appendix H: Field Evaluation of a Spiral and Cyclonic PM2 5 Size Selective Separator for the
MetOne Ambient Chemical Speciation Sampler-SASS
Appendix I: Evaluation of PM2 5 Size Selectors Used in Speciation Samplers (Peters et al. 2000)
Appendix J: Comparison of Particulate Organic and Elemental Carbon Measurements Made with the
IMPROVE and NIOSH Method 5040 Protocols
Part I, Page xv
-------
Parti
Introduction and Experimental Design
Part I, Page 1
-------
INTRODUCTION
On July 18, 1997, the U.S. EPA promulgated a new NAAQS for paniculate matter (PM) in 40 CFR
Parts 50, 53, and 58, Federal Register (EPA 1997a; EPA 1997b). In addition to slightly revising the
previous PM10 standard, EPA added a new standard for fine particles less than 2.5 |j,m in aerodynamic
diameter, known as PM2 5. To develop meaningful relationships between PM2 5 levels at receptors and
source emissions and for better understanding the causes of high PM2.5 concentrations, in particular
secondary components formed in the atmosphere through chemical reactions and condensation, it is
necessary not only to sample for PM2 5 mass, the NAAQS indicator, but also for the chemical
components of PM2 5. A sampling program of this type has been initiated by EPA (EPA 1999
Guidance Document) that will consist of up to 300 sites at which the major chemical components of
PM2 5 will be measured in the collected aerosol. Since information from this network will be used for
the identification of sources contributing to high PM2 5 mass concentrations, development and evaluation
of control strategies, measurement of trends, and support of health studies, it is important that there be
national consistency in the species concentrations measured by the PM2 5 speciation network. In
particular, 54 of these PM2 5 chemical speciation sites will become part of the National Air Sampling
Stations (NAMS) network and will provide nationally consistent data for assessment of trends (EPA
1997b).
Development of chemical speciation samplers for the National PM2 5 Sampler Procurement Contract
(National Sampler Contract) was based on performance, rather than design criteria. This has allowed
innovation in the development of these samplers and has resulted in the development of three slightly
different samplers for meeting the specified performance criteria. Also as a result of this approach, a
guidance document on chemical speciation of paniculate matter has been prepared by EPA (EPA,
1999) and reviewed by an external peer-review panel (Speciation Expert Panel; Koutrakis, 1998). In
their first review, the expert panel recommended an intercomparison among the chemical speciation
samplers. The intercomparison also should include other historically accepted samplers (e.g., the
improved IMPROVE sampler, the Harvard Sampler, or some other sampler) and the PM2 5 Federal
Reference Method (FRM). The chemical species to be determined should include those recommended
by the expert panel (Koutrakis, 1998) and as specified in the guidance document for chemical speciation
(EPA, 1999). The program plan for EPA's Chemical Speciation Sampler Evaluation Study (Solomon
et al. 1998) outlines the approach and details the implementation of the intercomparison study to
perform an initial evaluation of the chemical speciation samplers developed in response to the National
Sampler Contract and several other samplers developed earlier and independently of the EPA national
program.
About this Report
This draft final report provides results from EPA's Chemical Speciation Sampler Evaluation Study (4
City Study). The data presented in this report have been validated through Level 2b, that is, the data
have undergone multi-variate statistical analyzes for consistency and known physical relationships and
interpretive data analysis (NARSTO 1999). Part I of this report outlines the study, provides a
Part I, Page 2
-------
summary of the samplers and the chemical analysis methods, and outlines the major questions and
hypotheses to be addressed by this evaluation. Part II presents the results. First, quality assurance
results are summarized, including operations and maintenance and systems and performance audit results
followed by a summary of the chemical characteristics observed at each location. Next, results are
presented from the statistical evaluations of the data, including time series analysis, regression analysis,
difference analysis, T-test, and Analysis of Variance. In the Discussion Section, each hypothesis noted
in the program plan, and Part I of this document is addressed to the extent possible and within the
limitations of the study design. Lastly, an overall summary is provided.
Study Objectives
The objective of this sampler intercomparison study is to determine if there are differences among the
three PM2 5 chemical speciation samplers developed in response to the National Sampler Contract and
how these samplers compare relative to other historical samplers, and to the FRM. While the FRM is
the "gold" standard for mass, there are no such standards for the chemical components of PM2 5. Thus,
this intercomparison only establishes the relative equivalence of the samplers to each other on a species
by species basis. For semi-volatile species (those in dynamic equilibrium between the gas and particle
phases; e.g., for ammonium nitrate), the FRM using Teflon filters provides only a lower limit on the
expected mass loading, since there is potential for loss of nitrate and semi-volatile organic species
(SVOC) from the inert Teflon filters. For stable species, the FRM should provide an accurate estimate
of the mass loading for those species. Chemical speciation samplers used historically [e.g., the Versatile
Air Pollution Sampler (VAPS) developed under an EPA contract, the Caltech gray box sampler
(Solomon et al., 1989), or the South Coast Air Quality Management District's PM10 Technical
Enhancement Program (PTEP) sampler (SCAQMD, 1996) should provide a less biased value for semi-
volatile species (i.e., ammonium nitrate) and provide an additional set of samples for comparison;
however, they still can only be compared on equivalent bases.
Overview of the Intercomparison
Collecting atmospheric particulate matter using the FRM with Teflon filters can result in negative
sampling artifacts associated with the collected sample. Potential artifacts include the loss of volatile
species, such as ammonium nitrate (Solomon et al., 1988, Hering et al., 1988; Hering and Cass 1999)
and semi-volatile organic compounds (Cui et al., 1997; Eatough et al. 1995). Use of other filter media
also may result in negative or positive sampling artifacts. The magnitude of these potential artifacts
depends upon the atmospheric concentration of the species being affected, the temperature, relative
humidity, and other variables (e.g., for nitrate, Russell and Cass, 1986; Hering and Cass, 1999). The
chemical speciation samplers developed for National Sampler Contract have been designed to minimize
these potential biases or artifacts by the use of diffusion denuders to remove gas phase species and
reactive substrates to collect species that may volatilize during or after sampling from the inert filter (e.g.,
Teflon membrane) where the aerosol is collected. Therefore, to evaluate the performance of these
chemical speciation samplers they must be able to properly determine the chemical components of
PM2 5 under a variety of atmospheric conditions, each of which will place different stresses on the
Part I, Page 3
-------
performance of the sampler designs. For this study, this was accomplished by sampling at different
locations throughout the country, since the composition of the atmospheric aerosol is not uniform across
the country (Pace, 1998). For example, some areas have high nitrate and low sulfate levels (Los
Angeles, CA: Solomon et al, 1989), while others (e.g., the eastern part of the United States) have
relatively high sulfate and low nitrate levels (Hidy 1994, Pace, 1998). Still, other areas are dominated
by aerosol rich in organic compounds derived from automobile exhaust (Los Angeles, CA: Schauer,
1996) , by organic aerosol derived from wood smoke combustion (Fresno, CA: Schauer, 1998), or
from by organic aerosol derived from natural biogenic emissions (e.g., Southeast US). Some areas of
the country are highly influenced by crustal material (e.g., Southwest US: Pace 1998; Eldred et al.
1998a). In actuality, several of these conditions exist simultaneously, with one or two components being
higher then the others (Pace 1998; Eldred, 1998a, Solomon et al. 1989).
A variety of atmospheric chemical conditions also may be observed at one location during different seasons
(Pace, 1998). For example, sulfate is likely highest in the east during the summer when photochemistry is
high, while nitrate is highest in the west in the winter when cool temperatures drive the ammonium nitrate
equilibrium with nitric acid and ammonia to the aerosol phase. However, due to the need to have results
by mid-1999, the study was conducted over about an eight week period at four different locations to obtain
as wide a difference in chemical atmospheres as possible. These constraints, however, resulted in
limitations, and follow-on studies will have to occur to fully test the equivalency of these samplers under a
wider variety of conditions. For example, by sampling in the winter in the east, we missed the highest sulfate
concentrations which occur in the summer (Hidy, 1994), we did not sampling at a site with high wood
smoke emissions, we sampled in Phoenix for crustal material in the winter when the highest crustal
concentrations are likely to be observed in the hot dry summers, and the samplers did not experience
extreme cold temperatures as might be expected in the northern mid-west or hot humid summers as
experienced during the summer in the east.
Due to time and resource limitations, sampler evaluation is being conducted in four phases. Phase I is
centered on sampling in areas with the following atmospheric conditions: high sulfate and low nitrate
(east coast US), high nitrate and low sulfate (California), and high crustal material (Phoenix, AZ). The
fourth site is located near ORD headquarters in Research Triangle Park to allow for a more thorough
evaluation of the samplers and their in-field operational performance. Phase n is taking place in Seattle,
WA from March-July, 1999 and is evaluating the efficiency and capacity of organic diffusion denuders
and reactive back-up sorbents, including ones not currently planned for the chemical speciation
samplers. Phase UJ is an extensive comparison of the same speciation samplers used in the 4 City
Study, as well as several others that have been developed at universities. Comparisons in Phase UJ also
will be made to a number of species specific continuous methods for the major components of PM2 5.
Phase IV is a ten city study where the sites will have at least 2 speciation samplers and be operated by
the States.
The time schedule for Phase I of the study dictated that we sample more frequently than every 6th day,
as the results are needed by OAQPS by mid-June, 1999 for input into the decision process for choosing
Part I, Page 4
-------
chemical speciation samplers for the National Air Monitoring Stations (NAME) TRENDS network.
Therefore, samples were collected every-other-day. The statistical design required a minimum of 10-15
samples. To ensure that a sufficient number of samples were collected to meet that objective, 20
sampling periods were attempted. Samples were analyzed for the major chemical components using
standard analytical techniques as described below and recommended by the expert panel that reviewed
the guidance document (Koutrakis, 1998). Data analysis provided a robust test of the equivalency of
the samplers studied and, within the limitations of the study, reasons for differences among the methods
tested.
Phase n involves sampling in Seattle, WA with a focus on understanding the collection of organic
material (aerosol OC and semi-volatile organic compounds) under wood smoke conditions in a manner
that will minimize negative and positive sampling artifacts for organic species. These systems include a
denuder to remove semi-volatile organic compounds that are in the gas phase and may be collected by
the downstream quartz fiber filter, followed by a reactive sorbents (denuder, PUF, or impregnated
filter). The evaluation includes determining capacity, efficiency, and comparability of two denuder
systems and an evaluation of the sorbents located behind the quartz fiber filter. The first system uses
XAD-4 coated onto annular denuders as was proposed for use in two of the chemical speciation
samplers procured through the National Sampler Contract (University Research Glassware and
Andersen Instruments). The second system uses a multi-channel parallel plate denuder composed of
carbon impregnated filters (GIF) (Eatough et al., 1993). Both denuders are followed by quartz fiber
filters which are then followed either by second XAD-4 coated denuder, an GIF filter, an XAD-4
impregnated Whatman filter, PUF cartridge, or an XAD-4-sorbent bed. XAD-4, PUF cartridges, and
quartz fiber filters can be extracted and individual species can be determined to obtain a mass balance
between the SVOC, aerosol organic species collected on the quartz fiber filter, and the SVOC
volatilized from the quartz fiber filter and collected on the reactive back-up medium, on a species-by-
species basis. The GIF filter can be analyzed for organic carbon using thermal desorption.
Phase m will involve sampling in Atlanta, GA where biogenic VOC emissions are known to be high in
the summer (Chameides et al. 1988). The Atlanta intercomparison is an integral part of the EPA
Supersites Program (EPA 1998). The same set of chemical speciation monitors will be operated in
Atlanta as were operated in the 4 City Study. In addition, several other speciation samplers are
included in the intercomparison along with the potential for comparisons to a number of species specific
continuous methods for sulfate, nitrate, ammonium, trace elements (Na - Pb), organic carbon, and
elemental carbon. Details of the Atlanta study are described in Hering (1999).
Phase IV, the Ten City Study is still in planning. It is anticipated, that each site will have at least two
different chemical speciation samplers, operate on a 1 in 3 day schedule from about October 1999
through March 2000, and have chemical analysis performed in the national laboratories established to
support the chemical speciation sampling network. The goal of this study is to evaluate the samplers
Part I, Page 5
-------
under more severe extremes of temperature, as well as higher crustal material and wood smoke
loadings.
Study Design
The design of this program is constrained by limitations in the time frame allowed for the experiment and
in resources available to complete the program (e.g., number of samplers, personnel, and funding).
However, the statistical design was prepared understanding these limitations and the design chosen
provides a robust evaluation of the samplers relative to each other, to several samplers used historically
to obtain similar data, and to the FRM. The overall design is detailed below.
Statistical Design
The primary objective of this study is to determine if there are differences in the measured
concentrations of the chemical components of PM2 5 mass as determined by the three PM2 5 chemical
speciation samplers available on the National Sampler Contract. Comparisons also will be made to
two historical samplers and to the FRM using these samplers as a relative reference. A secondary
objective of this study is to evaluate the operational performance or practicality of the samplers in the
field, that is, reliability, ruggedness, ease of use, and maintenance requirements.
There are three major scientific hypotheses to be addressed by this intercomparison study.
*• One is associated with reconstructing the FRM mass.
> The second is associated with comparing the measured chemical concentrations among
the various speciation samplers, which consists of two parts:
! The first part is associated with examining differences among the samplers,
without regard to why there are differences, if they exist.
! The second part examines why there are differences, if they exist. Some are
expected due to the slightly different methods employed.
> A third set of hypotheses is given dealing with the potential affect of different analytical
methods on measured concentrations of the chemical components of PM2 5. These
include the effect of vacuum X-ray fluorescence (XRF) or atmospheric pressure XRF
on nitrate concentrations measured on Teflon filters and the effect of thermal optical
reflectance (TOR) vs. thermal optical transmittance (TOT) on the determination of
organic and elemental carbon (OC/EC) concentrations from pre-baked quartz fiber
filters.
The first two hypotheses are predicated on the assumption that the cutpoints (50% collection efficiency)
for the samplers used in this study have essentially the slope and 50% cutpoint. This is a required
Part I, Page 6
-------
assumption to address these hypotheses. Also, it is important to establish the precision of the
instruments, which was obtained by collocating samples at one site (Rubidoux, CA). While this
provides only a limited assessment of the precision, it provides a first cut estimate of the precision for the
statistical analyses performed to understand the data. If for example, the precision is estimated at 50%,
then determining differences among samplers is not as informative as if the precision were 10-15%. As
a benchmark, the coefficient of variation for the differences in concentrations from collocated FRM
instruments is required to be less than 10%, according to 40 CFR Part 58, Appendix A. Depending on
the species, we anticipate a range of precision from less than 10% to about 30%.
A detailed list of hypotheses is given in the Statistical Analysis section.
Part I, Page 7
-------
EXPERIMENTAL
Sampler Types and Rationale
Chemical speciation samplers have been developed and built by three different manufacturers under the
National Sampler Contract procurement. The need for PM2 5 chemical speciation monitoring is
described under 40 CFR, Parts 53 and 58 (EPA 1997). The three samplers are the Reference Ambient
Air Sampler (RAAS) developed by Andersen Instruments Incorporated (Andersen), Mass Aerosol
Speciation Sampler (MASS) developed by University Research Glassware Corporation (URG), and
Spiral Ambient Speciation Sampler (SASS) developed by Met One Instruments (MetOne). The
external peer-review committee (Koutrakis, 1998) recommended comparison of these samplers under
field conditions in different areas of the country and different seasons. They also recommended
comparison to samplers used previously that have been accepted historically as providing data of known
uncertainty, and to the FRM.
Historical methods included in this study were the National Park Services' IMPROVE (Interagency
Monitoring of Protected Visual Environments) sampler modified to include 47 mm filters as suggested by
the expert review panel (Koutrakis, 1998), the Versatile Air Pollution Sampler (VAPS)(URG
Corporation; four available), and the PTEP sampler (SCAQMD, 1996) operated by the South Coast
Air Quality Management District (SCAQMD) at their Rubidoux, CA site. These samplers are well
characterized for collecting relatively unbiased samples suitable for chemical analysis of major PM
composition.
Two FRM samplers were operated at each site to allow for chemical characterization of the collected
sample similar to that being obtained by the chemical speciation samplers. One FRM collects aerosol
samples on Teflon filters for mass and trace elements (Na - Pb), while the other FRM used quartz-fiber
filters for determination of ions (SO4=, NO3", and NtV), OC, and EC.
The FRM should provide a suitable reference for stable species, such as many of the trace metals and
sulfate. The historical samplers should provide a reference for labile compounds (nitrate ion and semi-
volatile organic compounds [SVOC]) as they used diffusion denuders and reactive backup filters, similar
to the chemical speciation samplers, thus minimizing the potential gain or loss of these species when
using only Teflon or quartz fiber filters. The IMPROVE sampler should provide nearly artifact free data
for nitrate, while the VAPS should provide nearly artifact free data for nitrate and organic carbon.
During Phase I, only the VAPS used a denuder for removing gas-phase semi-volatile organic
compounds (referred to here after as an organic denuder), as there is currently considerable uncertainty
in using organic denuders as well as the desire to leave research oriented approaches to more careful
examination. Collection of organic carbon using denuders and reactive collection media is addressed in
Phase II activities.
Both the VAPS and the IMPROVE samplers have been used and evaluated in numerous studies over
the last decade, and thus, provide a reference to many other databases (Shaibal et al. 1997;
Part I, Page 8
-------
Sommerville et al. 1994; Stevens et al., 1993; Pinto et al. 1998; Mathai et al. 1990; Cahill, 1993). The
PTEP sampler, only operated at Rubidoux, also falls into this category as it has been used for nearly a
decade by the South Coast Air Quality Management District (SCAQMD) in southern California
(Teffera et al., 1996; SCAQMD, 1996). The PTEP sampler also uses methods similar to the chemical
speciation samplers.
Sample analysis, which is described in more detail later, included mass by gravimetric analysis, ions
(sulfate, nitrate, and ammonium) by ion chromatography (1C), OC/EC by thermal-optical reflectance
(TOR), and elemental analysis by energy dispersive X-ray fluorescence (XRF). Mass was always
determined on Teflon filters following FRM protocol for filter equilibration and weighing.
Concentrations of trace elements (Na - Pb), were measured on the same filter used for mass
determinations. Ions are determined from aqueous extracts of either Teflon (wet with 50 jil ethanol
before extraction), quartz-fiber, or nylon filters. Nylon filters analyzed for only for nitrate were extracted
in 1C eluent and those analyzed for nitrate, sulfate, and ammonium ions were extracted in water. OC
and EC were measured on quartz-fiber filters that have been baked at 600°C for 2 hours to lower
background carbon levels below 0.2 |j,g/cm2 total carbon. Quartz-fiber filters analyzed for ions were
split to allow for carbon and ions analysis. All other filters were kept whole for analysis.
Sampler Descriptions - The Chemical Speciation Samplers
Design of the three chemical speciation samplers for the National PM2 5 Network can be found in the
EPA chemical speciation guidance document (EPA, 1999). The draft guidance document outlines the
general design of these samplers as envisioned for the PM2 5 network; although they are not likely the
final designs to be implemented, as this and future field evaluations of the samplers may result in
modifications to the samplers. Specific designs of the samplers for this intercomparison are given below.
In general, each sampler draws air at a specified flow rate through a size selective inlet that removes
particles greater than a specified size with a 50% collection efficiency or cutpoint. For the samplers
employed in this study the cutpoint is 2.5 jim. As recommended by the expert peer-review panel
(Koutrakis, 1998), the efficiency of collection (slope and cutpoint) for each sampler should closely
resemble that of the FRM, and that was under the control of the manufacturers. Described below are
the three samplers provided to EPA for the National Sampler Contract procurement by URG, MetOne,
and Andersen.
Reference Ambient Air Sampler (RAAS) developed by Andersen Instruments
A schematic flow diagram of the Andersen RAAS is shown in Figure la, with a picture of the sampler
given in Figure Ib. It consists of a size selective inlet followed by two PM2 5 cyclones in parallel, the
outlets of which are connected to separate sampling manifolds. These cyclones are used to remove
particles greater than 2.5 micrometers with a 50% collection efficiency, when operated at 24 Lpm. The
flow is then split in each manifold into 2 channels (maximum of 3) for at total of up to 6 channels. Of the
four channels used in this study, the first channel (labeled 1 in Figure la) is used to estimate atmospheric
concentrations of paniculate organic and elemental carbon (OC/EC). The flow rate in this channel is
Part I, Page 9
-------
7.3 Lpm. In the second channel (labeled 2 in Figure la), particulate matter is collected on a Teflon filter
for analysis of mass and trace elements (Na - Pb) by energy dispersive X-ray fluorescence (XRF). The
flow rate through this channel 2 is 16.7 Lpm. In the third channel (labeled 3 in Figure la) particulate
matter also is collected on a Teflon filter, which is extracted in water and analyzed for sulfate, nitrate,
and ammonium ion concentrations by ion chromatography (1C). The last channel (labeled 4 in Figure
la) is used to obtain a nearly unbiased estimate of fine particle nitrate by removing acidic gases (e.g.,
HNO3) from the air stream using a diffusion denuder coated with MgO and collecting aerosol nitrate on
a reactive Nylasorb (nylon) backup filter. This assumes the denuder is efficient for HNO3 and other
acidic gases that might be collected on the nylon filter and analyzed as nitrate and that the nylon filter
does not collect NO2. The filter is extracted in 1C eluent and analyzed by 1C for nitrate. In all channels,
critical orifices control the flow and the flow rates are monitored using electronic mass flow sensors. All
internal components before the filter holders or denuders are Teflon® coated and no grease or oil is used
in the sampler's design. The system also monitors continuously relative humidity (RH), barometric
pressure (BP), orifice pressure (OP), ambient temperature (T), manifold temperature (MT), meter
temperature (MeT) and cabinet temperature (CT). Data can be downloaded through a RS-232C serial
port, which also allows for two way remote communication (Andersen, 1999).
Spiral Ambient Speciation Sampler (SASS) developed by MetOne
A schematic flow diagram for the MetOne SASS sampler is presented in Figure 2a, with a picture of the
sampler shown in Figure 2b. The SASS has 5 separate channels, operated through a common
controller and pump. For the current Four City Study, each channel contained a spiral impactor
designed to give a 2.5 |im cut-point (50% collection efficiency) with a slope and cutpoint similar to the
FRM when operated at 6.7 Lpm (MetOne, 1999). (Note, results from this study indicted that under
high coarse particle loading conditions, the Spiral impactor allowed large particles to penetrate to the
filter. The Spiral is being replaced by a sharp cutpoint cyclone (SCC) developed by BGI, Incorporated.
The rest of the design for the SASS sampler is staying essentially the same.} The first channel (labeled
1 in Figure 2a) collects particulate matter on a Teflon filter that is analyzed for atmospheric
concentrations of PM2 5 mass and trace elements (Na - Pb). The second channel (labeled 2 in Figure
2a) also collects particulate matter on a Teflon filter that is analyzed for sulfate, nitrate, and ammonium
ion concentrations. A MgO coated aluminum honeycomb diffusion denuder is located behind the spiral
impactor in the third channel (labeled 3 in Figure 2a). This denuder is used to remove acidic gases (e.g.,
HNO3) from the sampled air stream. The MgO denuder is followed by a Nylon filter that is analyzed
for nitrate as described above. As in the RAAS sampler, the denuder/reactive filter pair is used to
obtain a nearly unbiased estimate of aerosol nitrate. This assumes the denuder is efficient for HNO3 and
other acidic species that might be analyzed as nitrate, and that the nylon filter does not collect NO2. The
fourth channel (labeled 4 in Figure 2a) contains two baked quartz-fiber filters located behind the spiral
impactor. The first quartz-fiber filter is analyzed for OC/EC by thermal-optical reflectance, while the
second quartz-fiber filter is archived. The fifth channel (labeled 5 in Figure 2a) also contains 2 baked
quartz-fiber filters as a replicate set to channel 4. This set of quartz fiber filters are archived for future
use. In Phase HI (Atlanta), it is anticipated that a elemental carbon honeycomb diffusion denuder will
be available for use in channel 5. This denuder is used to remove semi-volatile organic compounds that
Part I, Page 10
-------
may interfere, as a positive artifact, with the OC measurement. The flow rate through each channel is
nominally 6.7 Lpm and is controlled by a critical orifice. The flow rate in this instrument is monitored
using electronic mass flow sensors.
Mass Aerosol Speciation Sampler (MASS) developed by University Research Glassware
(URG)The URG MASS sampler is shown in Figure 3a with a picture of this sampler given in Figure 3b.
This sampler consists of two modules (URG MASS 400 and MASS 450), each with an FRM PM10size
selective inlet and a WINS impactor for the collection of PM2 5 aerosol. The MASS 400 is equipped
with a Na2CO3 denuder before the WINS impactor but after the PM10 size selective inlet. This denuder
is used to remove acidic gases much like the MgO denuders discussed above. The particles less than
2.5 |im are collected on the top filter of a dual filter pack, which is an inert Teflon filter that is analyzed
for PM2 5 mass and trace elements (Na - Pb). The backup nylon filter efficiently collects nitrate that
may have vaporized from the front Teflon filter during sampling. Nitrate ion is quantified using 1C after
extraction from the Teflon and nylon filters as described above for the RAAS sampler. The sum of
nitrate measured on the Teflon and nylon filters provides a nearly bias free estimate of fine particle
nitrate. This assumes the denuder is efficient for HNO3 and that the nylon filter does not collect NO2.
The MASS 450 contains a single filter pack containing one pre-baked quartz-fiber filter. This filter is
split in half with OC and EC determined from one half and sulfate, nitrate, and ammonium ions
determined on the other half. An organic denuder (XAD coated annular denuder) is not used here, but
will be used in Phase m of the study following recommendations from Phase n. The flow rate through
each module is nominally 16.7 Lpm. Flow is monitored using a dry gas meter with a feed back loop to
the controller to adjust for variations in flow rate as particles are collected on the filter.
Sampler Descriptions - Historical Samplers
Historical samplers include the IMPROVE, VAPS, FRM, and PTEP samplers, the latter being operated
only at Rubidoux as part of a SCAQMD PM chemical characterization study (SCAQMD, 1996).
IMPROVE Sampler
Detailed descriptions of the IMPROVE sampler can be found in Eldred et al. (1998b). A schematic
diagram of the IMPROVE is given in Figure 4a with a picture of the sampler given in Figure 4b. In
general, the IMPROVE sampler consists of several modules each of which is dedicated to collecting a
series of related chemical components of the atmospheric aerosol. Each module consists of a size
selective inlet, a cyclone to provide a PM2 5 size cutpoint based on the specified flow rate, filter media
for sample collection, a critical orifice that provides the proper flow rate for the desired size cutoff, and a
vacuum pump to produce the flow. Flow rate is not monitored continuously, but are verified prior to
and after each sampling period. The IMPROVE samplers consist of up to four parallel modules, and a
common controller (timer) as described in Eldred et al. (1998). Only three modules are used in this
study, as the fourth is typically used to collect PM10. The first module (labeled 1 in Figure 4a) collects
PM2 5 on a Teflon filter, for determining atmospheric concentrations of PM2 5 mass and trace elements
(Na - Pb). The second module (labeled 2 in Figure 4a) includes a Na2CO3 denuder before the PM2 5
Part I, Page 11
-------
cyclone to remove acidic gases (e.g., HNO3) followed by the cyclone and a nylon filter. This nylon filter
is analyzed for sulfate, nitrate, and ammonium ions. The third module (labeled 3 in Figure 4a) collects
PM on a pre-baked quartz-fiber filter. This filter is analyzed for OC and EC.
Versatile Air Pollution Sampler
The VAPS sampler is shown in Figure 5a with a picture of the sampler given in Figure 5b. A PM2 5
cutpoint is obtained using a size selective impactor followed by a virtual impactor with a PM2.5 cutpoint.
The coarse particles follow the minor flow (3 Lpm) and are collected on a Teflon filter from which
coarse (PM10-PM2 5) particles mass is obtained. The fine (< PM2 5) particle flow (30 Lpm) is split
evenly between two channels. One channel (labeled 1 in Figure 5a) contains a diffusion denuder coated
with Na2CO3 followed by Teflon/nylon filter pack as described above. The Teflon filter will be analyzed
for mass and trace elements (Na - Pb). The Na2CO3 denuder is extracted and analyzed for nitrate to
give an estimate of ambient nitric acid concentrations. The second channel (labeled 2 in Figure 5a),
contains an XAD coated annular denuder, designed specifically for the VAPS (Gundel, personal
communication) to remove gas phase semi-volatile organic compounds that might be collected by the
quartz-fiber filter that follows the denuder. The quartz-fiber filter is analyzed for OC and EC
concentrations.
Sampler Descriptions - Federal Reference Method
The experimental design of the two FRM samplers is schematically illustrated in Figure 6a with a picture
of the samplers given in Figure 6b. Two FRM samplers will be used at each site to obtain a chemical
characterization of the collected aerosol in a manner similar to the other samplers. One FRM uses a
Teflon filter to obtain PM2 5 mass and trace elements (Na - Pb). The second FRM uses a pre-baked
quartz-fiber filter that is split in half with one half being analyzed for OC and EC and the other half for
sulfate, nitrate, and ammonium ions. As mentioned above, the FRM is the reference method for PM2 5
mass and should provide a suitable reference for non-volatile species, such as sulfate and many of the
trace elements determined by XRF. The semi-volatile species, such as ammonium nitrate and some of
the organic species are collected with less bias by the VAPS sampler and in Rubidoux by the PTEP
sampler. Thus, the VAPS will provide a reference for semi-volatile species.
SCAQMD PTEMP Sampler
The PTEP sampler, like the Andersen sampler is based on the design of the Caltech Gray Box sampler
(Solomon 1989). Air is drawn through an inlet and a PM2 5 cyclone to obtain the desired cut-point. Air
is split into several sample streams, with a fraction of the air passing through denuders and into filter
packs or directly into filter packs. The PTEP sampler is schematically illustrated in Figure 1-7 and
described below. Additional details of the design and the network this sampler is employed can be
found in SCAQMD (1996).
As shown in Figure 1-7, the PTEP sampler has four channels and ten sampling lines for measurement of
PM10 and PM2 5 mass, and chemical and gaseous components. : PM2 5 is sampled in Channels n (Lines
Part I, Page 12
-------
3, 4 & 5) and IE (Lines 6 and 7). A Teflon-coated AfflL Cyclone (John and Reischi, 1980) is used to
obtain a nominal PM2 5 size fraction in Channel U. Three sampling lines are located below Channel U
for the measurement of aerosol nitrate and ammonium and their gas phase counter parts, nitric acid and
ammonia. Ammonia and nitric acid losses were minimized by the use of a short Teflon line into the
cyclone and coating the cyclone internally with Teflon. Channel U contains two stainless steel denuders
used for ammonia and nitric acid. Line 3 feeds into the ammonia denuder columnar box consisting of
strips of citric acid impregnated quartz filters that are efficient scavengers of ammonia gas (Stevens et al.,
1985). Due to the high ammonia levels sometimes found in the Los Angeles Basin, these ammonia
denuders were changed every month. An acid impregnated filter in a Gelman aluminum filter holder is
connected to the ammonia denuder. Line 4 feeds into the nitric acid denuder, which consists of a
stainless steel columnar box with anodized aluminum plates. A dual filter pack, quartz followed by
nylon, is mounted below this denuder. The quartz filter collects the paniculate nitrate and the nylon filter
is used to quantitatively trap any gaseous nitric acid that has penetrated through the denuder and
volatilized from the front quartz filter.
Line 5 consists of an all-Teflon filter pack (Savillex) with three stages. A quartz filter followed by a
Nylasorb (Gelman) and then a citric acid impregnated quartz filter are all mounted in series in line 5.
This line collects PM2 5, nitric acid, and ammonia gas, and is used as the non-denuded leg of the denuder
system. This line measures total nitrate and ammonium (gas and particle). The difference between this
line and lines 3 and 4 provide an estimate of gas phase nitric acid and ammonia by the denuder
difference method (Solomon et al., 1988).
Channel UJ (Lines 6 & 7): PM2.5 mass, organic and elemental carbon, and inorganic trace metals are
obtained from Channel UJ (Lines 6 & 7). PM2 5 size fractionation is obtained using a stainless steel
Sensydyne model 240 cyclone (Lippmann and Chan, 1970). A stainless steel bowl with stainless steel
mesh protects the inlet of the cyclone. Because of the high-volume flow characteristics (110 Lpm) of
the cyclone, a stilling or mixing chamber coated with Teflon is used prior to the splitting of the flow into
two lines (Fitz et al., 1989). Since the carbon analysis and trace elemental analysis utilizes techniques
that are precision-sensitive to the homogeneity of particle deposits on the filter, flow homogenizers were
used. The homogenizers are 30 cm long stainless steel tubes with internal diameters of 4.5 cm. Line 6
samples PM2.5 carbon while line 7 collects aerosol samples for the determination of mass and inorganic
trace element concentrations.
Chemical Speciation and Chemical Analysis
The chemical components of PM2.5 measured in this study are the same as those specified for the
National PM2 5 Chemical Speciation Network (EPA, 1998) and recommended by the expert peer-
review panel (Koutrakis, 1998). Chemical characterization includes mass, sulfate, nitrate, and
ammonium ions, elements (Na through Pb), organic carbon (OC) and elemental carbon (EC).
Appropriate filter media were used to allow for chemical analysis by routine methods as described in
EPA (1998), Koutrakis (1998), Chow (1995), and recommended by the vendors. As described
Part I, Page 13
-------
above, these media combined with appropriately coated diffusion denuders should minimize sampling
artifacts. The field study described here, however, will not involve comparisons to independent certified
methods that would allow for an estimate of accuracy. However, comparison to the historical samplers
(IMPROVE, VAPS, and FRM) provide for a comparison to samplers that have been operated under a
number of conditions. Differences in nitrate losses and possibly losses (negative artifact) or gains
(positive artifacts) of SVOCs can be initially evaluated as a result of this intercomparison.
Chemical analysis of aerosol on the collected filters is by routine methods as described in EPA (1998)
and Chow (1995). Figures 1-1 through 1-7 illustrate the experimental design for each sampler and show
which analytes were determined on which filters. A tabular summary of the species measured by each
sampler is given in Table 1-1. Appendix A summarizes the chemical analysis methods. Detailed
standard operating procedures (SOPs) have been prepared (RTI, 1999), and are listed in Table 1-2,
and can be found in Appendix B. These SOPs were followed for all analyses. In general, PM2 5 mass
is determined gravimetrically on Teflon filters. Elements (Na - Pb) are determined on the same filter as
PM2.5 mass by energy dispersive X-ray fluorescence (XRF). Anions (sulfate and nitrate), and
ammonium ion are determined from aerosol collected on several different filter media (Teflon, quartz-
fiber, or nylon). Each filter is extracted in water or a carbonate/bicarbonate buffer solution (1C eluent
for anions if only anions are being determined from the filter) and quantified in the extract using ion
chromatography. The nylon filter is analyzed only for nitrate, except for the IMPROVE sampler, where
nitrate, sulfate, and ammonium ion concentrations are determined from the sampler collected on the
nylon filter. Organic and elemental carbon (OC/EC) are determined on the quartz-fiber filters using
thermal-optical reflectance (TOR).
The following provides a brief description of the chemical analysis methods used in this study by species.
PM25Mass
PM2.5 mass, is determined gravimetrically on Teflon filters using a microbalance (see Appendix B)
following procedures outlined in the Federal Register for PM2 5 FRM mass measurements in ambient air.
Prior to sampling filters are equilibrated for 30 days at the specified temperature (T) and relative
humidity (RH), followed by a one week equilibration period in the temperature range from 20-25 C and
an RH in the range of 20-30%. Filters are weighed, sealed in petri dishes, and stored until they are sent
out to the field. During storage and transport, filters are maintained at < 4 C. Prior to weighing sampled
filters, they are again equilibrated at the same T and RH as they were for pre-weights. PM2 5 mass is
determined by the difference between the post- and pre-weighed filters. Atmospheric concentrations
are obtained by dividing the mass per filter by the volume of air sampled.
Part I, Page 14
-------
Trace Elements (Na-Pb)
Teflon filters analyzed for mass also are analyzed for trace elements from Na to Pb by atmospheric
pressure X-ray fluorescence (see Appendix B). In this method, the filter is open to the atmosphere, but
surrounded by a sheath of He gas. Secondary x-rays are used primarily as the excitation source
resulting in virtually no heating of the filter or collected sample. Quantification of XRF spectra are
obtained by comparing to standards of known concentration as described in the SOP. Atmospheric
concentrations are obtained by dividing the loadings per filter, usually in nanograms (ng) by the volume
of air sampled.
Sulfate, Nitrate, and Ammonium Ions
Sulfate, nitrate, and ammonium ions are determined in filter extracts from Teflon or quartz-fiber filters by
ion chromatography (1C). Filters used for ion analysis are identified Figures 1-1 to 1-3,1-5, and 1-6
(also see SOPs in Appendix B, and). For the IMPROVE sampler, anions (i.e., sulfate and nitrate) and
ammonium ion are determined from the nylon filters used in that sampler. Volatilized nitrate is
determined directly in the extract from the nylon filters located behind the Teflon filter used for mass and
XRF analysis in the URG and VAPS samplers. Anions are determined from a section of the quartz-
fiber filter in the URG 450, VAPS, and FRM samplers. These are being compared to anions
determined from extracts of Teflon filters used in the MetOne and Andersen samplers. This helps to
ensure that nitrate and sulfate collected on the quartz-fiber filter can be used for anion and cation
determinations if nitrate and ammonium are lost from the Teflon filter during XRF analysis. Standards
are run according to the procedures outlined in the SOP (Appendix B) and used to quantify the
concentrations of the anions and cations in the extract. Atmospheric concentrations are obtained by
dividing the loadings per filter by the volume of air sampled.
Organic and Elemental Carbon
Organic and elemental carbon collected on pre-baked quartz-fiber filters are determined by the
thermal/optical reflectance method (TOR) (see SOP in Appendix B). In this method, a portion of the
quartz-fiber filter is heated first in He to remove organic material and then in He with 2% oxygen to
remove elemental carbon. The volatilized carbon is converted to CO and then to methane, which is
detected by an flame ionization detector. Optical reflectance of the sample is monitored to correct the
TOR OC/EC analysis for possible charring during the highest temperature step in 100% He.
Concentrations are determined by comparison to standards of known amounts. Atmospheric
concentrations are obtained based on the amount of filter used and the volume of air sampled.
Special Studies: XRF and Thermal Analysis for OC/EC
Loss of Nitrate During XRF Analysis
Atmospheric pressure XRF, with secondary ion excitation will likely minimize loss of volatile species
e.g., nitrate and condensed SVOCs, during XRF analysis relative to vacuum XRF, thus, these filters
might be able to be analyzed for nitrate, sulfate, and ammonium at a later date, or archived for other uses
(e.g., QC check on final mass). However, most analytical laboratories use vacuum XRF and both
Part I, Page 15
-------
primary and secondary excitation procedures, all of which would likely result in a significant loss of
volatile species from the filter and limit it use for other analyses. Therefore, determining the effect of
vacuum XRF on volatile species is important for two reasons. First, the URG MASS sampler, as
specified from the manufacture uses the same filter to obtain mass, trace elements by XRF, and ions
(sulfate and nitrate). If volatile species, i.e., nitrate and ammonium, are lost during vacuum XRF, then
subsequent determinations of those species will be biased by the amount lost. Secondly, the FRM
sampler, in the compliance network is being used only for mass determination. If vacuum XRF does not
bias the nitrate, ammonium, and organic carbon determinations, then these filters can be archived and, if
needed re-weighed at a later time, or analyzed for sulfate, nitrate, and ammonium to provide a more
detailed chemical composition of the collected aerosol from the FRM sampler. One alternative would
be requiring atmospheric pressure XRF analysis of all Teflon filters, assuming it does not drive off semi-
volatile species in the analysis process. The other alternative would be not using the filters for further
chemical analysis or mass determinations. To examine the potential loss of volatile species from the
collected Teflon filter during vacuum XRF (see SOPs and Appendix B), 40 filters are analyzed by
vacuum XRF, after atmospheric XRF analysis, and then analyzed for sulfate, nitrate, and ammonium by
extraction and 1C analysis as described below. These ions are compared to their concentrations
collected by the same sampler and by collocated samplers.
As just described, analysis of Teflon filters by atmospheric pressure XRF also may result in the loss of
volatile species due to the phase equilibrium shifting to the gas phase as He passes over the sample. Teflon
filters previously analyzed by atmospheric pressure XRF are being analyzed for sulfate and nitrate
concentrations. These are being compared to nitrate and sulfate concentrations obtained by the same
sampler and by collocated samplers.
TOR vs TOT Analysis for OC and EC
Two methods have been widely used for bulk analysis of OC and EC on quartz-fiber filters; thermal
optical reflectance (TOR) and thermal optical transmittance (TOT). TOT is the MOSH 5040 method
that is being used by the national laboratories for OC/EC determinations. At the names imply TOR
employs reflectance to help adjust the OC/EC analysis for charring during the thermal evolution of OC,
while TOT uses transmittance to accomplish the same objective. There are other differences between
the methods. For example, the temperature ramps are different and the maximum temperature used for
obtaining OC and EC are different. For these reasons, investigators have observed differences between
the two methods for OC and EC determinations. Therefore, in this special study, a series of filters will
be analyzed by both methods, including standards of known concentrations.
Part I, Page 16
-------
Splitting Filters for Multiple Analyses
As described above, some of the filters are used for more then one analysis or the analytical method
itself requires only a section of the filter. For example, Teflon filters for anion and cation analysis are
split in half so that each half can be extracted using the appropriate solution. Quartz-fiber filters are
sectioned and only a small section (about 1 to 1.5 cm2) is used for analysis. As well, in the case of the
URG chemical speciation sampler, the VAPS, and the FRM the filter is split in half, with one half used
for ion analysis and the other for TOR analysis. The SOP for sample sectioning is found in Appendix B.
Filter and Denuder Preparation
Several of the filters require pretreatment to lower blank levels and diffusion denuders need to be coated
with a reactive substance to allow for efficient removal of specific gas phase species. For example,
Teflon filters are equilibrated at specified T and RH as described earlier, quartz-fiber filters used for
OC/EC analysis are baked for several hours (Chow, 1995) at 900 C to lower blank levels to 1 ug C
cm"2 of filter material, while nylon filters must be cleaned before use to ensure consistently low blank
levels if acceptance testing indicates variable blank levels or contamination greater then 1 ug NO3" per
filter. Nylon filters are cleaned by soaking in a NO2CO3 solution followed by a thorough rinse using DI
water. Table 1-3 lists the filters by sampler type and indicates general filter preparation needs.
Denuders must be coated initially, cleaned or refurbished, and recoated as needed. As described in
Table 1-3, MgO denuders only require the initial coating as they are believed to have sufficient capacity
for the 20 day study and are not extracted for chemical analysis. The Na2CO3 coated denuder, requires
cleaning and re-coating after every use, or at least after every three uses. In the VAPS, this denuder
was extracted after each sampling period and analyzed for FDSTO3. The XAD denuders, must be
refurbished after every sampling period, and re-coated after every tenth sampling period.
Sampling Locations and Rationale
Sampling locations are identified based upon the following criteria. First, the statistical design requires
testing each sampler under different chemical atmospheres and varying environmental conditions.
Secondly, locations are needed where PM sampling is ongoing with preference given to locations where
PM chemical speciation sampling is occurring at the time of the study. Finally, sufficient infrastructure
needs to be available with local support to assist with filter changing and sampler operations. Four
locations were chosen that meet these criteria: Philadelphia, PA, Phoenix, AZ, Rubidoux, CA, and
Research Triangle Park, NC. Philadelphia represents a typical east coast situation where high sulfate
and organic material are present in the aerosol, but nitrate is typically low (Pace, 1998). Phoenix
represents an area with the potential for high crustal material, which typically is the dominant material
above 2.5 //m, but with a tail in the less than 2.5 //m size range (Pace, 1998; Solomon et al., 1986).
Phoenix also has a strong nitrate and organic material component. Rubidoux represents an area with
very high nitrate, moderate organic material, low sulfate, and relatively low crustal material (Solomon et
al., 1989; SCAQMD, 1996). The RTF site is to allow for a more thorough evaluation of sampler
performance and provide a site where PM levels are near the lower limit of detection for the species
measured by the samplers being tested.
Part I, Page 17
-------
Of the four sites, Rubidoux is the prime site because it provides the most stringent test of the samplers
for examining collection efficiencies of nitrate and semi-volatile condensed organic compounds, has a full
complement of PM, gaseous, and meteorological sampling equipment, including full chemical speciation
using the SCAQMD's PTEP sampler, and the characteristics of the air at Rubidoux have been well
characterized by several studies over the last decade (e.g., Solomon et al., 1989). Two sets of samplers
are collocated at Rubidoux to obtain precision estimates. Table 1-4 outlines the existing sampler
equipment located at Rubidoux, CA. Table 1-5 lists the existing equipment located at Phoenix, AZ.
These two sites are well equipped to support this study with both additional PM measurements,
meteorological measurements (the most important of which are relative humidity and temperature), and
supporting gas phase measurements, such as ozone, nitrogen oxides, and sulfur oxides. PM10 sampler
meteorological data are collected at the Philadelphia site. At RTF, samplers were installed at the new
NERL sampling platform; however, supporting data are not available at this site.
These sites represent Phase I of this program to evaluate the chemical speciation samplers for use in the
National Chemical Speciation Network. We recognize however, that the study is limited in scope, not
only geographically, but seasonally. Conditions that were not represented are the high sulfate season on
the east coast and areas with either high biogenic organic material or high wood smoke emissions. The
highest season for crustal material in Phoenix is during the summer, thus, the samplers were not
challenged with the highest concentrations of crustal material. The samplers were not evaluated for
operations in either very cold or very hot conditions, nor under conditions of severe weather. As
discussed earlier, these other conditions will be tested during Phases n and m of this evaluation.
Figures 1-8 through 1-11 show the samplers at each site.
Program Schedule
Overall Program Schedule
Table 1-6 summarizes the overall schedule for this study. The schedule was driven by three criteria: 1) a
draft report was due to OAQPS by the middle of March, 1998, 2) 20 sample sets would be collected
at each site to help ensure that a sufficient number of samples would be collected simultaneously on all
samplers to meet the statistical design objectives, and 3) the study could not begin until all five sets of
the three chemical speciation samplers and the IMPROVE sampler were delivered to ORD (the original
delivery date was August 15, 1998, and only MetOne met that schedule). The latter included delivery
of a sufficient number of spare parts, extra filter holders, and denuders to allow for every-other-day
sampling. These three criteria uniquely define the schedule for the program and dictated that sampling
must be performed simultaneously at the four locations chosen for this study. Sampling was to begin
around September 1, 1998. However, all samplers and spare parts were not delivered until nearly the
end of November 1998 (Andersen was the last sampler to arrive), which with seasonal holidays delayed
the start of sampling until nearly the middle of January, 1999. The due date for submission of the draft
final report to OAQPS was then re-scheduled for the end of June 1999.
Sampling Schedule
Part I, Page 18
-------
Sampling was conducted in January and February of 1999. Samplers were operated for 24-hr sampling
periods every other day, except at Rubidoux. Sampling at Rubidoux was every third day to sample
simultaneously with the PTEP sampler.
To meet the every other day sampling schedule, filters and holders were shipped overnight to the
contractor immediately after collection according to the sampling schedule illustrated in Table 1-7.
Filters, the XAD denuder in the VAPS, and all Na2CO3 denuders were shipped by overnight mail.
Three full sets of filter holders and denuders were available for this purpose, which required continuous
shipping of filters to and from the laboratory. This turned out to be a rigorous schedule to maintain with
site operators and laboratory personnel working 7 days per week. Delays only occurred when the
overnight service failed to delivery the filters as expected.
Preliminary Evaluation and Training
Once samplers were received at EPA in Research Triangle Park, they were configured for use in this
field study and underwent a quick shakedown. One representative from each manufacturer was
available to ensure proper assembly of their samplers and to train contractor personnel. The shakedown
included, for example, running each sampler for two-12 to 24 hour periods to ensure that samplers were
turning on and off as expected, flow rate checks, flow control checks, leak checks, etc. Standard
operating procedures in the form of operating manuals were made available from the manufacturers
when the samplers were delivered to RTF. The contractor prepared condensed installation and
operating SOPs for easy use by field and laboratory operators. A list of SOPs is given in Table 1-2 and
Appendix B.
Site Installation and Decommissioning
All sites were equipped with one of each of the three chemical speciation samplers, one IMPROVE,
one VAPS, and one set of two FRM samplers. Rubidoux had a collocated second set of all samplers
except the VAPS sampler. EPA contractor personnel installed the samplers at all sites, trained site
operators at Rubidoux and Phoenix, and conducted initial systems and performance audits prior to the
start of the study. At the end of the study, EPA contractors performed a final audit and then
decommissioned the site and returned the equipment to EPA at Research Triangle Park, NC. A mid-
study audit was also conducted.
Power and platforms were provided by the states or local districts. At Rubidoux and Phoenix, local
state or local district operators normally working at those sites operated the samplers. At Philadelphia
and RTF, the sites were operated by EPA contractors.
Part I, Page 19
-------
Sampler Operation and Filter Shipping and Storage
Samplers were operated according to the SOPs as given for each sampler and referenced Table 1-2
and in Appendix B. Three full sets of filter holders and denuders were required for each sampler to
meet the rigorous schedule described above. The three sets of filter holders and denuders also allowed
time for these items to be shipped to the contractor for exchange, thus, removing site-to-site variability
for filter changing. Filters were shipped in coolers at reduced temperature with a max/min thermometer.
Each cooler contained sufficient blue ice, sealed in plastic bags or some other closed system, to keep the
filters cool for 24-hours. At the field site, the filters were stored before and after sampling at reduced
temperatures in a refrigerator, especially after sample collection. Except for equilibration of filters for
mass determination, filters with collected aerosol were stored in sealed petri dishes at reduced
temperatures at or below freezing.
Quality Assurance/Quality Control (QA/QC)
A thorough QA/QC effort was implemented for this study to ensure the quality of the data. These
efforts included audits of the samplers prior to, during, and after the field program, even though the study
was only 2 months in duration. Laboratory quality assurance and control followed already established
procedures and included external audits of the analyzers in conjunction with ongoing studies.
Field QA/QC
Quality assurance consisted of system and performance audits at the beginning, middle, and end of the
study. These audits were conducted by contractor staff. Initial QA audits of the samplers were
performed prior to the first sample if possible, or as quickly after the initiation of the program as
possible. A second set of system and performance audits was performed in the middle of the program,
between the 8th and 10th sampling periods, depending on the site. A third set of audits was performed
after the last sample, but before decommissioning of the samplers by the EPA contractor. Initial system
audits checked installation of the samplers, site setting, site operations and operators, and custody
management. Performance audits checked flow rates, check for vacuum leaks, and other performance
characteristics of the samplers. Quality control consisted of checking flow rates before and after each
sample on every filter unless automatically logged by the sampler, maintaining appropriate operations
logs, checking filter holders and denuders prior to and after each sampling period, and other items as
specified on the data log sheets.
Part I, Page 20
-------
Laboratory QA/QC
Laboratory systems and performance audits were conducted for each species according to schedules
already in place in the laboratories. Results will be reported to EPA staff in the contractors final report,
which has not been received to date.
STATISTICAL ANALYSIS
Questions or Statistical Hypotheses Being Tested
As described above in the introduction, the primary objective of this study is to determine if there are
differences among the three chemical speciation monitors for their estimation of the concentrations of the
chemical components of PM2.5 mass. Comparisons also are made to two historical samplers
(IMPROVE and VAPS) and to the FRM using these samplers as a relative reference. Reference
standards do not exist for any of the species, only for mass and that is defined by the FRM; therefore,
this study is testing the equivalency of the samplers to each other, rather then comparing the samplers to
a known value that provides an estimate of their accuracy.
There are three major questions to be addressed by this intercomparison study.
1. How do the concentrations of the measured chemical components as determined by the FRM
compare to those measured by the chemical speciation monitors?
2. Can FRM mass be reconstructed within expected uncertainties from the sum of the chemical
components as measured by the chemical speciation monitors?
3. How well do the concentrations of the measured chemical components of PM2 5 from the
various speciation samplers agree?
4. What are the causes of the differences, if they exist. Some are expected due to the different
methods employed by the different samplers? and
5. What is the potential effect of different analytical methods on measured concentration of the
chemical components of PM2.5. These include the effect of vacuum XRF vs. atmospheric
pressure XRF on nitrate concentrations measured on Teflon filters and the effect of thermal
optical reflectance vs. thermal optical transmittance on OC/EC concentrations.
The following hypotheses are predicated on the assumption that the collection efficiency (slope of the
efficiency curve and cutpoint) for the samplers in this study are essentially the same. Also, it is important
to establish the precision of the instruments and this was done by collocating samples at Rubidoux.
While this provides only a limited assessment of the precision (one site, one time of the year, one set of
atmospheric conditions), it does provide a first cut estimate of the precision needed for the statistical
analyses performed here. If for example, the precision is estimated at 50%, then determining differences
Part I, Page 21
-------
among samplers is not as informative as if the precision were say 10-15%. As a benchmark, the
coefficient of variation for the differences in mass concentrations from collocated FRM instruments is
required to be less than 10%, according to 40 CFR Part 58, Appendix A. Depending on the species,
based on the authors' prior experiences, we anticipate a range of precision from less than 10% to about
30%.
In the following discussion, the phrase "speciation samplers" includes the Andersen RAAS, the URG
MASS, the Met-One SASS, the IMPROVE, and the VAPS.
The following hypotheses are related to the first two questions, how do the samplers compare to the
FRM and can FRM mass be reconstructed from the species measured by the speciation samplers. For
each of these hypotheses, the concentrations, on a species-by-species basis, from each of the speciation
samplers is compared to the respective concentration from the FRM. The concentrations from the
speciation samplers are not compared to each other. The specific hypotheses are:
1. PM2 5 mass concentrations measured on the FRM by weighing a Teflon filter is compared to
PM2 5 mass as measured by weighing filters collected by the Teflon filters in the chemical
speciation monitors. The hypothesis is that the mass concentration from each of the speciation
samplers is not statistically different from the mass on the FRM filter. This is the starting point.
The next eight hypotheses delve further into understanding why the mass concentrations do or
do not compare favorably.
2. PM2.5 trace elements or groups of trace elements determined on samples collected by the Teflon
filter in the FRM as determined by XRF are compared to trace elements determined on samples
collected on Teflon filters collected by the speciation samplers. The hypothesis is that there is no
statistical difference between the concentrations from each of the speciation samplers to that
determined from samples collected by the FRM. These concentrations should be comparable
since these species are stable.
3. PM2 5 sulfate on the FRM quartz filter is compared to sulfate on the Teflon (RAAS, SASS),
quartz-fiber (MASS, VAPS), or nylon (IMPROVE) filters in the speciation samplers. The
hypothesis is that there is no difference between the concentrations from each of the speciation
samplers to that determined from samples collected by the FRM. These concentrations should
be comparable since sulfate is a stable species.
4. PM2 5 ammonium determined from samples collected by the quartz-fiber filter on the FRM is
compared to ammonium ion determined from samples collected on the Teflon (RAAS, SASS),
quartz (MASS, VAPS), or nylon (IMPROVE) filters in the speciation samplers. The hypothesis
is that there is no difference between the concentrations from each of the speciation samplers to
Part I, Page 22
-------
that determined from samples collected by the FRM. These concentrations may not be
comparable since ammonium is volatile when in the form of ammonium nitrate.
5. PM2 5 nitrate determined from samples collected on the quartz-fiber filter in the FRM is
compared to nitrate determined from samples collected on the Teflon (RAAS, SASS), quartz-
fiber (MASS, VAPS), or nylon (IMPROVE) filters in the speciation samplers. The hypotheses
is that there is no difference between the concentrations from each of the speciation samplers to
that determined from samples collected by the FRM. These concentrations should be
comparable since the collection of nitrate on these filters should all be biased by the loss of
nitrate due to volatilization, with the exception of the IMPROVE.
6. PM2 5 nitrate on the FRM quartz filter is compared to nitrate on the nylon (RAAS, SASS,
IMPROVE), or quartz+nylon (MASS, VAPS) filters in the speciation samplers. The hypothesis
is that the nitrate from each of the speciation samplers is greater than or equal to the nitrate on
the FRM Teflon filter, due to negative artifacts in the FRM. This addresses questions about the
amount of nitrate volatilized from the FRM Teflon filter.
7. PM2 5 elemental carbon (EC) determined on samples collected on quartz-fiber filters by the
FRM is compared to EC determined on quartz-fiber filters collected by the chemical speciation
samplers. The hypothesis is that there is no difference between the concentrations from each of
the speciation samplers to that from the FRM. These concentrations should be comparable
since elemental carbon is stable.
8. PM2 5 organic carbon (OC) determined on quartz-fiber filters collected by the FRM is
compared to OC determined on quartz fiber (MASS, RAAS, SASS, IMPROVE) filters in the
speciation samplers. The hypothesis is that there is no difference between the concentrations
from each of the speciation samplers to that determined from samples collected by the FRM.
These concentrations should be comparable since as designed in this study, all the speciation
samplers potentially suffer from negative or positive artifacts.
9. PM2.5 OC determined on samples collected on the quartz-fiber filter in the FRM is compared to
OC determined on samples collected on the quartz-fiber filter from the VAPS. The hypothesis
is that OC collected by the FRM is greater then the OC collected by the VAPS denuded
channel if there are positive artifacts, or less than the VAPS denuded channel if there are
negative artifacts. No difference would be inconclusive.
The following hypotheses compare the concentrations of the chemical components determined on
samples collected by the chemical speciation samplers to address why there are differences among
measured concentrations, if they exist. For each of these hypotheses, only the concentrations from the
speciation samplers are compared.
Part I, Page 23
-------
10. PM2 5 mass and chemical composition as determined according to the manufacturer's guidelines
are compared among the speciation samplers. The hypothesis is that all species of interest are
comparable among the chemical speciation samplers when concentrations are determined
according to manufacturer's guidelines.
11. PM2 5 mass concentrations determined by gravimetric analysis using Teflon filters are compared
among the samplers. The hypothesis is that there is no difference in these masses.
12. PM2 5 trace element concentrations (individually or in groups) determined by XRF from samples
collected by Teflon filters are compared among the samplers. The hypothesis is that there is no
difference in these concentrations as these elements are stable during sampling and analysis.
13. PM2 5 nitrate concentrations determined from samples collected using nylon (RAAS, SASS,
IMPROVE) or Teflon+nylon (MASS, VAPS) filters are compared among the samplers. The
hypothesis is that there is no difference in these concentrations.
14. PM2 5 sulfate determined from samples collected using Teflon (RAAS, SASS), quartz-fiber
(MASS, VAPS), or nylon (IMPROVE) filters are compared among the samplers. The
hypothesis is that there is no difference in sulfate concentrations as sulfate is stable during
sampling and analysis. We also will look at sulfate estimated from XRF sulfur (S times 3 is
approximately equal to sulfate) to see how well XRF sulfur estimates sulfur determined by 1C
from both Teflon or quartz filters.
15. PM2 5 organic and elemental carbon determined from samples collected by quartz-fiber filters
where no denuder is used is compared among the speciation samplers. Since the VAPS used a
denuder, it will not be included in this hypothesis. The hypothesis is that the concentrations of
OC and EC are the same provided no denuder is employed.
16. PM2 5 organic and elemental carbon determined from samples collected by quartz-fiber filters,
including speciation samplers where a denuder is used is compared among the samplers. The
hypothesis is that samplers that use denuders (VAPS) will have lower OC concentrations since
the potential for positive artifact due to organic vapors has been minimized. Nothing can be said
about negative artifacts.
17. PM2.5 ammonium determined from samples collected by the Teflon (RAAS, SASS), quartz-
fiber (MASS, VAPS), or nylon (IMPROVE) filters are compared among the samplers. The
hypothesis is that the concentrations of ammonium are the same for all speciation samplers as all
potentially suffer from negative artifacts.
Part I, Page 24
-------
18. Ammonium nitrate equilibrium is a function of temperature and relative humidity and thus, nitrate
concentrations measured on reactive filters behind a base coated denuder are compared as a
function of temperature and RH to determine if these factors bias sample collection. The
hypothesis is that there is no difference in the samplers as a function of temperature or RH.
19. Crustal related elements are typically associated with particles greater than 2.5 um AD,
however, differences in the inlet the efficiency (slope and cutpoint) may result in different
concentrations of crustal related material being measured by these samplers. The hypotheses is
that the crustal related material as determined from summing the oxides of Fe, Ca, and Si
(Solomon et al, 1989) are not statistically different among the samplers. If differences are
observed they are related to the coarse particle mass as measured by the VAPS or by other
collocated PM10 monitors. Wind speed also is a variable of interest in this analysis.
The above hypotheses examine the first four major questions stated above. Statistical differences are
examined with in sites using primarily the paired t-test at alpha = 0.05. Added variability due to site-to-
site variations in the chemical composition limit the paired t-test to individual sites, as missing data are
minimized. The data also are examined by looking at means, time series analysis, regression analysis,
difference analysis on a species by species basis relative to the FRM.
Several tests are being conducted to examine the affect of different analytical methods on the
concentration of the species measured. These include the effect of vacuum XRF or atmospheric XRF
on nitrate concentrations measured on the filter after XRF analysis and the use of TOR vs. TOT for
OC/EC analysis as stated in the fifth question presented at the beginning of this section.
a. Examine the effect of vacuum vs. atmospheric pressure XRF on nitrate concentrations
measured on the Teflon filter after XRF analysis. The hypothesis is two fold. First,
there will be a greater loss of nitrate from Teflon filters after vacuum XRF than after
atmospheric pressure XRF. Secondly, losses that occur during atmospheric pressure
are minimized, such that nitrate measured after atmospheric pressure XRF is not
different than nitrate measured on a Teflon filter that has not undergone XRF analysis. It
is important to understand the magnitude of the loss of nitrate from Teflon filters after
vacuum or atmospheric XRF because the proposed design of the URG sampler has ion
analysis occurring after XRF analysis of the only Teflon filter in the system. There is also
great potential for the FRM Teflon filter to be used for XRF analysis and ion analysis
after mass determination to obtain additional speciation data at FRM sites.
b. Examine difference between OC and EC as measured by TOR and TOT. The
hypothesis is that these two methods are statistically not different from each other for
determining OC and EC from samples collected on quartz-fiber filters. However,
differences have been noted, especially with samples collected in areas with high wood
smoke emissions.
Part I, Page 25
-------
Finally, the samplers will be evaluated in terms of their operational performance or practicality of use in
the field, that is, reliability, ruggedness, ease of use, and maintenance requirements. Field operators
were asked to maintain detailed logs of their operational performance and they completed an extensive
survey after the study on these issues. As well cost data are presented for the samplers and spare parts.
Part I, Page 26
-------
Data Reporting Format
The concentrations for each analyte will be compiled as a flat ASCII file organized as follows:
1} Format
ASCII text file with defined columns and rows.
2}
4)
Site IP's:
1
2
3
4
Rubidoux
Phoenix
Philadelphia
Research Triangle Park
Sampler IP's:
FRM - Teflon
FRM - Quartz
VAPS
Met One-SASS
URG-MASS
Andersen RAAS
Improve
Analytes (mass/m3)
C-l
C-2
C-3
C-4
C-5
C-6
C-7
C-8.Cn
FRM-T
FRM-Q
VAPS
MET
URG
AND
IMP
PM2.5
SO4
NO3
NH4
OC
EC
S-XRF
Individual metals - XRF
5) Reporting Duplicate Sampler Results (Rubidoux^
Identify as "Repeat 1" and "Repeat 2".
Part I, Page 27
-------
6} Limit of Detection (LOP) or MDL for Total Method (Sampling Plus Analysis^
The LOD for mass is determined based on Federal Reference Method procedure (40 CFR,
Part 50, Appendix L). The LOD sulfate, nitrate, and ammonium based on 3 time the noise in
the baseline noise in the chromatogram since field blanks were non-detectable, while for OC
and EC it is based on three times the standard deviation of the field blanks. XRF limits of
detection are based on propagating errors associated with the analytical method and flow rates.
In general, LOD values were within the expected ranges.
Part I, Page 28
-------
REFERENCES
Andersen. 1999 Operations Manual for the RAAS sampler. Andersen Instruments, GA.
Cahill, Thomas A. and Paul Wakabayashi. Compositional analysis of
size-segregated aerosol samples. Chapter in the ACS book Measurement
Challenges in Atmospheric Chemistry. Leonard Newman, Editor. Chapter 7,
Pp. 211-228(1993).
Chameides, W.L., Lindsay, R.W., Richardson, 1, Kiang, C.S., 1988. The role of biogenic
hydrocarbons in urban photochemical smog: Atlanta as a case study. Science 241, 1473-1475.
CFR 1997. National Primary and Secondary Ambient Air Quality Standards for Particulate Matter.
Code Federal Regulations, 40 CFR 50.7. Published by the Office of Federal Register, National
Archives and Records Administration.
Chow, J.C. 1995. Measurement Methods to Determine Compliance with Ambient Air Quality
standards for suspended particles. Journal of the Air and Waste Management Association, 45:320.
Cui, W., J. Machir, L. Lewis, DJ. Eatough, and N.L. Eatough (1997). Fine Parti culate Organic
Material at Meadview During the Project MOHAVE Summer Intensive Study, J. Air and Waste
Mgmt. Assoc., Volume 47.
Eatough, D.J., Tang, H.; Cui, W.; Machir, J. (1995). Determination of the Size Distribution and
Chemical Composition of Fine Particulate Semi-volatile Organic Material in Urban Environments
Using Diffusion Denuder Technology, Inhalation Toxicology, 1995, 7, 691-710.
Eldred, R.A., PJ. Feeny, and P.K. Wakabayashi. 1998a. The Major Components of PM2.5 at
Remote Sites Across the United States. Proceedings of an International Speciality Conference - PM2 5:
A Fine Particle Standard, Long Beach, CA, Jan. 28-30, A&WMA, Pittsburgh, PA, pp. 13-27.
Eldred, R.A., PJ. Feeny, P.K. Wakabayashi, J.C. Chow, E. Hardison. 1998b. Methodology for
Chemical Speciation Measurements in the IMPROVE Network. Proceedings of an International
Speciality Conference - PM2.5: A Fine Particle Standard, Long Beach, CA, Jan. 28-30, A&WMA,
Pittsburgh, PA, pp. 352-364.
EPA. 1998. Draft Supersites Conceptual Plan. Prepared for the Technical Subcommittee on Fine
Particle Monitoring of the Clean Air Scientific Advisory Committee by the Office of Air Quality Planning
Part I, Page 29
-------
and Standards and Office of Research and Development. US EPA, Research Triangle Park, NC.
Available at http://www.epa.gov/ttn/amtic/supsites.html.
EPA. 1999. Particulate Matter (PM2 5) Speciation Guidance Document, Monitoring and Quality
Assurance Group, EMAD, OAQPS, Research Triangle Park, NC. Available at
http://www.epa.gov/ttn/amtic/pmspec.html.
EPA. 1997a. 62 FR 38761. National Ambient Air Quality Standards For Paniculate Matter;
Availability of Supplemental Information and Request For Comments, Final Rule, U.S. Environmental
Protection Agency, 40 Code of Federal Regulations, Part 50, Federal Register (62 FR 38761) , July
18.
EPA 1997b. Revised Requirements For Designation of Reference and Equivalent Methods For PM2 5
and Ambient Air Quality Surveillance for Particulate Matter, Final Rule, U.S.
Environmental Protection Agency, 40 Code of Federal Regulations Parts 53 and 58, Federal Register
(62 FR 38763), July 18.
Hering, S.V., Lawson, D.R., Allegrini, I, Febo, A., Perrino, C., Possanzini, M., Sickles U, I.E., Anlauf,
K.G., Wiebe, A., Appel, B.R., John, W., Ondo, J., Wall, S., Braman, R.S., Sutton, R., Cass, G.R.,
Solomon, P.A., Eatough, D.J., Eatough, N.L., Ellis, E.G., Grosjean, D., Hicks, B.B., Womack, J.D.,
Horrocks, J., Knapp, K.T., Ellestad, T.G., Paur, R.J., Mitchell, W.J., Pleasant, M., Peake, E.,
MacLean, A., Pierson, W.R., Brachaczek, W., Schiff, H.I., Mackay, G.I., Spicer, C.W., Stedman,
D.H., Winer, A.M., Biermann, H.W., and Tuazon, E.C, 1988. The Nitric Acid Shootout: Field
Comparison of Measurement Methods. Atmospheric Environment 22, 1519-1539, Special Issue. Also
see other papers in this special issue.
Hering, S. and Cass, G. 1999. The Magnitude of Bias in the Measurement of PM2.5 Arising from
Volatilization of Particulate Nitrate from Teflon Filter. JAWMA, 49, 725-733.
Hering, S. 1999. Program Plan for the Atlanta Supersites Study. Prepared by Aerosol Dynamics,
Inc., Berkeley, CA.
Hidy, G.M., 1994. Atmospheric Sulfur and Nitrogen Oxides. Academic Press, San Diego, CA.
Koutrakis, P. 1998. Recommendations of the Expert Panel on the EPA Speciation Network. Report
to EPA, OAQPS. School of Public Health, Harvard University, Boston, MA
(http://www.epa.gov/ttn/amtic/pmspec.html). http://www.epa.gov/ttnamtil/whatsnew.html
Mathai, C.V., John G. Watson, Jr., C. Fred Rogers, Judith C. Chow, Ivar, Tombach, Judith O.
Zwicker, Thomas Cahill, Patrick Feeney, Robert Eldred, Marc Pitchford, and Peter K. Mueller. 1990.
Part I, Page 30
-------
Intercomparison of ambient aerosol samplers used in western visibility and air quality studies.
Environmental Science & Technology. Vol. 24, No. 7, pp. 1090-1099.
MetOne. 1999 Operations Manual for the SASS sampler. MetOne, Instruments. GA.
NARSTO. 1999. NARSTO Quality Systems Management Plan. Publication number 4785, Version
1.0, September 30. NARSTO Quality Systems Science Center, Oak Ridge National Laboratory, Oak
Ridge, TN. (Available from http://cdiac.esd.ornl.gov/programs/NARSTO/).
Pace, T. 1998. Composition of PM2 5 in Ambient Air. In Proceedings of an International Speciality
Conference - PM2 5: A Fine Particle Standard, Long Beach, CA, Jan. 28-30, A&WMA, Pittsburgh,
PA, pp. 3-12.
Pinto, J.P., R.K. Stevens, R.D. Willis, R. Kellogg, Y. Mamane, J. Novak, J. Santroch, I. Benes, J.
Lenicek, and V Bures. 1998. Chech Air Quality Monitoring and Receptor Modeling Study. Environ.
Sci. Technol. 32,843-854.
RTI. 1999. Quality Assurance Project Plan for Four-City PM2 5 Chemical
Speciation Sampler Evaluation Study, RTI Project Number 07263-030. RTI, RTF,
NC.
Russell, A.G. and G.R., Cass. 1986. Verification of a Mathematical Model for Aerosol Nitrate and
Nitric Acid Formation an its Use for control Measure Evaluation. Atmos. Environ., 20(2), 2,011-
2,025.
SCAQMD. 1996. Final 1997 Air Quality Management Plan. South Coast Air Quality Management
District, Diamond Bar, CA.
Schauer, J.J., W.F. Rogge, L.M. Hildemann, L.M. Mazurek, GR. Cass, and GR. Simoneit. 1996.
Source Apportionment of Airborne Particulate Matter Using Organic Compounds as Tracers. Atmos.
Environ. 30, 3,837-3,855.
Schauer, J.J. 1998. Source Contributions to Atmospheric Organic Compound Concentrations:
Emissions Measurements and Model Predictions. Chapter 10. Dissertation, California Institute of
Technology, Pasadena, CA.
Mukerjee, S., W.D. Ellenson, R.G Lewis, RK. Stevens, M.C. Somerville, and D.S. Shadwick. 1997.
An Environmental Scoping Study in the Lower Rio Grande Valley of Texas - 1. Comparative
Assessment of Air Sampling Methods. Environmental International 23(5), 61-628.
Part I, Page 31
-------
Solomon, P.A. and J.L. Moyers. 1986. A Chemical Characterization of Wintertime Haze in Phoenix,
Arizona. Atmospheric Environment, 20(1), 207-213.
Solomon, P.A., L. Salmon, T. Fall, and G.R. Cass. 1992. "The Spatial and Temporal Distribution of
Atmospheric Nitric Acid and Particulate Nitrate Concentrations in Los Angeles." Environ. Sci. Technol.
26(8):1594-1601.
Solomon, P.A., T. Fall, L. Salmon, G.R. Cass, H.A. Gray, and A. Davidson. 1989. Chemical
Characteristics of PM-10 Aerosols Collected in the Los Angeles Area. J. Air Pollut. Control Assn
39(2):154-163.
Solomon, P.A., W. Mitchell, D. Gemmill, M. Tolocka, J. Suggs, J. Homolya, J. Rice, and S. Eberly.
1999. Field Program Plan For The PM2 5 Chemical Speciation Sampler Evaluation Study. Prepared
by US EPA, ORD, NERL, Research Triangle Park, NC. Available at
http://www.epa.gov/ttn/amtic/pmspec.html
Sommerville, M.C., S. Mukerjee, D.L. Fox, and R.K. Stevens. 1994. Statistical Approaches in Wind
Sector Analyses for Assessing Local Source Impacts. Atmos. Environ. 28,3483-3493.
Stevens, R., J. Pinto, Y. Mamane, J. Ondov, M. Abdulraheem, N. Al-Majed, M. Sadek, W. Cofer,
W. Ellenson, and R. Kellogg. 1993. Chemical and Physical Properties of Emissions from Kuwaiti Oil
Fires. Water Sci. Techonl. 27, 223-233.
Teffera, S., S. Barbosa, and W.G Bope. 1996. Design, testing, and implementation of a multi-channel
sampler for use in an enhanced fine particulate monitoring program at the South Coast Air Basin. Paper
number 96-MP3.07. Proceedings Air & Waste Management Association, 89th Annual Meeting and
Exhibition, Nashville, TN, A&WMA, Pittsburgh, PA.
Part I, Page 32
-------
Table 1-1. Analyte Listing for Speciation Sampler Intercomparison
Analytical
Method
Analyte
(Concentration
Units)
Gravimetric
Ton
Chromatography
Thermal/Optical
Ref. Method
Atmospheric
Pressure XRF
PM2 5 Mass
SO4=, NO3; NFV,
OC/EC
S, Si, Ca, Mn, Fe,
Cu, Zn, Pb,
Groupings of
metals, etc.,
Sampling Systems
Met
One
X
X
X
X
MASS
X
X
X
X
RAAS
X
X
X
X
IMPROVE
X
X
X
X
FRM-T
X
-
-
X
FRM-Q
-
X
X
VAPS
X
X
X
X
Part I Tables, Page 1
-------
Table 1-2. List of Standard Operating Procedures for Field and Laboratory Efforts During the Chemical Speciation Monitor Evaluation
Study.
Field Related SOPs
SOP
Prepared By
Laboratory Related SOPs
Spiral Ambient Speciation Sampler MetOne
Reference Ambient Air Sampler Andersen
Mass Aerosol Speciation Sampler URG Corp.
IMPROVE Sampler UC Davis
Versatile Air Pollution Sample RTI
PM2.5 Federal Reference Method RTI
Filter Pack and Cassette Handling in the Field RTI
Denuder Handling and Shipping in the Field RTI
Receiving, Laboratory and Field RTI
Shipping at Reduced Temperatures RTI
SOP
Prepared By
Filter Pack and Cassette Handling in the Lab
Denuder Preparation - NajCC^
Denuder Extraction - Na2CO3
Denuder Preparation - MgO
Denuder Preparation - XAD
Denuder Regeneration - XAD
Denuder Extraction - XAD
PUF Cartridge Cleaning and Preparation
Filter Preparation - Nylasorb (nylon)
Filter Preparation - Teflon
Filter Preparation - Quartz-Fiber
Acceptance Testing of Filters
Filter Sectioning Procedures
Filter Extraction - Nylasorb
Filter Extraction ~ Teflon
Filter Extraction - Quartz-Fiber
Mass by Gravimetric Analysis1
Anions by Ion Chromatography
Cations by Ion Chromatography
XRF - Atmospheric Pressure
XRF - Under Vacuum
Thermal-Optical Reflectance (TOR)
Thermal-Optical Transmittance (TOT)
Data Validation
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
Mantech/NERL
DRI/Reno
DRI/Reno
Sunset Labs
RTI
Federal Register for FRM.
Part I Tables, Page
-------
Table 1-3. Experimental Design Including Filter and Denuder Preparation.
Sampler
Andersen
RAAS
Met-One
SASS
URGMASS
IMPROVE
VAPS
FRM
Filter Type
(Channel)
Teflon (2)
Teflon (3)
Nylon (4)
Quartz (1)
Teflon (1)
Teflon (2)
Nylon (3)
Quartz (4)
Quartz (5)
Teflon
Nylon
Quartz
Teflon
Nylon
Quartz
Teflon
Nylon
Quartz
Teflon
Quartz
Denuder
Type
MgO
MgO
Na2C03
Na2C03
Na2C03
Na2CO3
XAD-4
Analysis
(2) Mass and
elements1 (3) ions2
Fine particle NO3
OC/EC
(1) Mass and
elements (2) ions
Fine particle NO3"
OC/EC
OC/EC
Mass and
elements
Volatilized nitrate
OC/EC, ions
Mass and
elements
Volatilized nitrate
OC/EC
Mass and
elements Denuder
forHN03
Volatilized nitrate
Filter for OC/EC,
ions Denuder for
SVOCs
Mass and
elements
Ions and OC/EC
Filter and Denuder Preparation
Filters are equilibrated before weighing
Filters are wet with ethanol before extraction
Nylon filters are cleaned before use and extracted
in 1C eluent; MgO denuders are not changed or
cleaned during the study
Quartz filters are baked before use
Filters are equilibrated before weighing Filters are
wet with ethanol before extraction
Nylon filters are cleaned before use and extracted
in 1C eluent; MgO denuders are not changed or
cleaned during the study
Quartz filters are baked before use
Quartz filters are baked before use; Carbon
denuders need to be heat treated every 7 samples
Filters are equilibrated before weighing
Nylon filters are cleaned before use and are
extracted in 1C eluent; Na2CO3 denuders are
cleaned and re-coated daily
Quartz filters are baked before use; Filters are
split in half to allow for ion analysis
Filters are equilibrated before weighing
Nylon filters are cleaned before use and are
extracted in water since NH4+ also is analyzed in
the extract; Na2CO3 denuders are cleaned and re-
coated daily
Quartz filters are baked before use
Filters are equilibrated before weighing Na2CO3
denuders are extracted and re-coated daily
Nylon filters are cleaned before use. Nylon is the
backup filter in a Teflon/nylon filter pack
Quartz filters are baked before use; XAD-4
denuders must be rinsed daily, re-coated every
10 samples
Filters are equilibrated before weighing
Quartz filters are baked before use; Filters are
split in half to allow for ion analysis
1. Elements refer to those obtained from XRF analysis, potential Na to Pb.
2. Ions include nitrate, sulfate and ammonium.
Part I Tables, Page 3
-------
Table 1-4. Measurements Made at Rubidoux, CA in Conjunction with the Chemical Speciation
Sampler Evaluation Study.
Species
PM2.5 and PM10 Mass and
Chemistry
PM2.5 Elemental Carbon
Light scattering
PM2.5 EC/OC
PM10 Mass
Methane/NMHC
NOx/NOy
03
CO
NO,N02,NOx
Wind Speed
Wind Direction
Standard Deviation of Wind
Direction
Air Temperature
Relative Humidity
Delta Temperature
Measurement Method Duration/Frequency1
Sierra Anderson Dichotomous
Samplers, PTEP Sampler
Magee Scientific aethalometer with
a Harvard Impactor PM2.5 inlet
Optec nephelometer
Rupprecht & Patashnick Total
Carbon Analyzer
Rupprecht & Patashnick PM10
Federal Equivalent Method TEOM
TEI hydrocarbon analyzer
TEI NOx/NOy analyzers
One Dasibi ozone analyzer
TEI CO analyzer
TEI NOx analyzer
AGL
AGL
AGL
2 meters AGL, nephelometer height
2 meters AGL, nephelometer height
24 hrs/1 in 3 days
5 minute average, C
1 hour average, C
1 hour average, C
1 hour average, C
1 hour average, C
1 hour average, C
1 hour average, C
1 hour average, C
1 hour average, C
1 hour average, C
1 hour average, C
1 hour average, C
1 hour average, C
1 hour average, C
C
O6-one in six days; C-continuous; D-Daily; OD-Every other day.
Part I Tables, Page 4
-------
Table 1-5. Measurements Made at Phoenix, AZ in Conjunction with the Chemical Speciation Sampler
Evaluation Study.
Species
PM2 5, Coarse PM, & PM10
Mass and Chemistry
PM2 5, Coarse PM,&PM10
Mass and Chemistry
PM25
Elemental Carbon / Light
Absorption
Ambient light scattering
PM2 5 EC/OC
PM10Mass
Methane/NMHC
NOx/NOy
03
CO
NO,N02,NOx
SO2
Wind Speed
Wind Direction
Standard Deviation of Wind
Direction
Measurement Method
4 - Sierra Anderson Dichotomous
Samplers
2 - IMPROVE 25mm 4-module
samplers
2 - Magee Scientific aethalometers,
each with a Harvard Impactor PM2 5
inlet
Optec nephelometer (NGN-2)
Rupprecht & Patashnick Total
Carbon Analyzer
Rupprecht & Patashnick PM10
Federal Equivalent Method TEOM
TEI hydrocarbon analyzer
TEI NOx/NOy Trace level analyzers
Dasibi O3 analyzer
TEI CO analyzer
TEI NOx analyzer
Monitor Labs SO2 analyzer
10 meters AGL
10 meters AGL
10 meters AGL
Duration/Frequency2
6 and 24 hours/06, (2 w/ quartz, 2 w/
Teflon filters)
24 hours/OD, (each w/ Teflon,
nylon, quartz, & PM10 Teflon filters,
respectively)
5 minute average/C
1 hour average/C
1 hour average/C
1 hour average/C
1 hour average/C
1 hour average/C
1 hour average/C
1 hour average/C
1 hour average/C
1 hour average/C
1 hour average/C
1 hour average/C
1 hour average/C
O6-one in six days; C-continuous; D-Daily; OD-Every other day.
Part I Tables, Page 5
-------
Table 1-6. Overall Planned Study Schedule1.
Date
Deliverable
1998
November 10
October 1-Nov. 20
Nov 1 - 30
Nov 12-24
Nov20-DeclO
Dec 8-Dec 12
Dec 14
Standard operating procedures received from the 3 chemical speciation sampler
manufacturers and for the IMPROVE and VAPS samplers
Sites prepared for sampler installation. Power and platforms installed as needed to
accommodate all samplers (see Tables 1 and 2 for listing of samplers at each site).
Samplers will undergo an evaluation and RTF field personnel will be trained on installation,
operation, and quality control
External review by the expert peer-review panel on chemical speciation.
Samplers shipped to sampling sites and local operators trained.
Systems and performance audits performed with help from local agency (Rubidoux and
Phoenix)
First day of sampling. (Rubidoux and Phoenix)
1999
Jan 4-Jan 5 Systems and performance audits performed with help from local agency (Phoenix and
Philadelphia)
Jan 6 First day of sampling (Philadelphia and RTF)
Jan. 1 -4 Systems and performance audits performed with help from local agency. (optional)Chemical
analysis of first half of samples should be started.
March 1 Fastday of sampling. Systemsandperformanceauditsperformedwithhelpoflocalagency.
March 2-5 Decommission sites.
March 20 Chemical analyses complete.
April 15 Fevel II data delivered to EPA for all chemical analyses.
April 15-May 1 EPA reviews data and begins initial statistical analysis2
May 1-May-15 All final statistics prepared, including plots, tables, and figures
May 15-May 30 Prepare report and submit to internal NERL review group for comments
June 15 Address comments from internal review group
June 25 Provide OAQPS with draft report.
1 Actual schedule for data delivery was delayed. However, OAQPS also delayed implementation of samplers and
both schedules then coincided.
2 Detailed initial results are forwarded to OAQPS at this point or earlier.
Part I Tables, Page 6
-------
Table 1-7. Field Operations Sampling Schedule.
Sample Day
1
2
3
4
5
6
7
8
9
10
10 (blank)
11
12
13
14
15
16
17
17 (blank)
18
19
20
20 (blank)
Rubidoux
1/6/99
1/9/99
1/15/99
1/18/99
1/21/99
1/24/99
1/27/99
1/30/99
2/2/99
2/5/99
2/9/99
2/11/99
2/14/99
2/17/99
2/20/99
2/23/99
2/26/99
3/1/99
3/2/99
n/a
n/a
n/a
n/a
Phoenix
1/6/99
1/8/99
1/14/99
1/16/99
1/18/99
1/20/99
1/22/99
1/24/99
1/26/99
1/28/99
1/29/99
2/3/99
2/5/99
2/7/99
2/9/99
2/11/99
2/13/99
2/15/99
n/a
2/17/99
2/19/99
2/21/99
2/24/99
Philadelphia
1/17/99
1/19/99
1/21/99
1/23/99
1/25/99
1/27/99
1/29/99
1/31/99
2/2/99
2/4/99
2/5/99
2/8/99
2/10/99
2/12/99
2/14/99
2/16/99
2/18/99
2/20/99
n/a
2/22/99
2/24/99
2/26/99
2/27/99
RTP
1/15/99
1/17/99
1/19/99
1/21/99
1/23/99
1/25/99
1/27/99
1/29/99
1/31/99
2/2/99
2/5/99
2/6/99
2/8/99
2/10/99
2/12/99
2/14/99
2/16/99
2/18/99
n/a
2/20/99
2/22/99
2/24/99
2/25/99
F:\USER\BMILLAR\amticweb\fourcities\02 Part I Tables v8-f.wpd
Part I Tables, Page 7
-------
Inlet
Cyclone
Fractionator
Manifold
Quartz Filter
OC/EC
Teflon Filter
Mass,
Elements by XRF
Cyclone
Fractionator
1
Manifold
(D
(D
Q
O
too
Feflon Filter
Nylon Filter
SO/, NO - NH
3 ' ^"4
Fine Particle
Nitrate
Pump
Sampler
Housing
Figure I-la. Schematic of the Andersen RAAS Sampler.
Part I Figures, Page 1
-------
Figure I-lb. Picture of the Andersen RAAS Sampler Deployed in the Field at RTF.
Part I Figures, Page 2
-------
(D
a
Empty
Teflon Filter
Empty
Spiral Impactor
(D
-^J
(D
a
Empty
Teflon Filter
Empty
Spiral Impactor
Mass, SO42-, NO3-,
Elements by XRF NH4+
Empty
Nylon Filter
MgO Coated
Al Honeycomb
Denuder
Spiral Impactor
Fine Particle
Nitrate
a
ft
Quartz Filter
Quartz Filter
Empty
Spiral Impactor
OC/EC
(D
a
Quartz Filter
Quartz Filter
Empty
Spiral Impactor
Replicate OC/EC
Figure I-2a. Schematic of the MetOne SASS Sampler.
Part I Figures, Page 3
-------
-.-.,
Figure I-2b. Picture of MetOne Sampler Deployed in the Field at RTF. Left - Inlet, Filter Holders, and Meteorological Sensor. Right-Pump
box.
Part I Figures, Page 4
-------
B
OH
Size selective
Inlet (PM10)
Sodium Carbonate
Denuder
Sampler
Housing
HNO,
WINS
Impactor
Teflon Filter
Nylasorb Filter
Mass, Elements by XRF
Volatilized Nitrate
Pump
MASS 400
Size selective
Inlet (PM10)
WINS
Impactor
Quartz Filter
Pump
Sampler
Housing
OC/EC,
so42-, NO3-,
NH4+
MASS 450
Figure I-3a. Schematic of the URG MASS Sampler.
Part I Figures, Page 5
-------
Figure I-3b. Picture of the URG MASS Sampler Deployed in the Field at RTF.
Part I Figures, Page 6
-------
a
&H
H-l
(N
(N
1
Size selective
Inlet
Size selective
Cyclone
Teflon Filter
Pump
Mass,
Elements by XRF
Sampler
Housing
Figure I-4a. Schematic of the IMPROVE Sampler.
Size selective
Inlet
C
(L>
Q
m
O
o
Size selective
Cyclone
Nylon Filter
S042-, N03-,NH4+
Sampler
Housings
Size selective
Inlet
Size selective
Cyclone
OC/EC
Part I Figures, Page 7
-------
rj£33
Figure I-4b. Picture of the Filter Position, Control Box, and Timer for One of the Three Modules of the IMPROVE Sampler as Deployed in
the Field at RTF.
Part I Figures, Page £
-------
Size Selective
Inlet
a
2 3
m /-^
m /\
S >
OH
h-J
^ fe
3
^
x4 e
& £>+ ^ Sampler
hJ ^ Housing
m ^H
3
1 Teflon Filter |
H,
* 1
Teflon Filter
Nylon Filter
-------
Figure I-5b. Picture of VAPS Sampler Deployed in the Field at RTF. Left - VAPS Filter Box Showing XAD Denuder on Right and Sodium
Carbonate Coated Annular Denuder on Left. Right - Pump Box and Dry Gas Meters for Monitoring Flow.
Part I Figures, Page 10
-------
OH
Size selective
Inlet (PM10)
WINS
Impactor
Teflon Filter
Pump
Sampler
Housing
Mass,
Elements by XRF
Figure I-6a. Schematic of the Federal Reference Method Samplers
Size selective
Inlet (PM10)
WINS
Impactor
Quartz Filter
Pump
OC/EC,
S042; N03-,NH4H
Part I Figures, Page 11
-------
Andersen
BGI
R&P
Figure I-6b. Picture of FRM Samplers Deployed in the Field at RTF.
Part I Figures, Page 12
-------
PM2.5
TEFLON COATED CYCLONE
PM2.5
CYCLONE
C
NH3
DEIMUDER
MEDIUM-VOL PM10
INLET
MIXING
CHAMBER
HNO3
DENUDER
Qa
10LPM
Q
N
HQMQ-
GEIMIZER
TUBES
8LPM
TSP IONS
PM2.5 IONS, HN03, NH3
c
T
Q
T
Q
20LPM
PM2.5 CARBON, MASS
CRUSTAL ELEMENTS
C
1DLPM
PM10 CARBON, MASS
IONS, CRUSTAL ELEMEh
Q-Quartz N=>
Qa=acid impregnated quai
T= Teflon
Figure 1-7. Schematic of the SCAQMD Multi-Channel Fine Particulate Sampler.
Part I Figures, Page 13
-------
Figure 1-8. Top - Samplers on the Platform at Rubidoux, CA. Bottom - View of the Sampling
Platform and Surrounding Area at Rubidoux, CA.
Part I Figures, Page 14
-------
Figure 1-9. Sampling Platform at Phoenix, AZ.
Part I Figures, Page 15
-------
Figure I-10. Philadelphia Sampling Site. Top - Roof View. Bottom -View From the Road.
Part I Figures, Page 16
-------
Figure 1-11. Research Triangle Park Sampling Site.
C:\WINDOWS\TEMP\03 Part I FIGURES v8-f.wpd
Part I Figures, Page 17
-------
Part II
Results, Discussion, Conclusions, and Implications
-------
Table of Contents - Part II
Table of Contents - Part II Part H Page ii
List of Tables - Part II Part II. Page v
List of Figures - Part II Part n. Page vi
RESULTS Part II. Page 1
Quality Assurance Results Part II. Page 1
Performance Audit Methods and Verification Part II. Page 1
Flow Rate Audit Equipment Part II. Page 1
Barometric Pressure Audit Equipment Part II. Page 1
Temperature Audit Equipment Part II. Page 1
Field AuditResults PartII. Page 2
Limits of Detection Part II. Page 3
Field Blanks Part II. Page 3
Precision Part II. Page 4
Data Validation Part II. Page 4
Level la Data Validation Part II. Page 4
Level Ib Data Validation Part II. Page 5
Level 2b Data Validation Part II. Page 5
Data Capture Part II. Page 5
Chemical Characteristics of PM2 5 Part II. Page 7
Chemical Components Reported in the 4 City Study Database Part II. Page 7
PM2.5 and Chemical Components Collected by the FRM Part II. Page 7
Volatilization of Aerosol Nitrate Part II. Page 10
Statistical Analysis Part II. Page 11
Means of PM2.5 Mass and Major Chemical Components Part II. Page 11
Explanations of Anomalous Data Part II. Page 11
Comparison of Means Among Samplers Part II. Page 13
Comparison To Performance Criteria Part II. Page 14
Time Series Analysis Part II. Page 15
Regression Analysis Part II. Page 16
Mass Part II. Page 16
Sulfate Part II. Page 17
Nitrate Part II. Page 17
Ammonium Part II. Page 18
Part II, Page ii
-------
Organic and Elemental Carbon Part II. Page 18
Trace Elements Part II. Page 19
T-TEST Results Part II. Page 19
Results from the Paired T-Tests Part II. Page 21
Paired T-Test Results for FRM and Andersen Samplers .... Part II. Page 22
Paired T-Test Results for FRM and MetOne Samplers Part II. Page 22
Paired T-Test Results for FRM and IMPROVE Samplers
Part II. Page 22
Paired T-Test Results for FRM and URG Samplers Part II. Page 22
Paired T-Test Results for the FRM and VAPS Samplers . . . Part II. Page 23
Results from the ANOVA on Particle Nitrate Data Part II. Page 23
Aerosol Nitrate Part II. Page 23
Organic and Elemental Carbon Part II. Page 25
Denuded vs Non-Denuded Organic Carbon Results Part II. Page 25
Comparison between TOR and TOT for OC and EC in PM2.5 Part II. Page 26
Loss of Nitrate During Vacuum XRF Analysis Part II. Page 27
Field Experience with the Speciation Samplers Part II. Page 28
Andersen RAAS Part II. Page 29
Met One SASS Part II. Page 29
URG MASS Part II. Page 29
IMPROVE Part II. Page 30
VAPS Part II. Page 30
FRM Part II. Page 30
Sampler Costs and Related Spare Parts Part II. Page 31
DISCUSSION Part II. Page 32
Expectations Part II. Page 32
Major Questions Addressed Part II. Page 32
Q1. How well do PM2 5 mass and the chemical components of mass agree between the
FRM and the chemical speciation samplers tested in this study?
Part II. Page 3 3
Q2. How well can the FRM mass be reconstructed by summing the chemical components
measured by the speciation samplers Part II. Page 34
Specific Hypotheses Related to Questions Ql and Q2 Part II. Page 3 5
Denuded vs Non-Denuded Organic Carbon Results Part II. Page 38
Q3. How well do the measured concentrations from the various speciation s
-------
Q4. What are the causes of the differences among the speciation samplers for measured
concentrations of mass and the components of mass if they exist.
Part II. Page 3 9
Specific Hypotheses Related to Questions Q3 and Q4 Part II. Page 39
SUMMARY AND CONCLUSIONS Part II. Page 44
Site Characteristics Part II. Page 44
Operations Part II. Page 44
Measurement Performance Part II. Page 45
Implications Part II. Page 45
Nitrate Part II. Page 45
Carbon Part II. Page 46
REFERENCES Part II. Page 47
Tables - Part H Part II Tables, Page 1
Figures - Part n Part n Figures, Page 1
Part II, Page iv
-------
List of Tables
Table H-1.
Table II-2a.
Table II-2b.
Table II-2c.
Table H-3.
Table H-4.
Table H-5.
Table H-6.
Table H-7.
Table II-8a.
Table II-8b.
Table H-9.
Table 11-10.
Table 11-11.
Table II-12a.
Table II-12b.
Table II-12c.
Table II-12d.
Table II-13 a.
Table II-13b.
Table II-13c.
Table II-13d.
Table 11-14.
Table 11-15.
Table 11-16.
- Part II
Summary of Flow Audit Results.
Limits of Detection in ng/m3.
Average Field Blank Data for All Species and Samplers Averaged Across All Sites In
Atmospheric Concentrations.
Precision (as % CV) Achieved by FRM and Speciation Samplers Based on Results
from the Collocated Samplers at Rubidoux.
Valid Data Capture in Percent by Sampler and Major Species.
Summary of Problems Encountered In the Field During Operations of Sampler
Evaluated in this Study.
Species Concentration Data for the FRM at Each Location of the 4-City Study.
Estimated PM2.5 Mass Balance of Species versus Measured PM2.5 Mass (ug/m3) for
the FRM at Each Site.
Average Volatilized Nitrate (NO3V) in ug/m3 Observed for Each Sampler at Each
City.
Mean Analyte Concentrations for Each Sampler at All Sites.
Ratio of Speciation Sampler to FRM for Chemical Components by Site.
Regression Statistics of FRM (x-axis) versus Speciation Samplers (y-axis) for All Sites,
Samplers, and Major Species.
Results from the Paired t-Tests Between the Speciation Samplers and FRM Samplers
for Each Analyte/Site.
Results from the ANOVA for Examining Equivalency Among the Samplers for
Paniculate Nitrate.
Nitrate Concentrations (ug/m3) Measured on Teflon (T) or Quartz-Fiber (Q) Filters by
Sampler Type Averaged Over the Study Period.
Total Particle Nitrate Concentrations (ug/m3) Measured by Each Sampler Averaged
Over the Study Period.
Volatilized Nitrate Concentrations (ug/m3) Measured by Each Sampler Averaged Over
the Study Period.
Sulfate Concentrations (ug/m3) Measured on Teflon (T) or Quartz-Fiber (Q) Filters
Averaged Over the Study Period.
Nitrate Concentrations (ug/m3) Measured on Teflon (T) or Quartz-Fiber (Q) Filters
Averaged Over the Study Period.
Total Particle Nitrate Concentrations (ug/m3) Measured by Different Denuder-Filter
Pack Methods Averaged Over the Study Period.
Volatilized Nitrate Concentrations (ug/m3) Measured by Different Denuder-Filter Pack
Methods Averaged Over the Study Period.
Sulfate Concentrations (ug/m3) Measured on Teflon and Quartz Filters Averaged Over
the Study Period.
Loss of Nitrate Resulting from Analysis of Teflon Filter by Vacuum XRF.
Summary of Site Operators Surveys Regarding Speciation Sampler Setup and
operation.
Recommended Spare Parts and Supplies for Use of Chemical Speciation Samplers and
FRM Used in the Chemical Speciation Evaluation Study.
Part II, Page v
-------
List of Figures - Part II
Figure II-l. Frequency Distributions, Given as Box and Whisker Plots of PM2.5 Species at Each of
the Four Cities Studies
Figure II-2. Chemical Mass Balance of PM2.5 at Each City.
Figure II-3. Time Series Plots.
Figure II-4. Regression Analysis Plots.
Figure II-5. Denuded and Non-Denuded OC Samples as a Function of the TOR Carbon Fractions.
Figure II-6. Organic Carbon concentrations from FRM and VAPS versus Total FRM PM2.5
Mass.
Figure II-7. Organic and Elemental Carbon as a function of Face Velocity.
Figure II-8. Loss of Aerosol Nitrate from Teflon filters Due to Vacuum XRF Analysis.
Part II, Page vi
-------
RESULTS
Quality Assurance Results
Performance Audit Methods and Verification
As part of the overall field evaluation of the candidate speciation samplers, periodic performance
evaluations were conducted at each of the four sampling sites. Following the initial calibration of each
sampler and site installation, a pre-sampling evaluation was conducted to ensure that each sampler was
operating according to manufacturer's specifications. The pre-sampling evaluation consisted of single-
point verification of each sampler's operating flow rate and ambient temperature, and pressure
measurements. Subsequent mid- and post-study performance evaluations consisted of similar
evaluations. Pre-sampling audits were typically conducted approximately one day prior to the initiation
of Run 1, while the mid-study audit was conducted following Run 10. The final site audit was
conducted shortly after the last sampling period.
Flow Rate Audit Equipment
The flow rate audit consisted of a single-point measurement conducted at the operational flow rate of
each channel for each sampler. Depending upon the sampler and the specific channel being audited,
one of two flow audit devices was typically used. A Chinook Engineering Streamline flow transfer
standard (FTS #980819) was used for the majority of the channel audits. Prior to its purchase, the
Streamline FTS was calibrated by the manufacturer against a set of precision MST traceable critical
flow Venturis. Following its purchase, its performance was verified in the laboratory at RTI. The
second in field flow audit device used in this study was a Schlumberger dry gas meter. Its traceability
was established in the laboratory using EPA Method 40 CFR60, Appendix A, Method 5, Section 7.
Barometric Pressure Audit Equipment
Laboratory calibrations and field audits of speciation sampler pressure sensors were conducted using a
PSI Tronix Model PG 2000 digital pressure gauge. Laboratory validation of the Model PG 2000's
response was obtained using a certified mercury barometer previously adjusted to local station
pressure.
Temperature Audit Equipment
Two instruments were used for laboratory calibration and auditing of speciation sampler ambient
temperature response. A Tegum Model 847 thermo-calibrator was used to calibrate and audit Type K
thermocouples used in some speciation sampler designs. Other samplers were audited using a Fluke
Model 8022A digital multi-meter in conjunction with a Model SOT-1 SOU temperature probe.
Traceability of the Fluke probe was established using a traceable Hewlett Packard quartz probe (S/N
725A Model 2833A).
Part II, Page 1
-------
Field Audit Results
Results of the pre-, mid-, and post-study field flow rate audit results are summarized in
Table II-1. Values are given for flow audit results that were outside of < 10% and > 10%, all others
are within ±10% of the audit device. All flow channels were audited as well as audits of all temperature
and pressure measurement devices associated with each sampler. Pressure and temperature audit
results for the IMPROVE and VAPS were not obtained, since these devices do not measure these
quantities.
As seen in the Table D-l, only 9 flow audit values were outside the range of < 10% and > 10%. This
represents less than 3% of the total number of possible flow audits. Of those outside the 10% range,
only three were greater then 20%. The two highest values were associated with the Andersen sampler
and were apparently due to malfunctioning flow sensors; although the flow rates were likely correct.
However, since the actual flow rate was not known, these data were invalidated in the data archive.
Nineteen values fell in the range of+5% to 10% and (-5%) to
(-10%) of which 8 were associated with the IMPROVE sampler at the Phoenix site that used
25 mm filters, thus, leaving only 11 values in this range overall sampling lines.
Ambient pressure response of the speciation samplers was generally acceptable with the majority of
measured values within 5 torr of actual pressure. Occasional relative biases in excess of 7 torr were
experienced with the Andersen and URG speciation samplers. Of the approximately 80 individual
audits, only one value exceeded 10% of the audit value.
Audits of the ambient temperature monitors associated with the speciation samplers indicated excellent
agreement with the audit devices. All temperature audits, throughout the study, agreed to within 1.6 C
of the audit device or to within 5%.
Noted biases in sampler flow rates require additional attention since they adversely influence overall
sampler measurement accuracy to a greater degree than do biases in ambient temperature and pressure
response. Response problems with the Andersen's flow sensing module resulted in indicated flow rates
differing dramatically from audited values. Mean flow bias for all channels in the Andersen samplers
over all sites was +2.9% with a standard deviation of approximately 23%. In particular, the two
Andersen speciation samplers in Rubidoux experienced higher flow variations than the other three sites.
In one extreme instance, a bias in the indicated flow rate of+168% was observed. However, since the
Andersen sampler uses critical flow orifices, the actual flow rate was likely within acceptable limits of
the channel's design value. When making post-sampling calculations incorporating the sampler's
indicated flow rate, however, biases in indicated flow rates will result in calculated biases in species
concentrations.
Flow rates indicated by the Met One's five sampling channels generally agreed well with the audit
values. Mean flow bias for all channels at all sites was -0.6% with a standard deviation of only 1.2%.
A maximum indicated flow bias of-4.3% was observed during the entire study for the Met One
speciation sampler.
Part II, Page 2
-------
Flow problems were initially observed with the URG MASS 400 sampler due to limitations of the
sampler's pumping system in conjunction with the inherently high pressure drop associated with use of
the nylon filter in the Teflon/nylon filter pack. This problem was corrected by the manufacturer by
adding a second pump. Following the correction of the system's flow design, mean flow bias for the
URG sampler was -0.9% with a standard deviation of 2.0%.
Unlike the previous three samplers, the IMPROVE sampler does not provide a direct reading of
sampling flow rate. For purposes of the audit, therefore, biases are expressed in the audited flow rate
relative to the design flow rate of 23 1pm. Flow audit results showed that the sampler's actual response
agreed well with the design value. Mean flow bias for the IMPROVE sampler was -0.6% with a
standard deviation of 2.6%. Maximum flow bias in the IMPROVE sampler during the study was -
13.3%.
Similar to the IMPROVE sampler, the VAPS sampler does not provide an instantaneous reading of the
flow rate in liters per minute. In the VAPS, dry gas meters are used to indicate integrated sample
volumes over a known time period and the mean flow rate is then calculated. At Rubidoux, Phoenix,
and RTF measured flows agreed well with the audit flow rates, while at Philadelphia flow bias was
more variable, particularly near the end of the sampling program. Mean flow bias for the VAPS
sampler at all sites was +2.4% with a standard deviation of 5.7%.
Averaged over all four sites, the PM2 5 FRM samplers provided the lowest overall flow bias with a
mean value of-0.1% and a standard deviation of 2.7%. Deviations were consistently low at all sites
with average measured biases of-0.5%, +0.2%, -0.3%, and -0.1% observed at Rubidoux, Phoenix,
Philadelphia, and RTF, respectively.
Limits of Detection
Table H-2a lists the limit of detection (LOD) for each species and the analytical methods used to
determine the species concentrations. The LOD for mass is determined based on Federal Reference
Method procedure (40 CFR, Part 50, Appendix L). The LOD sulfate, nitrate, and ammonium based
on 3 time the noise in the baseline noise in the chromatogram since field blanks were non-detectable,
while for OC and EC it is based on three times the standard deviation of the field blanks. XRF limits of
detection are based on propagating errors associated with the analytical method and flow rates. In
general, LOD values were within the expected ranges.
Field Blanks
Two field blanks were collected at each site for every channel in all samplers. Filter samples used for
field blanks were handled in the same manner as filters used on regular sampling days, however, the
filters were loaded and unloaded a few minutes later and the sampler was not turned on for field blanks.
Table n-2b summarizes average field blank data by sampler averaged across sites, as well as the
standard deviation of the 10 blanks taken on each sampler over all sites. On the average, for the major
species (those listed in |ig/m3), field blank values were close to the limit of detection for the analytical
method employed and in virtually all cases less than the associated uncertainty of the measurement.
Part II, Page 3
-------
Precision
Precision results, as coefficient of variation reported in percent (CV%), are presented in Table n-2c.
Precision estimates were obtained from collocating samplers at the Rubidoux site for all sampling
periods. These precision data are limited, that is, they provide a set of data for one time period and
under one set of atmospheric conditions. Nonetheless, they provide important information needed to
evaluate the samplers performance relative to each other. In general, precision values are in the range
expected from past experience. For the major species, the CV% is typically less then 10 percent,
except for particle nitrate which is less then 20 percent. Trace elements have reported precision
estimates that are typically less then 20 percent, except Si which has reported precision values in the
range of 20- 30 percent, which may be due to the higher uncertainty in the field blank data (Table n-
Ib). The cause for the higher precision values for OC measured on the IMPROVE sampler, relative to
the other samplers, has not been determined.
Data Validation
Level la Data Validation
The contractor managing the field and laboratory operations was responsible for validating the data
through Level la. This first validation process occurred during all phases of laboratory and field
operations. As part of field operations, site operators were first responsible for ensuring that samples
received from the preparation laboratory were complete, properly labeled, and correctly matched the
entries on the supplied field data sheets. Following sample collection, the site operator was responsible
for ensuring that pertinent run data were entered on the field data sheet with particular attention given to
records of atypical sampling events such as unusual weather conditions, sampler malfunctions, or
unusual site conditions. The site operator also was responsible for ensuring that all collected field
samples were properly labeled, packed, and accompanied by the completed field data sheets and chain
of custody records.
Upon receipt of the collected field samples at the laboratory, the sample cooler was carefully unpacked
and inventoried to ensure that all samples, field data sheets, and chain of custody records were
correctly received. Personnel responsible for sample retrieval also were responsible for recording
unusual observations associated with the samples, such as torn filters, improperly loaded filter cassettes,
etc., which might adversely affect data quality. All field data sheets were then transferred to the Data
Manager who inspected the sheets for completeness. Missing or suspect information on the field data
sheets were brought to the attention of the Program Manager who was then responsible for contacting
the appropriate site operator to obtain the correct information. Once each data sheet was inspected,
the data manager manually entered the field data into the appropriate spreadsheet for subsequent
analysis. As in all phases of data entry, the Data Manager was responsible validating data entry and
correcting errors due to transcribing information from hard copy to digital format.
The Data Manager also was responsible for collecting and validating analysis results obtained from each
of the analytical laboratories responsible. Data received were entered into the appropriate spreadsheet
and inspected for possible outliers. If suspect data could not be resolved with accompanying notes
from field or laboratory data sheets, the appropriate analytical laboratory was contacted and asked to
verify the reported results. If the analysis results could not be resolved through review of the analytical
Part II, Page 4
-------
or reporting procedures, the Laboratory Manager would be contacted to determine if re-analysis of the
suspect sample would be appropriate.
Potential data outliers that could not be adequately resolved at this point were brought to the attention
of the Program Manager. Based on discussions with site personnel, laboratory support technicians,
and the Data Manager, the Program Manager was then responsible for determining the validity of the
data under review. Once the Level la validity of all data was performed, the data set was formally
released to EPA for Level Ib and Level 2 data validation and subsequent interpretation.
A complete listing of Level la data validation comments is given in Appendix C.
Level Ib Data Validation
EPA scientists were responsible for Level Ib data validation (single variate comparison of the data).
First a printout of the data set was scanned visually on a site-by-site, sample-by-sample, analyte-by-
analyte, run-by-run basis to identify obviously anomalous values. For each anomalous value identified,
the laboratory and field records associated with that sample were examined to see if there was an
explanation that would permit the value to be either adjusted (e.g. transcription error, incorrect units,
etc.) or invalidated. A computer scan also was conducted on the data set to identify values that were
less than 5 times the MDL for that analyte. The values were to be excluded from the statistical analysis
because of the greater uncertainty associated with data near the limits of detection. The remaining data
were considered valid for this analysis.
Level 2b Data Validation
Data have been validated through Level 2b through continued diagnostic and interpretive data analysis
(see Part n results and discussion). This includes multi-variate analyses in space and time where
appropriate and cross-comparisons to other samplers, in the case of this study. Physical relationships
among variables were also reexamined among samplers.
Data Capture
Table II-3 summarizes the percent valid data capture (at data validation Level 2) for each sampler by
groups of species collected on the same filter. The shaded boxes in the table indicate valid data capture
less than 90% percent. The Andersen sampler had better then 94 percent valid data capture at all sites
for all analytes. The FRM, IMPROVE, and MetOne samplers also all had excellent valid data capture
with most over 95 percent, but a few were as low as 85 percent. Only the URG and VAPS samplers
had data capture below 80 percent. The low percentages for the URG are likely do to flow problems
associated with the flow through the MASS 400 sampler. The pump initially installed with that sampler
was unable to pull 16.7 Lpm through the Teflon/nylon filter pack. Flow rates on the order of 9.5 Lpm
were noted. This problem was noted during the initial days of Phoenix and Rubidoux, the first two sites
installed. As can be seen, the problem was quickly solved by URG and better data capture was
achieved at Philadelphia and RTF. The overall poorest data capture was observed on the VAPS
sampler. Operational problems are discussed later in this report.
Part II, Page 5
-------
Problems resulting in loss of data are presented in detail in Appendix C. Sampler specific problems are
summarized in Table II-4. Problems with sampler operations were brought to the attention of the
manufacturer as soon as they were identified so they could be corrected as soon as possible. For
example, with the Andersen the indicated flows were not stable and often reported values different from
the audit flow; the lack of performance with the URG pump achieving only 60 percent of the desired
flow due to the added pressure drop of using a Teflon/nylon filter pack, and the coarse particle
penetration observed with the MetOne sampler. In all these cases and others, the manufacturer
attempted to solve the problem as expeditiously as possible. Andersen has subsequently upgraded their
flow indicator system, URG has installed a larger pump on their new units, and MetOne has developed
a new inlet based on the Sharp Cut Cyclone (BGI) (Kenny et al. 1999).
Part II, Page 6
-------
Chemical Characteristics of PM
2.5
Chemical Components Reported in the 4 City Study Database
Atmospheric particulate matter in the PM2 5 size fraction was collected on all samplers as discussed in
the introduction. Coarse particle mass also was measured by the VAPS sampler. Mass and major and
trace chemical components were determined for each sampling period and sampler. Methods are
described in the Introduction. The major components include PM2.5 mass, sulfate, nitrate, ammonium,
organic carbon, and elemental carbon. We limited the analysis of trace chemical components to a few
species to allow for a manageable data base and because many of the species measured by XRF are
often below the limits of detection for ambient samples by the XRF method. As well, to perform a
meaningful comparison among samplers, species concentrations needed to be at least 3 to 5 times their
limit of detection and present on all or at least most samplers simultaneously. XRF data reported by the
laboratory, therefore, were limited to S, Si, K, Ca, Mn, Fe, Zn, Cu, As, and Pb, knowing a priori that
these elements were the most likely to be observed above the limit of detection most of the time.
Subsequently we have limited the data base further by minimizing our use of Mn, Cu, As, and Pb, due
to their low levels (i.e., since significant number of values for these species are reported below the
methods limit of detection). The complete database, including Mn, Cu, As, and Pb is provided in
Appendix D and Dl for replicate data at Rubidoux.
It also is important to note that sulfur measured by XRF times three was in very good agreement with
sulfate measured by ion chromatography. This comparison was done using data from the FRM, thus,
sulfur by XRF was determined from Teflon filters, while sulfate by 1C was determined using quartz-fiber
filters. The ratio (S*3/SO4= ranged from 1.06 at Rubidoux to 1.01 at RTI. This also indicates that
Teflon or quartz-fiber filters are suitable for the measurement of SO4= in PM2 5 samples under the
conditions of this experiment.
PM2 5 and Chemical Components Collected by the FRM
One of the requirements of this experiment was to collect chemically speciated PM2 5 samples in
different regions of the country where we would expect to sample different chemical atmospheres, as
described in the Introduction of this report. The purpose of this was to stress the samplers over as
wide a range of PM2 5 chemical composition as possible in a limited time frame. To examine the
success of meeting the above requirement we describe below the general chemical characteristics of the
data using the PM2.5 FRM data as a reference, noting the possibility that the aerosol nitrate measured
on the PM2 5 FRM Teflon filter and the organic carbon measured on the PM2 5 FRM might be biased
low due to volatilization of ammonium nitrate or organic compounds, respectively, during sample
collection and storage or the organic carbon might be biased high due to positive sampling artifacts. A
more thorough discussion of the chemical characteristics of PM2 5 mass and composition observed
during this 4-City study are given in Tolocka et al.(2000).
Average concentrations of PM2 5 mass and chemical components as measured by the FRM samplers
are given in Table II-5. These data do not include the first two days of sampling at each of the sites, as
they were considered shakedown periods. Mass and trace elements were determined from the sample
collected on the Teflon filter, while ions and OC/EC were determined from the sample collected on the
Part II, Page 7
-------
quartz-fiber filter. Average PM2 5 mass concentrations ranged from 1 1 |ig/m3 at RTF to 28.4 |ig/m3 at
Rubidoux. Standard deviations about the means ranged from about 5 |ig/m3 at RTF to 20.6 |ig/m3,
indicating considerable day-to-day variability at each site. Maximum values at each site were about
twice the mean, except at Rubidoux where the maximum value was about three times the mean. Of the
possible 20 samples that could be collected at each site, except Rubidoux, which had a maximum
number 16, 12 valid samples were collected at Rubidoux, 17 at Phoenix and Philadelphia, and 16 at
RTF. Reasons for lost samples were discussed earlier and are summarized in Tables II-4 and
presented in detail in Appendix C.
As is typically observed, the most abundant species, of those measured, were NO3", SO4=, MV, OC,
EC, and Si, Fe, and Ca; although relative abundances varied from site-to-site. For example, OC was
the most abundant species at all sites except Rubidoux, where nitrate was the most abundant species at
11.8 jig/m3 Nitrate was lowest at RTF at 0.7 jig/m3 as expected, with moderate values at Phoenix and
Philadelphia. Also as expected, SO4= was highest at the east coast sites around 3-4 jig/m? and lowest
at the western sites, in the range of 1-2 jig/m3 At Philadelphia and RTF, SO4= was the second most
abundant species. In general, the other species, primarily Fe, Ca, and Si accounted for less than a total
of 1 jig/m3, not adjusting for their oxide form.
Frequency distributions are illustrated in Figures II- 1 for the major species as measured on the FRM
using box and whisker plots. These plots show the average, median, 25%, 75%, 5%, and 95% values
observed in the data base. Each plot shows the data for a single species at all four sites to allow for
easy comparisons between sites. Note, the scales are different on each plot, so care should be taken in
comparing among the different species. These data help to show the variability of the concentrations
from site-to-site and by species. Except for sulfate, lowest concentrations of all variables shown in
Figures II- 1 are observed in RTF. Highest values were observed at Rubidoux for mass, nitrate,
ammonium ion, and calcium with the largest variability, as well. Phoenix had the highest OC and Fe
concentrations, all with equivalent variability for a given species. Under the conditions observed during
this study, nitrate, OC, EC, and crustal related elements appear to be high in the western part of the US
then in the east. On the other hand, sulfate clearly dominates the east coast PM2 5 aerosol. Ammonium
ion is more complex as it is associated with both nitrate and sulfate.
A rough mass balance of the collected aerosol can be obtained by adjusting OC for missing hydrogen
and oxygen atoms to obtain organic material and major elements (Fe, Ca, and Si) for missing oxygen
atoms. Here, OC is adjusted by multiplying OC by 1.4 and Fe, Ca, and Si are adjusted by multiplying
by 1.43 for an estimate of Fe2O3, 1.4 for an estimate of CaO, and 2.14 for an estimate of SiO2
(Solomon et al., 1989, Eldred et al., 1998). The sum of Fe, Ca, and Si oxides provides an estimate for
crustal material, although this is a low estimate because Al, Mg, and Ti oxides also are used typically in
obtaining an estimate of crustal material (Solomon et al. 1989, Chow and Egami, 1997). Other species
are used directly with no adjustment. Table n-6 presents the results of the mass balance calculation,
while Figure D-2 illustrates the results graphically. As indicated above, nitrate was the most abundant
species at Rubidoux accounting for about 40% of the mass, followed by organic material (27% of the
mass), while organic material was the most abundant species at the other three sites, ranging from 4.8
to 10.7 |ig/m3 or 3 1% to 53%. The highest organic material concentrations were observed at Phoenix
10.7 |ig/m3 or about 53% of the measured mass. Sulfate was highest at the east coast sites accounting
Part II, Page 8
-------
for nearly a quarter of the mass on average between the two sites. Sulfate was a factor of two to four
lower at the western sites, accounting for only about 5% of the measured mass. At all sites, the
estimated crustal material represented only a minor fraction of the total mass, ranging from only 0.3 to
1.2 jig/m3 or about 2% to 6% of the measured mass. The coarse particle data collected by the VAPS,
however, indicated high coarse particle loadings in Phoenix (41% coarse/PMlO ratio) as opposed to
the east coast sites where the coarse/PMlO ratio was around 20%. In fact, at Phoenix, the
coarse/PMlO ratio exceeded 50% on three occasions, while at the east coast sites the ratio only
exceeded 30% once.
The sum of the components are compared to the measured mass in Table II-6 for the FRM samples
collected at each site. Except for Phoenix, it appears that within the error of the measurements the sum
of the components accounted for the measured mass. In actuality, there is likely an over accounting of
mass as the water associated with the particles has not been included in the sum of the components. At
Phoenix, the difference between the sum of the components and the measured mass is even larger.
Over estimates of the measured mass at all sites may be due to the use of collecting and measuring the
components of PM mass on different filter media to allow for chemical characterization of the mass
(i.e., potential positive artifacts associated with using a quart-fiber filter such as the filter collecting
HNO3 or gas phase semi-volatile organic material) or the use of an acid denuder in the Teflon filter
channel as opposed to the other channel that did not use a denuder. For the latter, it is possible that 1)
particles are being lost in the denuder (reducing the measured mass), 2) there is a shift in the MtjNOs
equilibrium due HNO3 free air are passing over the Teflon filter during sampling (reducing nitrate
concentrations on the filter, thus reducing the measured mass), 3) another reason still to be established.
Results presented in this section indicate that the sampling locations used in this study met the objective
of stressing the samplers with different chemical and meteorological ambient atmospheres. As
anticipated, Rubidoux had high nitrate with moderate to high carbon, Phoenix had the highest crustal
material with high organic material and moderate nitrate, Philadelphia and RTF had high sulfate, RTF
had low nitrate as well as the lowest overall concentrations for PM. The only unexpected result,
relative to past measurements, was the relatively high concentration of nitrate at Philadelphia. However,
most past measurements in the east have occurred during summertime conditions, when the NB^NOs
equilibrium would be shifted to the gas phase. During the winter, under conditions of this experiment,
the equilibrium would be shifted to the particle phase, assuming sufficient ammonia is available (see for
example, Russell et al. 1983; Solomon et al. 1992; Hering and Cass, 1999).
Part II, Page 9
-------
Volatilization of Aerosol Nitrate
Ammonium nitrate in air exists in dynamic equilibrium with nitric acid and ammonia (Russell et al. 1983).
This equilibrium is dependent on temperature, relative humidity, and the ambient concentrations of nitric
acid and ammonia. When collecting on a filter, this equilibrium also is dependent on the pressure drop
across the filter (face velocity or flow rate for filters that are the same size) and the air temperature that
exists in the sampler around the filter (Hering and Cass, 1999). For the chemical speciation samplers
used in this study, aerosol nitrate, with minimum bias, is measured either directly or by the sum of nitrate
measured on two filters. For reference, see Figures 1-1 to 1-5. The direct method, used by the
Andersen, MetOne, and IMPROVE samplers, uses a reactive filter (nylon) behind either a MgO
coated annular denuder (MetOne, Andersen) or a Na2CO3 coated annular denuder IMPROVE (see
Figures 1-1,1-2, and 1-4). The nylon filter is an efficient sink for HNO3; and therefore, aerosol nitrate.
The nylon filter is extracted and nitrate is determined by ion chromatography. In the second approach,
used by the URG and VAPS samplers, aerosol nitrate is collected by a two-stage filter pack located
downstream of a Na2CO3 annular denuder. The two-stage filter pack consists of a Teflon filter
followed by a nylon filter. The front filter collects particles, while the nylon backup filter collects nitrate
that has volatilized from the Teflon pre-filter (see Figures 1-3 and 1-5). Historically, the Teflon and
nylon filters in a filter pack of this type would each be extracted separately and analyzed for nitrate by
ion chromatography. The sum of the measured nitrate on the Teflon and nylon filter should be
equivalent to the nitrate measured on just the nylon filters in the direct method. However, in the design
used in these two speciation samplers, the Teflon filter is analyzed by vacuum XRF. It was believed,
and subsequently proven, that aerosol nitrate, assumed to be in the from of ammonium nitrate, would be
lost during the XRF analysis (see discussion below). Therefore, in this study, nitrate was measured on
the quartz-fiber filter collected simultaneously and also analyzed for OC. This nitrate was added to the
nitrate measured on the back-up nylon filter.
Volatilized nitrate (NO3V) is determined in a similar manner as aerosol nitrate, but the direct method
now applies to the URG and VAPS samplers. For these samplers, NO3V is measured directly from
the nylon back-up filter (see Figures 1-3 and 1-5). For the other samplers (Andersen, MetOne, and
IMPROVE), NO3 V is determined as the difference between nitrate measured on the nylon filter behind
the denuder and nitrate measured on the Teflon filter (see Figures 1-1,1-2, and 1-4). Table II-7
presents the average volatilized nitrate concentrations for each site by sampler as measured under the
conditions of this study. Also given are the average temperatures for each site over all study days, as
well as the average maximum temperature for all sampling days. For reference, average nitrate over all
samplers and periods at a given site is included with the number of samples (n) in the average. Finally,
nominal flow rate data are provided for the channels where volatilized nitrate concentrations were
determined.
As shown in Table D-7, absolute concentrations of NO3V measured on each of the samplers at a given
site were similar with the most variation observed at Rubidoux. On the average, volatilized nitrate only
exceeded 1 |ig/m3 at Phoenix, where average and maximum temperatures were higher then at the east
coast sites, although the same as at Rubidoux. However, volatilization is likely less at Rubidoux due to
the large amounts of ammonia transported to that site from the up-wind dairy farms (Russell et al. 1983;
Solomon et al. 1988; Solomon et al. 1992). Absolute values at Phoenix were on the order of 1.5
Part II, Page 10
-------
lig/m3; however, that represented about half of the total particle nitrate measured. Relative to total
particle nitrate, less then 10% of the nitrate was volatilized from Teflon filters at Rubidoux (note,
Rubidoux has extremely high ammonia concentrations) and Philadelphia, where the coolest
temperatures were observed. At RTF, about half of the total particle nitrate was NO3V; however,
nitrate levels in general were low at RTF (average of about 1 |ig/m3) so there is greater uncertainty in
the data.
Flow rates ranged from 6.71pm to 22.7 1pm through the Teflon filters where nitrate was measured. No
apparent affect due to flow rate (face velocity or pressure drop as all samplers use the same diameter
and type of filter in this analysis) can be observed in the data for NO3V. however, values are low and
the uncertainty is likely too high to be able to observe a meaningful trend.
Finally, an estimate of PM2 5 mass is provided in Table D-7 for comparison to the amount of NO3V.
The ratio of NO3V to PM2.5 mass also is given. As can be seen, under the conditions observed during
this study, volatilized nitrate represented only a small fraction of the total PM2 5 mass at all sites, ranging
from less than one percent at Rubidoux to about eight percent at Phoenix, where the highest absolute
values were observed.
Statistical Analysis
In this section, data are presented and described in several ways to provide the reader flexibility in
interpreting the data beyond what we present. Data are presented as means, in time series, regressions
relative to the FRM, differences relative to the FRM, and with the application of the t-test. The entire
data base is given in tabular form in Appendix D and Dl, time series in Appendix E, scatter plots with
trend lines in Appendix F, and absolute and percent differences in Appendix G.
Means of PM2 5 Mass and Major Chemical Components
Average concentrations (in |ig/m3) for all samplers, at all sites, for PM2 5 mass and the major
components of PM2 5 (sulfate, nitrate, ammonium, EC, and OC) as well as Si, K, Ca, Fe, and Zn are
given in Table II-8. Means are calculated only for sample sets for a given species and site when valid
data were available on all samplers at that site, thus, for the FRM, there are fewer values in the
averages in Table n-8 then in Table n-5.
Explanations of Anomalous Data
Data were reviewed and checked against field and laboratory logs. Where problems were indicated in
the logs, the data were removed or adjusted as indicated. Thus, data that were identified due to known
physical problems with the samplers or with the chemical analysis (see Table II-4 and Appendix C)
were removed from the data. In a number of cases, filters were re-analyzed to validated or invalidate
suspect data if no physical reason could be identified. Other inconsistent data (i.e., data that appears
as an outlier from the bulk of the data) remain in the data set, as no direct reason could be established
to indicate it as invalid. Detailed flow audit results indicate very few problems as described earlier. The
exceptions include five cases where the indicated flow on one channel of the Andersen sampler differed
from the audit value by more then 10 percent; one case where one of the FRM samplers differed by
Part II, Page 11
-------
more then 10 percent; and two cases where one channel of the IMPROVE differed form the audit
value by more then 10 percent. Of these, only three values differed by more then 15 percent.
The shaded data in Table n-8 indicate average values that are inconsistent with the corresponding
values obtained by the other samplers. Reasons for these inconsistencies for most of the data are given
below. All data presented in this report are validated to Level 2b, that is, multi-variate analyses of the
data have been conducted.
This section attempts to explain or at least suggest possible reasons for some of the inconsistent data
remaining in the database that is presented in this report. It is important to clarify these points before
moving on to the rest of the statistical analysis sections to help explain the results. The entire data base
is given in Appendix D for repeat 1, with replicate data at the Rubidoux site given in Appendix Dl.
Mean values presented in Table II-8 for PM2.5 mass for the MetOne sampler at Rubidoux and Phoenix
appear to be high. This is believed to be due to the passage of coarse particles through the spiral inlet
at relatively high coarse particle loadings (see Figure D-2 and Table II-6 for coarse particle
concentrations observed at each site). This also is illustrated in the data for Si, Ca, and Fe, which are
species typically associated primarily with soil dust and in the coarse particle mode of the atmospheric
aerosol (Chow et al. 1996). Silicon, Ca, and Fe are considerably higher for the MetOne at these sites
then for the FRM, confirming coarse particle penetration. Recent test by Peters et al. (2000) indicate
that the efficiency curve of the Spiral Impactor is somewhat more shallow than the FRM and may
experience particle bounce problems for large particles. As a direct result of this 4-City study, MetOne
has replaced the spiral inlet with a sharp cut cyclone (BGI) (Merrifield, personal communication) and
results of a study conducted by MetOne of the new inlet compared to the FRM are presented in
Appendix H. Peters et al. (2000) included the BGI sharp cut cyclone in their evaluation as well, and
the efficiency of the cyclone was much closer to that of the FRM (see Appendix I for a draft of Peters
et al. 2000).
At Rubidoux, PM2 5 mass by the URG sampler seems to be high relative to the FRM as well. Looking
at individual values, there appears to be a consistently high bias at the higher mass concentrations, with
lower concentrations being essentially equivalent. This difference needs to be investigated further,
because it was not observed at the other sites. The VAPS sampler at Rubidoux also appears to be
high relative to the FRM, Andersen, and IMPROVE samplers. As explained above, the collection jet
for the VAPS was left out after the first sample, thus, both the fine and coarse particle legs of the
VAPS essentially collected PM10. This did not result in a large bias for all species, just for nitrate and
the crustal related elements (Si, Ca, and Fe). This is reasonable as most of the other species are
primarily observed in the fine particle mode. Note, the same observation is true for the MetOne
sampler, verifying that under the conditions of this study, SO4=, NFV, EC, OC, and most of the nitrate
are primarily found in the fine particles.
The mean elemental carbon concentration reported for the primary IMPROVE sampler at Rubidoux
appears to be low. This is due to low values relative to the FRM and the collocated URG sampler for
sampling periods froml/15/99 to 2/2/99. The replicate sampler concentrations are more in line with the
other samplers at Rubidoux, with an average value of 2.7 ug/m3. It is possible that a leak occurred
Part II, Page 12
-------
during sampling that was not corrected until the mid-study audit, however, the audit results are well with
in 10 percent, the audit criteria for sampler flow rates. The data remain in the data base as valid data.
The mean organic carbon concentration reported in Table II-8 for the primary Andersen sampler at
Rubidoux are high relative to the other samplers. All individual values for that sampler are high, while
the replicate appears to in line with the other samplers during all sampling periods. Quality assurance
checks indicated that the flow rate reported by the sampler was a factor of about 2 lower then the
measured audit flow rate. Andersen Instrument Company (W. Davis, personal communication) has
subsequently noted a problem with the reported indicated flow and has modified (upgraded) their
system accordingly. Validation will occur during the Atlanta field program this August. The replicate
OC data were used in place of the primary samplers OC data. The higher value for OC by the
Andersen sampler at Phoenix appears to be due to a systematic bias. Audit data indicate the indicated
flow rate was high, therefore, the concentration would be low by ten percent.
High crustal related trace elements (Fe, Si, and Ca) for the VAPS samplers at Rubidoux and Phoenix
suggest that the cutpoint (slope and 50% collection efficiency) of the VAPS sampler is not as efficient
as the other samplers. By design, the WINS impactor associated with the FRM and used in the URG
speciation sampler have steeper slopes then the VAPS virtual impactor (Peters and Vanderpool,
1996). Additional studies are needed to confirm if the slope in the efficiency curve for the VAPS virtual
impactor is the reason for the higher values for the crustal related trace elements at these sites where
there was significant coarse particle loadings, relative to Philadelphia and RTF.
Comparison of Means Among Samplers
Mean values for PM2 5 mass and the major chemical components are presented in Table D-8 by
species, site, and sampler to allow for comparison of individual species among each of the samplers and
across sites. Means are calculated pair-wise when all samplers reported a value and the first two days
were not included in the average as they were considered to be the shake down period at each site.
Thus, the number of samples in the average is somewhat lower than the 20 attempted. It is readily
apparent, with the exceptions mentioned above and highlighted in the table, that all the samplers agree
within reasonable expectations (based on the precision data) with each other; although real differences
among the samplers exist for organic carbon and nitrate. Mass for example ranging from 10 to 20
|ig/m3 typically only has absolute differences among the samplers of about 1 |ig/m3, excluding the few
exceptions noted above, i.e., the VAPS, which uses a virtual impactor and the MetOne, which used the
spiral inlet that has subsequently been shown to allow coarse particle penetration (Tom Merrifield,
MetOne, personal communication). Differences for most of the other major components (sulfate,
ammonium, and EC) are on the order of a few tenths to 0.5 |ig/m3 and typically on the order of 10 to
20 percent for most species. Nitrate and OC appear to be the most variable with differences in the
average values ranging from 1-3 |ig/m3 and 1-4 |ig/m3, respectively. However, the IMPROVE sampler
tends to produce lower ammonium values on average at all sites then the other samplers, although it is
not a large practical difference. This may be due to attempting to collect a basic substance on a basic
material; ideal for collecting and maintaining acidic species, but not basic species, such as ammonia, one
of the volatilization products of ammonium nitrate. Differences observed for OC will be discussed
below in the Organic Carbon section (Section F, Organic and Elemental Carbon). Nitrate tends to be
higher at all four sites for the URG and VAPS sampler and low for the IMPROVE sampler at Phoenix.
Part II, Page 13
-------
Reasons for these differences will be discussed below under the nitrate section (Section E, Aerosol
Nitrate). Trace element concentrations are typically within 0.1 |ig/m3 or less. This agreement and the
effect of potential outliers in the data will be seen more clearly in the sections that follow.
Comparison To Performance Criteria
EPA's Expert Panel on Chemical Speciation met in May 1999 to review the draft Guidance Document
for Chemical Speciation (EPA 1999) and provide comments on results from the Chemical Speciation
Sampler Evaluation Study described in this report. The Expert Panel provided recommendations in
June 1999 (Koutrakis 1999) and a final report in August 1999. Recommendations for the Chemical
Speciation Sampler Evaluation Study included suggested performance criteria by which to judge the
relative performance of the individual samplers on a species-by-species basis relative to a sampler
designated as the relative reference. Performance criteria were given for mass, sulfate, nitrate, and
ammonium. Performance criteria for OC and EC were not recommended due to the poor
understanding of how to collect OC with minimal bias. Performance criteria were not established for
trace elements. However, results from this study should help to define performance criteria for future
studies and uses of the speciation samplers.
The expert Panel established the following performance criteria: for mass, paniculate nitrate, and
ammonium ions, the ratio of the test sampler to the relative reference should be 1 ± 0.1 with a linear
regression correlation coefficient (R2) of 0.9; while sulfate should have a ratio of 1 ± 0.05 and a R2 of
0.05. The precision data (Table 2c) suggest the performance criteria established by the Expert Panel
is, in general, reasonable. However, a ratio of 1 ±0.15 might be more reasonable for particle nitrate
and organic carbon. Based on the precision data in Table 2c, performance criteria for the other species
could be suggested and would be in the range of 1 ± 0.15 for OC, EC, K, Ca, Fe, Zn, while Si and
Mn might have criteria in the range of 1 ± 0.25.
Table U-8b presents the ratio of each speciation sample relative to the FRM for all major species and
most trace species. OC is given relative to the VAPS OC measured on a quartz-fiber filter behind the
XAD denuder. In general, the Expert Panel's performance criteria were obtained for the four EPA test
samplers (Andersen, MetOne, IMPROVE, and URG) at Philadelphia and RTF, where the coarse
particle loadings were low, temperatures were low, or concentrations were low; although a few
exceptions are noted for nitrate and ammonium ion. The URG samplers exceeded the performance
criteria for mass, sulfate, particulate nitrate and ammonium ion in either or both Phoenix and Rubidoux;
although for sulfate the ratio was only 1.06 and likely well within acceptable error of the 5%
performance criteria. Positive bias was indicated for mass, sulfate, and particle nitrate and negative bias
for ammonium ion. The MetOne sampler exceeded the performance criteria for mass and sulfate at
Phoenix and Rubidoux and for ammonium ion at Phoenix; although the MetOne was acceptable for
nitrate at Phoenix, suggesting an individual problem at Phoenix for ammonium. The ratio for the
IMPROVE sampler was low at all sites for ammonium and exceeded the performance criteria at
Rubidoux, Phoenix, and Philadelphia.
Given the performance criteria suggested above for the other species, the data are mixed with samplers
meeting the criteria at some sites for one species, but not for another. However, in general, most ratios
fall within 20% of 1 for all species and samplers with a few exceptions. For example, OC exceeds
Part II, Page 14
-------
even this criteria at most sites and for most samplers. However, as discussed below, this is likely due to
differences in face velocity through the filter or pressure drop across the filter. As well, crustal related
elements far exceed the criteria for the MetOne sampler, due to the inlet/fractionator problems noted
earlier.
Time Series Analysis
Time series of all the data are given in Appendix E with examples given in Figure II-3a-i for mass,
sulfate, nitrate, ammonium, OC, EC, Si, Ca, and Fe. These species were chosen as they represent the
major components of the PM2 5 mass and the indicators for crustal material (Si, Ca, and Fe) typically
associated with coarse (>PM2.5 um) particles. Prior to this, results from the VAPS sampler at
Rubidoux have been included to show the identified problem with that sampler. For the time series
analysis, the VAPS data at Rubidoux have been removed. At this point, it was our intention to remove
the data for the first two days of sampling, as these were considered shake down periods, however, the
first two days of data remain in the time series plots for mass, sulfate, nitrate, and ammonium, but have
been removed from the plots for the other species. Thus, the reason for the different sample numbers in
the time series plots. The first two days of data for all the species have been removed for analyses that
follow (i.e., regression analysis, difference analysis, and t-test) for all species.
In examining the time series plots it is important to take note of the range of concentrations observed at
each site, as there is considerable variability among the sites, as noted earlier in the discussion about the
means. PM2 5 mass exceeded the 24-hour PM2 5 standard of 65 |ig/m3
only once during the study at Rubidoux.
The time series plots show generally good agreement among the samplers for the major species, noting
the exceptions mentioned earlier under the discussion of the means. In general, data for a given site and
species tend to increase and decrease together; although at some locations and for certain species there
is considerable scatter among the samplers (e.g., nitrate at Phoenix and the OC data at all sites). A
more detailed discussion of how the samplers compare and the relative biases observed among the
samplers is given in the next section on regression analysis.
Regression Analysis
Regression analysis, slopes, intercepts, and correlation coefficients were obtained for each species and
sampler against the FRM. Results of this analysis for all sites and species are illustrated in scatter plots
in Appendix F with examples given in Figure E-4a-i for mass, sulfate, nitrate, ammonium, OC, EC, Si,
Ca, and Fe. These species were chosen as they represent the major components of the PM2 5 mass
and are indicators of crustal material (Si, Ca, and Fe). The scatter plots include a regression line
through the data, where the intercept is allowed to float (i.e., not set to zero). Table II-9 provides a
complete listing of the slopes, intercepts, and correlation coefficients (square root of the R2 value given)
for all species (excluding As and Pb), sites, and samplers against the FRM and their associated
standard errors. The first two days of data have been removed from all regression analyses, as
explained earlier, since those days were considered the shake down period. As a reminder, the VAPS
data at Rubidoux likely represent PM10 rather than PM2 5, since the acceleration nozzle was not placed
back into the sampler after the first sampling period.
Part II, Page 15
-------
A gross examination of the regression plots in Figures II-4a-i and in Appendix D show that the
samplers are, in general, providing similar data, where the slopes are parallel, reasonably closely
packed, and scattered around the 1:1 line. Regression statistics shown in Table II-9 also confirm the
generally good overall agreement with the majority of the slopes in the range from 0.8 to 1.2 and R2
values greater then 0.8 and even 0.9. Exceptions to this follow the discussion presented in the sections
above, that is, there are a few known problems with sampler operations and there are a few data points
that are either significantly above or below the trend line.
Typically lines that are not parallel with the group on a given plot have one or two points that are
relatively far from the trend line, and if these data points that are at high or low concentration they
greatly influence the slope, intercept, and correlation coefficient. These "inconsistent" data points have
been investigated thoroughly, including re-analysis of filters or extracts, and those that remain are
considered to be valid data, i.e., no explanation can be found to invalidate the data. The following
provides brief comments on each of the species presented in Figure U-4a-i. Comparisons are made to
the 1:1 line (not shown) of the speciation samplers to the FRM, since volatilized nitrate appeared to
represent only a small fraction of the nitrate collected, with the exception of Phoenix, where about half
of the nitrate was measured as volatilized nitrate. Similar data are not available for OC; although one
would expect more volatilization in the locations with higher temperatures, Phoenix and Rubidoux, then
at RTF and Philadelphia, the latter which had the coolest temperatures (see Table II-7 for average and
average maximum temperatures observed during sampling days). Comparisons are made to FRM,
since it is a well characterized aerosol sampler with a sharp cutpoint and under wintertime conditions,
volatilization should be minimal.
Mass
In general, and as expected from the time series results, the trend lines are tightly packed and parallel
with and scattered around the 1:1 line. A few of the trend lines are noticeably separated from the
others, being above the 1:1 line. This is the case for the URG and MetOne mass results at Rubidoux,
the MetOne mass results at Phoenix, and the for the VAPS at Philadelphia. Audit results for the URG
were within 5 percent for all audits, however, sulfate was also high, suggesting a small bias in the
reported flow rates. The MetOne sampler is high at all sites, and noticeably high at Rubidoux and
Phoenix due to the reasons indicated earlier (see section: Means of PM2 5 Mass and Major
Chemical Components).
Sulfate
Regression analysis of the sulfate data shows good agreement among the samplers. Within day
variations among the samplers are on the order of a few tenths of a |ig/m3, lines are, in general, parallel
and scattered about the 1:1 line. This suggests that the quality assurance and quality control objectives
were achieved in monitoring and validating flow rates and chemical analysis, at least for sulfate. It also
suggests that differences seen in what are typically considered coarse particle species (Fe, Ca, Si), are
due to the differences in the collection efficiency of the inlets, as has been identified for the MetOne
sampler and is suggested here for the VAPS sampler, which uses a virtual impactor to size fractionate
PM in to a fine (<2.5 um) and coarse (2.5-10 um) size ranges. At Rubidoux, Philadelphia, and RTF
the reported concentrations among the samplers appear mostly random with no one sampler showing a
Part II, Page 16
-------
systematic bias. At Phoenix, sulfate concentrations are lower then the other sites (range is from 0.2 - 2
|ig/m3) and while there looks to be more variability, the overall range (y-axis) is about one fourth the
other sites. In general, differences among the samplers, on a day-to-day basis, are on the order of a
few tenths of a |ig/m3, and certainly small compared to PM2 5 mass, and even to sulfate as observed in
the figures.
Nitrate
The variability in nitrate, as measured by the denuder/filter pack method is somewhat larger than sulfate
at Phoenix and RTF as well, the variability for nitrate is larger at Phoenix and RTF than at Rubidoux
and Philadelphia. The highest concentrations of nitrate were observed at Rubidoux and the lowest at
RTF, thus, it is important to note the range of concentrations given on the y-axis when looking at these
plots. Variability among samplers for a given day range from a few tenths of a |ig/m3 at low nitrate
concentrations to about 3 |ig/m3 at mid-range concentrations, and about 7 |ig/m3 on the highest
sampling day (maximum FRM nitrate was 38 |ig/m3, observed at Rubidoux). At Rubidoux, the URG
and MetOne are reporting slightly higher concentrations (2-3 |ig/m3) relative to a 1:1 line, while the
other samplers are slightly below by about the same amount. At Phoenix and RTF, the VAPS and
URG samplers are reporting nitrate concentrations higher then the 1:1 line (also seen in the means data,
Table n-8a, means differ by about 30% - 50% among samplers at these sites), while the IMPROVE
sampler is slightly below the 1:1 line relative to the FRM. However, audit flow rates for the IMPROVE
sampler were about 12 percent high relative to the indicated flow rate, possibly accounting for at least
part of the discrepancy.
These differences can be accounted for, in general, by the different methods used by the speciation
samplers for collecting aerosol nitrate. Nitrate can be determined directly on either Teflon or quartz-
fiber filters, or using the denuder/filter pack methods, one a direct measurement on a nylon filter behind
a denuder, the other requiring the sum of nitrate measured on a nylon filter behind a denuder and Teflon
filter plus nitrate on from a parallel quartz-fiber filter. Volatilized nitrate also is determined differently,
depending on how particle nitrate is determined. A more careful examination of nitrate measured by
these different methods is given below (Section E, Aerosol Nitrate).
Ammonium
Regression plots for ammonium ion are given in Figure II-4d. In general, the variability in these data
tend to follow a similar pattern to nitrate, with the exception that ammonium measured by the
IMPROVE sampler tends to result in the lowest slopes at each site (Table n-9 and Figure H-4d). This
was also observed in the comparison among the means (Table n-8a). It is suggested here, as
discussed earlier, that the nylon filter may not be the best medium for collecting ammonium, in
particular, ammonium associated with ammonium nitrate. The use of the nylon filter, with a basic pH,
allows HNO3 to be recaptured when volatilized after collection; however, ammonium would not be
recaptured and lost from the measurement. In fact, results here suggest the use of a nylon filter it may
even enhance the loss of ammonium from the filter.
Organic and Elemental Carbon
Part II, Page 17
-------
Organic carbon concentrations range from about 3 to 9 jig/m3 on average over all sites, while elemental
carbon concentrations range from about 1 to 4 jig/m3 (see Figures II-4d and U-4e and Table II-8).
Variability of OC and EC at all sites was large relative to these averages and often approaching 6-8
jig/m3 for OC and 2 - 3 for EC. However, as will be discussed later, the measured concentrations of
OC are a function of the face velocity or pressure drop across the filter, or for these samplers because
they all used the same size filters, the flow rate. As well, the VAPS sampler had an XAD coated
annular denuder designed to remove SVOC in the gas phase (Gundel and Lane, 1999; Gundel et al,
1995) that have been noted to cause positive artifacts for OC (Turpin et al., 1994).
In general, the regression lines for the OC measurements by the speciation samplers relative to the
FRM are parallel to and scattered about the 1:1 line; although there is considerable variability among
the samplers with parallel, but definitely consistent high or low biases relative to the 1:1 line for all
samplers, with the exception of the IMPROVE sampler at Rubidoux. For that sampler, the slope is
considerable lower then the other samplers and appears to be due to a series of 5-6 data points at the
beginning of the study, after which, the data agree better. The different slope was not observed at the
other sites for OC by the IMPROVE sampler; however, a similar slope was observed at Rubidoux for
EC, thus, suggesting a potential problem with that module during the beginning of the study. General
trends in bias are also observed among the samplers for OC. For example, the OC VAPS regression
line was typically the lowest (excluding Rubidoux where the collection cone was missing from the
sampler), while the Andersen and MetOne typically had the highest regression line on the plot, while
operating at the lowest flow rates among the samplers. The MetOne data however, may be
compromised due to the problem noted earlier with the spiral inlet.
Nonetheless, there was considerable variability among the samplers for measuring OC; although much
of the variation can be explained due to operational problems and more importantly due to differences
in how the samplers operate. The latter is likely due to differences in the face velocity or pressure drop
across the collection media in the different samplers or the effect an organic denuder has on removing
SVOC with subsequent collection on pre-baked quartz-fiber filters. OC data are discussed more fully
later in this report (Section F, Organic and Elemental Carbon).
Elemental carbon, a stable species like sulfate, shows somewhat less variability among the samplers
then OC, but the variability is still somewhat larger then for sulfate. In general, the regression lines are
parallel to the 1:1 and reasonably closely packed at Phoenix, RTF, and Philadelphia; although for
Philadelphia the regression line for the Andersen sampler is not parallel due to one high point at low
FRM concentrations and one low point at high FRM concentrations. Validation of the data, including
re-analysis of the samplers suggested these data to be valid. Other data from the Andersen fall close to
the 1:1 line. At Rubidoux, a series of low points for the IMPROVE sampler add considerably to the
observed variability at that site. For the IMPROVE sampler, as noted earlier, the data for the first 5-6
days of the study appear low, relative to the other samples, and then for the remainder of the study
agree much better. This is also true for the OC data; although it is not as noticeable with the OC data
due to the additional scatter in the plot. However, the EC concentration data for the IMPROVE
samplers at the other sites show better agreement. The best agreement was observed at RTF, where
the lowest EC concentrations were observed. EC data will have to be more carefully studied to
Part II, Page 18
-------
determine if the variability is due to the design of the samplers, operations by site operators, variations
in the laboratory analysis, or some other reason.
Trace Elements
The trace elements measured as part of this experiment are non-volatile and would be expected to
agree well among the samplers, and in general they do. Trend lines for trace elements are typically
clustered with parallel slopes; although there are a number of single data points significantly away from
the 1:1 line that sometimes skew the slope and intercept of the regression line, as well there are the
exceptions noted earlier. For example, the crustal related species (Si, Fe, and Ca) are skewed high for
the MetOne sampler due the bias noted with the inlet of that sampler.
T-TEST Results
The sample collection schedule at the four sites was designed to permit an analysis of variance
(ANOVA) to be conducted on the results. The plan was to have a single data set and to conduct
ANOVA on it on a site-by-site and analyte by analyte basis using the following linear model to establish
the comparability of the samplers:
yljk = u + D1 + Lj + Sk + DL:J + DSlk + LSjk +DLSljk
where:
y = Analyte
Di = Days with associated degrees of freedom equal to D - 1.
Lj = Locations with associated degrees of freedom of L - 1.
Sk = Samplers with associated degrees of freedom of S - 1.
DLy = The interaction between days and locations.
DSik = The interaction between days and samplers.
LSik = The interaction between locations and samplers.
DLSyk = The three way interaction term (used to estimate overall experimental error).
Part II, Page 19
-------
The results from the ANOVA would be used to test the following hypotheses.
There will be no statistical difference between the PM2 5 mass concentrations measured
by the FRM-Teflon filter and the Teflon filters on the other samplers.
There will be no statistical difference between the FRM-Teflon filter sampler, the
VAPS sampler, and each speciation sampler for concentrations of trace elements (by
atmospheric pressure XRF).
There will be no statistical difference between the FRM-quartz-filter-equipped sampler,
the VAPS sampler, and each speciation sampler for nitrate, ammonium, and sulfate (by
ion chromatography).
There will be no statistical difference between the FRM-quartz-filter-equipped sampler,
the VAPS sampler, and each speciation sampler for OC and EC (by thermal optical
reflectance).
At the completion of the Level Ib data validation, it was obvious that there were a large number of
missing values and that there were large variations in the concentrations measured for many of the
analytes at the four sites. These factors made it impractical to conduct the planned ANOVA, so
instead, a paired t-test (alpha = 0.05) approach was used to determine whether two samplers were
statistically similar. Based on the large numbers of values below five times the MDL, it also was
decided to exclude As, Cu, Pb, and Mn from this statistical analysis.
Part II, Page 20
-------
Results from the Paired T-Tests
Presented below are the results from those paired t-tests where the FRM value was used as the
reference standard. Although the FRM, Andersen, IMPROVE, Met One, and URG samplers were
operated in duplicate at Rubidoux, only the value from each primary sampler was used for the t-test.
The results are presented in tabular form in Tables n-10a through II-10e for each sampler (a-e) on a
site-by-site, analyte-by-analyte basis using the following parameters:
Actual P-value. P-values less than 0.05 mean that the FRM and the speciation
sampler's results are statistically different at the site for the analyte tested.
Statistical decision made based on the P-value. "Y" means the two results for
the FRM and speciation sampler are statistically the same and "N" means they
are statistically different.
• The number of times the FRM value was larger than the corresponding
speciation sampler value (numerator) compared to the total number of
differences used in the paired t-test for that site/analyte (denominator).
Appendices G-l and G-2 provide a complete listing absolute and percent
differences between the FRM and each speciation sampler by site and sampling
period. These data were used to calculate the above ratio.
Besides the t-test, an ANOVA was conducted using only the particle nitrate data from the four
speciation samplers. This ANOVA was conducted for the following reasons. First, the speciation
samplers were designed to provide an accurate measure of particle nitrate concentration, whereas, the
FRM was not. Second, the particle nitrate concentrations measured at a site did not vary substantially
across all the sampling runs. Third, except for the URG sampler at site 1, the particle nitrate data set
was reasonably complete, i.e. there were few missing values.
The sampling results for the first two days at each site were excluded from the t-tests and the ANOVA
because these were considered to be sampler shakedown/operator training days. The decision to
exclude the first two days from the statistical analysis was made before the first samples were taken.
Overall, there appears to be no consistent pattern in the T-test results presented in Table 10. This is
likely due to the reasonably high precision obtained by these samplers and due to the high correlations
among the samplers. Note, Table n-2c presents the sample precision, but the mean precision is about
4 times smaller (n = 14 or 18 in most cases). Below is a brief discussion on a sampler-by-sampler
comparisons relative the FRM.
Paired T-Test Results for FRM and Andersen Samplers
Table n-10a presents results from the paired t-tests for the FRM and Andersen samplers. The
Andersen was the only speciation sampler that produced PM2.5 mass concentration values that were
statistically equivalent to those of the FRM at all four sites. The only other analytes that were two
Part II, Page 21
-------
samplers were equivalent at the four sites are S (by XRF) and Zn. It is interesting to note that the two
samplers are different for sulfate (by 1C) at site 2 (P value = 0.0001) and site 3 (P value - 0.0028),
despite being statistically equivalent at these sites for S by XRF. This anomalous situation might be
attributable to analytical imprecision at the low sulfate concentrations measured. Table n-10a also
shows that at sites 1, 2 and 3, the Andersen sampler consistently produced lower particle NO3" and
" values and higher OC values than the FRM.
Paired T-Test Results for FRM and MetOne Samplers
Table II-10b presents the results from the paired t-tests for the FRM and the MetOne samplers.
Particle nitrate was the only analyte for which the two samplers produced statistically equivalent results
at the four sites. The PM2 5 concentrations and the Si, K, Ca and Fe concentrations measured by the
two samplers were statistically different either at all four sites (K, Ca, Fe) or at three of the four sites
(PM2 5, Si). The MetOne produced PM2 5, Ca Si, K, and Fe concentrations at the four sites that were
consistently larger than those from the FRM sampler (with the exception of PM2.5 mass at Site 3).
These results indicate strongly that the sampler inlet did not effectively remove coarse particles, as
discussed earlier. For the other analytes the agreement between the two samplers were inconsistent
across sites and analytes.
Paired T-Test Results for FRM and IMPROVE Samplers
Table II-10c presents the results from the paired t-tests for the FRM and the IMPROVE samplers.
The two samplers produced statistically equivalent results at all four sites for only three analytes: S (by
XRF), K, and Zn, but also produced results for three other analytes (MV, OC, and EC) that were
statistically different at all four sites. The statistical equivalence for the other analytes was highly variable
across the sites. A strong negative bias existed at all sites for the IMPROVE sampler for sulfate by 1C,
S by XRF, particle nitrate, ammonium, OC, and EC, and a consistent positive bias exists for the
sampler for Si, Ca, and Fe at sites 1 and 2, where there were significant coarse particle concentrations.
This may suggest a positive bias for coarse particle penetration relative to the FRM for the IMPROVE
sampler.
Paired T-Test Results for FRM and URG Samplers
Table n-10d presents the paired t-test results for the FRM and URG samplers. This sampler suffered
from a variety of operational problems at Site 1, which resulted in a large number of samples being lost
or declared invalid. From a statistical point of view, the performance of this sampler with respect to the
FRM was poor. This was unexpected since the URG sampler closely matches the FRM with respect
to its design and flow operation. Zn was the only analyte where the URG yielded results that were
statistically equivalent to the FRM at all four sites. Otherwise the statistical equivalence between the
FRM and URG samplers varied inconsistently from site-to-site and analyte-to-analyte. It is likely that
the mechanical problems associated with the specific URG samplers used in this study and not the
design of the sampler itself was the cause of the poor agreement between the FRM and the URG
sampler.
Paired T-Test Results for the FRM and VAPS Samplers
Table II-10e presents the paired t-test results for the FRM and the VAPS sampler. As noted earlier in
this report, the VAPS sampler's virtual impactor was not correctly assembled at Site 1 but was
Part II, Page 22
-------
included in the analysis to illustrate that point as well as for species, such as sulfate that are not affected
by a cutpoint greater than 2.5 |im. S by XRF and Zn were the only analytes for which the two
samplers produced statistically equivalent results at all four sites. For all the other analytes, the
statistical equivalency of the results differed from site-to-site. Mass and the trace elements showed a
consistent negative bias relative to the FRM, while the other species showed a consistent positive bias.
Results from the ANOVA on Particle Nitrate Data
The ANOVA was done on a site-by-site basis. At Rubidoux ANOVA involved 51 concentration
values covering 14 days of sampling, at Phoenix ANOVA involved 67 concentration values covering
18 days of sampling, at Philadelphia ANOVA involved 71 concentration values covering 18 days of
sampling, and at Research Triangle Park ANOVA involved 67 concentration values covering 18 days
of sampling. The results of the ANOVA are presented in Table II-11 in terms of the mean
concentrations and the number of values (in parentheses) used in calculating the mean. The last column
in Table D-l 1 identifies the samplers that are statistically equivalent, that is, samplers within parentheses
are statistically equivalent. For example, at Phoenix, the means for the Andersen and the IMPROVE
samplers are statistically equivalent as are those for the Andersen and MetOne samplers. Note,
however, that the means for the IMPROVE and MetOne samplers are not statistically equivalent. Only
the Andersen and the IMPROVE samplers apparently had statistically equivalent results across all four
sites for nitrate.
Aerosol Nitrate
As discussed above, aerosol nitrate as well as volatilized aerosol nitrate were determined by several
methods, and there appears to be differences in the nitrate concentrations reported by those different
methods. Tables II-12a through II-12c present data by sampler type and on a site-by-site basis for
nitrate measured by the different methods employed in this study. Sulfate data are presented in Table
H-12d, for comparison to a stable species, not influenced by volatilization. Tables n-13a through n-
13d present similar data, however like methods are averaged, as well, the ratio of the methods are
given to indicate the relative difference between methods.
The data in Table H-12a represent nitrate concentrations measured directly on Teflon and quartz-fiber
filters without a preceding denuder. The data in this table suggests a difference between the two filter
types for collecting aerosol nitrate. Chow (1995) indicates that both Teflon and quartz-fiber filters are
suitable for the collection of nitrate, that is, both experience similar negative and positive artifacts.
However, they are similar for quartz-fiber filters that have not been pre-treated (heated to 900 C) for
OC collection, as is the case in this study. Apparently, heating the quartz-fiber filter reactivates it and
allows it to adsorb gas phase nitrogen species (e.g., HNO3). Table II-13a shows the average of nitrate
measured on Teflon filters, quartz-fiber filters, the difference (Q-T), and the ratio of the quartz-fiber
filter nitrate results to those measured on the Teflon filter. Ratios range from about 1.2 at Philadelphia,
where temperatures are sufficiently cold to maintain the ammonium nitrate in the aerosol phase (no
HNO3 available to react with the quartz-fiber filter), to about 1.8 at Phoenix and RTF, where higher
temperatures were observed as well as the likely hood of lower ammonia concentrations relative to
Rubidoux, thus the possibility of HNO3 available to react with the quartz-fiber filter. Absolute
Part II, Page 23
-------
differences between nitrate concentrations measured on the quartz-fiber filter and the Teflon filter range
from about 0.2 |ig/m3 at Philadelphia to 1.0 |ig/m3 at Phoenix, in a similar pattern to the ratio.
As described above, particulate nitrate was measured either directly on a filter behind a denuder or
indirectly by summing two filters. Table II-12b summarizes these data by method or sampler. Once
again, samplers that use a quartz-fiber filter (the indirect method) appear to measure more particulate
nitrate then the direct method, which does not use a quartz-fiber filter. Table n-13b provides average
data for the two methods. The ratio of the indirect to the direct method ranges from 1.1 to 1.5 and
each value is slightly lower then the corresponding ratio in Table n-13a; although their associated
standard deviations suggest there is likely no difference. However, absolute differences (Indirect -
Direct Methods, where the indirect method includes the use of nitrate measured on a quartz-fiber filter)
are slightly larger then the differences observed for nitrate measured on the quartz-fiber filter and the
Teflon filter (Table II-13 a). This suggests another factor also is involved causing the apparent
difference between these methods. Hering (Aerosol Dynamics, Berkeley, CA, personal
communication) suggests there may be a small positive bias, possibly due to the adsorption of other
nitrogen containing compounds (e.g., NO2), when using Na2CO3 coated denuders with nylon filters
relative to using Na2CO3 impregnated filters with a Na2CO3 coated denuder. However, the
IMPROVE sampler shows a trend opposite that of the URG and VAPS; however, it operates at a
higher flow rate. A potential positive bias has been observed in this study using the indirect method to
measure particulate nitrate if that method used a pre-heated quartz-fiber filter to obtain particle nitrate
concentrations.
Volatilized nitrate concentrations measured by each method or sampler are given in Table H-12c. The
IMPROVE value was obtained relative to the FRM, since nitrate was only measured on the nylon filter
in that sampler. In determining volatilized nitrate, the quartz-fiber filter is not used by any sampler, thus
eliminating that variable from the uncertainty in the measurement. In general, there is no consistent
difference observed among the different samplers or methods. Table II-13c presents the averaged data
by method. The standard deviation of the average ratio at all four sites either exceed the average or
encompass 1, suggesting no difference can be distinguished for volatilized nitrate by these two methods.
On the other hand, the higher variability at Phoenix vs Philadelphia (similar concentration ranges) may
be due to how nitrate is measured and the amount of nitrate volatilized or in the gas phase as nitric acid.
Finally, to confirm that the differences are due to the method of collection, direct vs indirect, Teflon vs
quartz-fiber filters, data are presented by sampler in Table II-12d for sulfate as measured on either
Teflon or quartz-fiber filters. There is no difference between the use of the two filter types for
determination of aerosol sulfate. Table n-13d provides summary averages of the data in Table n-12d.
There is no difference between the two filter types, and the ratio at Rubidoux being different then one is
due to rounding errors, all ratios were within 5%.
Given the above discussion, there is an apparent positive bias when measuring nitrate on a quartz-fiber
filter that has been heat treated for carbon analysis. Therefore, it is recommended that a different
approach be used for the determination of particulate nitrate on those samplers that may use the quartz-
fiber filter in their determination of nitrate.
Part II, Page 24
-------
Organic and Elemental Carbon
Denuded vs Non-Denuded Organic Carbon Results
The PM2 5 Chemical Speciation Evaluation Study collected particulate carbon on quartz-fiber filters
from one sampler with an XAD denuder and five samplers without an XAD denuder. Data from
Phoenix were selected for this analysis because the Phoenix samples included an additional IMPROVE
sampler that used 25 mm quartz filters. Samples from the SASS were not included in this evaluation
due to potential carbonate contamination from soil, since the inlet allowed particles greater than 2.5 jim
to enter the inlet. The range of filter face velocities was from 11 to 104 cm/sec. Thermal Optical
Reflectance carbon concentrations were used to evaluate the effect of using the XAD denuder and to
investigate the differences in OC concentrations between the samplers.
Denuded and non-denuded samples were first compared using both the carbon fractions from TOR
and the mass concentration of OC, EC, and TC. Figure n-5a shows the average percent of OC for
OC1, OC2, OC3, OC4, pryrolized carbon; and the average mass concentrations for OC, EC, and TC
(EC + OC). Five samplers were used for the non-denuded average and 1 sampler was used for the
denuded sampler average. Seventeen 24-hour samples were used to calculate the average for each
sampler. Figure II-5b shows the average percentages and masses for the denuded sampler with a 15
liter per minute (1pm) flow rate (21.1 cm/sec face velocity) and the non-denuded samplers with 16.7
1pm flow rate (23.6 cm/sec face velocity). The OC1 mass percent was less for the denuded samples in
both Figures n-5a and II-5b. OC3 shows an increase for the denuded samples and little difference for
OC2 and OC4. Differences in the OC3 concentrations may be due to the shape of the cutpoint for the
VAPS compared to the other samplers. Organic carbon concentrations for the denuded sample were
less than the non-denuded sample and the EC concentrations were similar. The differences in the OC
concentrations are due to a combination of reduced positive artifact and a potential increase in the
negative artifact or stripping of volatile organic compounds from the collected particles.
The impact the denuder has on the collection of OC, under the conditions of this experiment, can be
seen in Figure n-6, where OC concentrations are plotted for the FRM sampler and the VAPS
samplers against PM2 5 mass. The VAPS sampler used a denuder in front of the quartz-fiber filter,
whereas the FRM sampler did not use a denuder. Except for Rubidoux, the regression lines of the two
samplers are parallel, but the FRM OC regression lines are offset high indicating a positive artifact in the
method relative to OC collected on a quartz-fiber filter behind a denuder. This assumes both filters
have similar negative artifacts. At Rubidoux, the VAPS sampler data likely represent PM10 rather then
PM2 5 and should not be considered valid data for OC or EC.
Figure n-6 also provides an estimate of the magnitude of the positive artifact for OC. Once again, at
zero PM2 5 mass the OC measured at Phoenix and Philadelphia by the FRM have intercepts of
between 1-3 |ig/m3, except at RTF, where it is closer to about 0.2 |ig/m3. OC by the VAPS sampler
has near zero intercepts, except at Rubidoux as just explained, suggesting that the use of the denuder
provides a more suitable estimate of OC, at least relative to what is measured on a Teflon filter, i.e.,
assumes similar negative artifacts, which are likely minimized since this was a wintertime study.
Differences RTF and the other cities may be due to the mix of OC compounds present in the
atmosphere during sampling, as RTF is a more rural location then the sites in Philadelphia or Phoenix.
Part II, Page 25
-------
Face-velocity effects were investigated using the ratio of the non-denuded to the denuded samples.
Figures U-7a and II-7b show the relationship between filter face velocity and the OC and EC
concentrations, respectively. Denuded OC1 to non-denuded OC1 ratios range from 9.7 at 11 cm/sec
to 4.8 at 104 cm/sec. The 95 percent confidence intervals show that the ratios have a fair amount of
scatter at each face velocity. EC2 and ECS concentrations show a increase that may be due to the
PM2 5 collection efficiency of the VAPS as compared to the RAAS. The effect efface velocity is
shown strongest in the lowest temperature carbon fraction determined by TOR (OC1). This fraction
may represent positive artifact that is due to SVOC that boil in the range of 120 C. Additional
experiments, like those conducted in Seattle, with back-up XAD impregnated quartz filters will help in
the interpretation of these data (Phase n of the evaluation of the chemical speciation samplers, Mitchell,
EPA, ORD, personal communication) .
Comparison between TOR and TOT for OC and EC in PM2 5
Carbon in atmospheric particles typically represents a large fraction (25 - 50%) of the PM2.5 mass.
Two methods are used currently for the analysis of particulate carbon: Thermal Optical Transmission
(TOT - NIOSH Method 5040) and Thermal Optical Reflectance (TOR). These two methods both
quantify carbon by heating filters and volatilizing the carbon that is oxidized in a granular bed of MnO2,
reduced to CH4 in a Ni methanator, and quantified as CF^ with a flame ionization detector. The TOT
and TOR methods use different temperature programs and use different techniques to correct for the
formation of pyrolysis products. These differences result in an operational definition of OC and EC by
each method and while total carbon values are typically the same, the split between OC and EC are
different, which usually results in a higher observed EC fraction by the TOR method. To better
understand this difference, the TOT and TOR methods were compared using samples from the
Chemical Speciation Monitor Evaluation Field Study. Additional wood smoke, and diesel exhaust
samples were also included in the evaluation. Appendix J provides a more detailed description of the
TOR and TOT methods comparison.
Both source and ambient wood smoke samples were evaluated. Thermal Optical Reflectance EC was
on average 175 percent greater for a triplicate analysis of the smoldering wood smoke source sample.
Measurement precision was similar for both techniques with a coefficient of variation (CV) of 21 and
23 percent for TOT and TOR, respectively. Spokane ambient wood smoke samples showed a lower
difference in the EC concentrations with a ratio of TOR EC to TOT EC of 1.21.
A diesel truck source sample also was evaluated in triplicate. The TOR method EC and TC were 30
and 7 percent higher than the TOT method, respectively. TOR TC was 7 percent higher than the TOT
TC. The CV for TOR EC (4 percent) was higher than the TOT EC (CV = 1 percent); however, both
CV values are small compared to the differences between the methods for EC. The diesel truck
exhaust was spiked with 10 |jL of Trona solution, a sodium carbonate ore containing both carbonate
and bicarbonate. The solution contained 10.16, 0.50, and 10.65 |j,g as C of OC, EC, and TC,
respectively. Neither analysis show an increase in the amount of EC measured, while the TOR method
showed a decrease in EC from 6.07 to 4.05 |j,g/cm2. The TOT technique quantifies CC as part of the
routine analysis and gave a value of 10.3 |j,g/cm2 for CC. TOR showed an increase in OC as opposed
to an increase in EC.
Part II, Page 26
-------
Samples were selected from the Phoenix (20) and Philadelphia (20) that represented a range of
loadings. Six days with a range of concentrations were selected from the co-located VAPS with an
XAD denuder, the IMPROVE sampler, and the Anderson RAAS. The VAPS, IMPROVE, and
RAAS sampled 21.6, 32.8, and 10.5 cubic meters of air through a 47 mm quartz-fiber filter,
respectively. The use of the different sample flow rates provided an average TC loading of 18.41
Hg/cm2 and wide range of TC loadings (4.77 to 41.78 |j,g/cm2). The average EC, OC, and TC
concentrations were significantly different (p < 0.001) when the TOR and TOT methods were
compared using a paired t-test and the Wilcoxon Signed Rank Test. Linear regressions were used to
evaluate the linear relationship between TOR and TOT in Phoenix and Philadelphia. Slopes for EC,
OC, and TC were not significantly different for each city, and the intercepts were not significantly
different from zero. Overall relationships between TOR and TOT (|j,g/cm2) were TOR OC = TOT OC
* (0.91 ± 0.02 (SE)) +(0.00 ± 0.35(SE)), r2= 0.98; TOR EC = TOT EC * (1.94 ± 0.08) - (0.22 ±
0.31), r2= 0.93; and TOR TC = TOT TC * (1.11 ± 0.02) - (0.07 ± 0.32), r2= 0.99.
Two conclusions can be made based on this methods comparison: [TOR]EC > [TOT]EC, and [TOR]TC
> [TOT]TC These conclusions are based on source and urban samples. Additional samples from
non-urban locations need to be evaluated to determine if the association is the same for urban motor-
vehicle dominated and non-urban biogenic/secondary aerosol samples.
Loss of Nitrate During Vacuum XRF Analysis
The analysis protocol for the FRM, URG, and VAPS samplers require mass, XRF, and ions analysis to
be performed on the same Teflon filter. Due to the potential loss of ammonium nitrate from filters
during vacuum XRF, nitrate, sulfate, and ammonium were determined on the quartz-fiber filter in these
samplers. However, the observed positive artifact noted above with the quartz-fiber filter for nitrate
required us to re-examine the use of the Teflon filter that has been used for XRF analysis in those
samplers for ions analysis. Paired Teflon filters (same fow rate, and usually in parallel lines in the same
sampler) were chosen for this analysis. One filter was analyzed directly for nitrate and sulfate, the
other, having been analyzed for atmospheric XRF was analyzed also by vacuum XRF followed by
determination of nitrate and sulfate on that filter. Another set of filters, analyzed only by atmospheric
pressure XRF, also were analyzed for nitrate and sulfate to see the effect of just atmospheric pressure
XRF on the potential loss of nitrate. The latter data are not yet available.
Table n-14 presents data for the loss of nitrate from Teflon filters as a result of XRF analysis. On the
average, about 40 ± 16% of the nitrate was lost during XRF analysis. Also presented in the table are
results for sulfate on the same filters. No sulfate was lost during the analysis. Figure 1-8 is a scatter plot
of the nitrate loss data as a function of the original nitrate concentrations. Data below original nitrate
levels of 500 ug/filter are grouped along one regression line with a correlation coefficient of 0.851. On
the other hand, data above 500 ug/filter fall well below the less than 500 ug/filter regression line. These
higher concentration data points may have lost less nitrate because of the nitrate present on the filter
may have been non-volatile (NaNO3, or metal(NO3)x}, rather then NFLNO3 (e.g., Eldering et al.
1991). Two of the three data points occurred on January 18th when Ca and Fe were near their highest
levels during the study. Sodium values were not measured.
Part II, Page 27
-------
These data indicate that filters used for vacuum XRF analysis can not be used for nitrate analysis;
although they can be used for sulfate analysis. One would also assume that the ammonium ion results
i.e., ammonium associated with NEtjNC^, also would be biased low as a result of vacuum XRF.
Field Experience with the Speciation Samplers
The overall objective of the chemical speciation sampler evaluation study was to characterize their field
performance under a range of atmospheric chemical, meteorological, and operating conditions. While
the primary evaluation criteria used to judge sampler performance was each sampler's ability to
representatively collect and provide equivalent chemically speciated atmospheric fine paniculate matter,
each sampler's ease of use, reliability, and field worthiness also were evaluated. For this reason, site
operators were instructed to record their daily experience with sampler setup and routine operation. At
the completion of the field sampling component of the overall study, a four-page field operations survey
was developed and distributed to allow site operators to document their overall experience with each of
the samplers. Evaluation criteria included ease of sampler installation, audits and calibrations,
programming, filter holder exchange, sampler maintenance, and data retrieval procedures.
Review of the daily site operations records and completed surveys revealed problems with each of the
speciation samplers as well as the collocated PM2 5 FRM samplers. Some problems were particular to
individual instruments of a given design at a given site and were typically associated with sampler
construction, handling, setup, or operation, rather than sampler design. Other problems were
associated with design, and thus, all samplers of that type were affected.
A summary of the site operator's surveys is presented in Table 11-15, while Table II-4 details noted
problems or deficiencies with each sampler as recorded from the daily field data sheets. A descriptive
summary of each sampler's field performance is listed below.
Andersen RAAS
The Andersen sampler was judged to be relatively straightforward to unpack and setup in the field;
although one site operator indicted that its main module weighed more than other sampling modules.
The ease of setting up the sampler in the field was judged to be good. The sampler is capable of being
temperature, pressure, and flow calibrated in the field and calibration procedures were judged to be
average to good. Good to excellent ratings were received for conducting the manufacturer's
temperature, pressure, and flow audit procedures. Few problems were noted during programming or
retrieving pertinent data after each sampling event.
At the Rubidoux site, the Andersen sampler lost the temperature, pressure, and flow calibrations on
two occasions requiring complete re-calibration of the instrument. The Rubidoux site operator also
noted a slight water intrusion problem following a rain event and indicated occasional drifts in reported
internal cabinet temperature. Two particular design problems were identified by each of the five site
operators. First, the sampler's indicated flow rate was difficult to calibrate due to electronic drifts in the
flow sensor response. Fluctuations in the indicated flow rate of each Andersen channel were observed
throughout the study at each of the four sites and caused uncertainties in overall volume sampled. The
Part II, Page 28
-------
second design problem observed by all site operators was the tendency for o-ring seals to fall out
during installation and removal of filter holders. This latter problems made routine filter holder exchange
awkward and raised concerns regarding potential sample contamination.
Met One SASS
The Met One speciation sampler was judged to be relatively easy to setup in the field and could be
accomplished by one person, with an overall rating for installation of good to excellent. The sampler's
response to ambient temperature, pressure, and flow rate only can be calibrated by the factory and
cannot be adjusted in the field. The manufacturer's procedures for auditing these quantities was rated
as good to excellent. Few problems were noted during exchange of the sample canisters in the field.
No problems were noted with post-sampling data retrieval.
Some problems were encountered during laboratory disassembly of the Met One canisters.
Specifically, collected filter samples were sometimes difficult to remove without damaging or
contaminating the sample. Occasional field problems were noted regarding failure to start as
programmed or with sampler premature shut-downs. These events were intermittent and varied by
sampling site. One operator experienced sharp edges on the sampler's upper and lower inlet shroud.
The sampler's screen was noted to be difficult to read in bright sunlight.
URG MASS
The URG sampler generally received favorable ratings regarding ease of field installation and initial
setup. Ambient temperature, pressure, and flow rate response for the URG sampler can be calibrated
in the field and the manufacturer's instructions for conducting these calibrations received average to
good operator responses. Programming and its reliability received excellent ratings. Responses
regarding sample exchange and data retrieval were rated as good.
The high pressure drop though the MASS 400 filter holder system exceeded the ability of the unit's
flow control system and resulted in low initial flow rates at all sites. Addition of a second air pump to
each flow system by the manufacturer effectively resolved the flow performance problem. Filter holder
temperature flags were often encountered at the Rubidoux and Phoenix sampling sites. Contact of
threaded Teflon components with threaded aluminum components in the MASS 400 resulted in
damage to Teflon filter holders.
IMPROVE
Site operators noted that setup of the IMPROVE sampler was generally more cumbersome and
required more time than the other speciation samplers. Unlike most of the other speciation samplers,
the IMPROVE does not provide ambient temperature or pressure measurement. Calibration of system
flow rates in the field were judged to be average to good and sampler programming was judged to be
average. Ease of exchanging filter holders in the field was rated as average to good.
Part II, Page 29
-------
Field reliability problems with the IMPROVE sampler were experienced at all four sites with most
problems relating to faulty timers, solenoids, and pumps. A slight water intrusion problem was noted at
RTF following two significant rain events.
VAPS
Installation and setup of the VAPS sampler was judged as average by operators at each site. Like the
IMPROVE sampler, the VAPS provides no measurement of ambient temperature and pressure, so
these parameters could not be calibrated nor audited in the field. The ease of the flow calibration and
flow audit procedures were both rated as average. Fair to average ratings were received regarding
exchanging of field samples and ease of data retrieval from the sampler.
Problems encountered with the VAPS were typically ones of programming and failure of the sampler to
properly shut down at the specified time. All site operators experienced problems with o-rings and
virtual impactor receiving tubes falling out during filter holder installation and removal.
FRM
Three different FRM models were used during the Four City study so comments received are primarily
sampler-specific. In general, however, the FRM samplers were judged to be relatively straightforward
to install and setup in the field. Manufacturer's procedures for calibration and auditing of temperature,
pressure, and flow rate sensors were generally judged to be good. Sampler programming and its
reliability were rated as good to excellent. Data retrieval from each sampler was judged to be good.
Specific problems encountered with the FRM were normally sampler specific and included problems
with pump failures, water intrusion, and failure to initiate the programmed sampling event.
Sampler Costs and Related Spare Parts
Table n-16 lists each sampler used in this study and parts needed to operate the samplers on a 1 in 6
day schedule. To operate the samplers on a more frequent schedule would require multiples of each
based on shipping and filter changing schedules. Operational costs for each of the speciation samplers
was considered to be essentially equivalent.
Part II, Page 30
-------
DISCUSSION
Expectations
The speciation samplers were designed to minimize sampling artifacts during the collection of
atmospheric paniculate matter, and thus, provide an improved estimate of the PM2.5 actually present in
the atmosphere, relative to the FRM. As stated in the introduction, for stable species, such as, sulfate
and the trace elements measured here, we would expect good agreement (approaching the precision
for that species) between the FRM and the chemical speciation samplers. This of course assumes the
inlets of the samplers are performing in a manner similar to the FRM. This was recommended highly by
the PM Expert Panel (Koutrakis, 1998).
For aerosol species in dynamic equilibrium with their gas phase counter parts (i.e., NB^NOs and semi-
volatile organic compounds) we would expect agreement between the FRM and the speciation
samplers under some conditions, but not under others. This is because these equilibrium processes are
dependent on temperature, relative humidity, pressure drop across the collection substrate
(proportional to flow rate with filters of the same diameter), and the concentration of the precursor
gases. The latter is especially true in the case for Nf^NOs in equilibrium with nitric acid and ammonia
(Russell et al. 1993, Hering and Cass, 2000), as they need to react to form aerosol, as opposed to just
condensing as might be the case for many of the SVOC. Since diffusion denuders and reactive
sorbents were not used to remove gas phase species or to efficiently collect SVOC, respectively, we
would expect there to be differences among the samplers for the measurement of OC as a function of
pressure drop across the collection substrate and other variables. On the hand, denuders and reactive
filters were used to collect NEtjNOs, and therefore, we would expect NFySTOs to agree among the
chemical speciation samplers for particulate nitrate, but perhaps not with the FRM depending on the
conditions stated above.
Before discussing differences among methods it also is important to review the precision results (Table
n-2c). Estimated precision, calculated as the coefficient of variation between collocated sampler pairs
was within 10 percent for most of the samplers for mass, sulfur, sulfate, ammonium, and potassium.
Precision values estimated for the other species, independent of sampler, were within 15 percent,
except for OC and EC by the IMPROVE sampler and silicon on all samplers. The latter 3 species had
measured precision values greater then 15 percent but usually less then 30 percent. The precision
results therefore, provide a lower bound for agreement between the speciation samplers and the FRM
and among the speciation samplers.
Major Questions Addressed
As described in the introduction, the primary objective of this study is to determine if there are
differences among the three chemical speciation monitors for their ability to estimate concentrations of
the chemical components of PM2 5 mass found in ambient air. Comparisons also will be made to two
historical samplers and to the FRM using these samplers as a relative reference.
Part II, Page 31
-------
There are four questions to be addressed by this intercomparison study with associated hypotheses
designed to address the questions. The questions and hypotheses are stated below with responses
based on the data provided from this study.
Ql. How well do PM2 5 mass and the chemical components of mass agree between the
FRM and the chemical speciation samplers tested in this study?
There is generally good agreement between the FRM and the chemical speciation samplers for the
measurement of PM.2.5 mass and it chemical components, under the conditions encountered in this
study, including meteorological and operating conditions. There are a few exceptions, due to
differences in sampler inlets or efficiency of the PM2 5 cutpoint. As well, real differences were observed
for nitrate and organic carbon. For nitrate the differences appear to be due to the use of pre-heated
quartz-fiber filters, and the believed associated positive artifact for nitrate with that filter material. For
organic carbon the differences appear to be due to the different face velocities across the quartz-fiber
filters used among the samplers. Tables n-8a lists average values for each species and sampler by site,
while Table II-8b lists the ratio of the speciation sampler to the FRM for each chemical component by
site and sampler. For example, there is excellent agreement for sulfate, suggesting that the quality
assurance and quality control objectives were achieved in monitoring and validating flow rates and for
chemical analysis, assuming sulfate results are representative of other species. PM2 5 mass agree
extremely well (with in 5%) at Philadelphia and RTF for all samplers and at Rubidoux and Phoenix for
the Andersen and IMPROVE samplers. More variation was observed for the URG and MetOne as
explained earlier.
There were individual data points or small groups of data that effectively biased the analysis (see
Figures II-3 and II-4). These data have been carefully reviewed and have been validated to Level 2b.
Any remaining inconsistencies in the data have been included in the analyses presented here as valid
data.
The spiral inlet of the MetOne Sampler produced results that were biased high relative to the FRM.
The data indicated (e.g., see Tables 8a and 8b) that coarse particles were passing through the inlet to
the collection substrate, especially at high coarse particle loadings. This problem has been addressed
by MetOne and the spiral inlet has been replaced with a Sharp Cut Cyclone (SCC). Results of
preliminary evaluations performed by MetOne of the MetOne sampler with the SCC are presented in
Appendix H. Results presented in Appendix H indicate much better agreement with the FRM than the
spiral impactor. Subsequently, the spiral inlet has been evaluated with test aerosol in the laboratory and
compared to the WINS impactor of the FRM and SCC now used in the MetOne sampler (Peters et al.
2000; given in Appendix I)
More variation was observed among the chemical speciation samplers relative to the FRM for organic
carbon than the other major species. Organic denuders and reactive sorbent collectors were not used
in this study to minimize negative and positive artifacts in the collection of organic carbon. If the method
for collecting OC was identical among the samplers, artifacts would not be an important factor, at least
for noting differences among the samplers. However, the samplers operated at different flow rates and
used the same size filter for organic carbon, except MetOne, which used a mask over the quartz-fiber
filter to achieve the same face velocity as the FRM. Therefore, variations due to pressure drop across
Part II, Page 32
-------
or face velocity through the filter were observed (see Figures n-7a and n-7b) and resulted in significant
variations for measured OC concentrations among the samplers tested in this study (e.g., see Figures
II-3e and II-4e).
More variation was observed among the nitrate at Phoenix then at Philadelphia, even though similar
nitrate concentrations were observed at both sites. This variation was partly attributed to the use of
pre-heated quartz-fiber filters originally designed for collecting OC and EC for subsequent analysis and
temperature differences (Philadelphia was colder). On an absolute basis, variations in nitrate are similar
at Philadelphia, RTF, and Rubidoux and on the order of 1 - 3 |ig/m3. Ammonium follows a similar
pattern to nitrate.
Variation among the chemical speciation samplers relative to the FRM for trace elements was varied.
Sulfur had excellent agreement, typically within 10%, while silicon varied up to a factor of 5 or more,
including the MetOne sampler. Excluding the MetOne sampler most differences were within 20-25%,
with the highest variations for all samplers being observed at Phoenix, where the highest coarse particle
loadings were observed. Similar results were observed for Ca and Fe, species typically associated
with the coarse particle size fraction.
Q2. How well can the FRM mass be reconstructed by summing the chemical components
measured by the speciation samplers.
Mass balance results only were calculated for the FRM sampler. Mass and trace elements were
measured on the Teflon filter, while ions and OC/EC were measured on the quartz-fiber filter as
indicated in Figure I-6a. In calculating the mass balance, Si, Fe, and Ca were converted to their oxides
and summed to give an estimate of crustal material and organic carbon was multiplied by 1.4 to give an
estimate of organic material. Other species were used directly. The sum of the species as defined
above accounted for 111, 135, 110, and 108 percent of the mass on average at Rubidoux, Phoenix,
Philadelphia, and RTF. Since mass is measured on an inert Teflon filter, where positive artifacts would
be minimized, it is likely that this overestimation in calculated mass is due to positive artifacts for OC
and nitrate on the quartz-fiber filter. Using data in Table n-13a (Q-T) to represent the positive artifact
for OC and regression intercepts from Figure II-6 the estimated corrected mass balance adjusted for
potential artifact for both species is 94%, 112%, 101%, and 104% at Rubidoux, Phoenix, Philadelphia,
and RTF, respectively. These values are more reasonable, however, they are still lower estimates since
Al, Ti, and Mg oxides have not been accounted for in the crustal material estimate (likely less than 1%
addition) and water has not been estimated and included. However, past experience suggests that
mass balance results of 100±20% is reasonable.
For the speciation samplers one would expect similar results, as there was reasonably good agreement
among the speciation samplers and the FRM for mass and its components, as discussed above. As
well, the FRM is likely a reasonable surrogate for the speciation samplers because 1) this was a
wintertime study and artifacts due to volatilization would be minimized, 2) the positive artifact observed
for nitrate, as measured on the quartz-fiber filter, resulted in similar particle nitrate concentrations
between the FRM and the speciation samplers, and 3) the fact that OC was measured by the FRM at
the standard face velocity of 16.71pm. The major difference between a calculated mass balance for the
speciation samplers relative to the FRM would be due to organic material as the OC measurement
Part II, Page 33
-------
continues to have both positive and negative artifacts that are a function of the sampler design
parameters (e.g., face velocity through the filter) and for nitrate for the samplers where paniculate
nitrate includes nitrate measured on a quartz-fiber filter.
Specific Hypotheses Related to Questions Ql and Q2
The following hypotheses are related to reconstructing the measured FRM mass. For each of these
hypotheses, the concentrations, on a species-by-species basis, from each of the speciation samplers is
compared to the respective concentration from the FRM. The concentrations from the speciation
samplers are not compared to each other. The specific hypotheses and brief statements regarding
those hypotheses are given below:
1. PM2 5 mass concentrations on the FRM Teflon filter measured by the gravimetric method are
compared to PM2 5 mass concentrations as measured on the Teflon filters in the chemical
speciation samplers by the gravimetric method. The hypothesis is that the mass
concentration from each of the speciation samplers is not statistically different from the
mass on the FRM filter. The next eight hypotheses delve further into understanding why the
mass concentrations do or do not compare favorably.
Tables n-10a through II-10e provide results of the paired t-test for all samplers and all species
measured in this study. Mass concentrations obtained by the speciation samplers are equivalent
for the Andersen sampler at all sites, while the other samplers are equivalent to the FRM mass
at some sites, but not others. However, while there are statistical differences at alpha equal to
0.05, the differences are small in absolute concentrations, for samplers operating properly, and
are within expectations based on the precision obtained from collocated samplers (Table n-
2c). The reason for these inconsistencies is due likely to differences in the inlet collection
efficiency (slope and cutpoint) of the samplers for PM2 5. The only unexplained exception is
mass measured by the URG sampler at Rubidoux. This sampler reported an average mass
concentration that was more then 30 percent greater then that report by the FRM (see Tables
n-8a and n-8b). The major species are high on the URG sampler relative to the other
samplers at Rubidoux (excepting the MetOne), as opposed to having a few data points that are
at an extreme value (see Figures II-3 and II-4). Trace elements associated with crustal material
are not high on the URG sampler relative to the other samplers at Rubidoux and these were
measured on the same Teflon filter as the mass. In addition, comparable results among the
URG sampler and the other samplers were observed at Phoenix, where coarse particle
concentrations were similar to Rubidoux, therefore, it is likely that the relatively high mass
concentrations observed at Rubidoux for the URG sampler were a problem only with that
particular URG sampler and the composition data do not provide the answer as they did for the
MetOne sampler.
Part II, Page 34
-------
2. PM2 5 trace element concentrations or groups of elements on the FRM Teflon filter as
determined by XRF are compared to trace element concentrations or groups of elements on
the Teflon filters in the speciation samplers as determined by XRF. The hypothesis is that
there is no statistical difference between the concentrations from each of the speciation
samplers to that from the FRM. These concentrations should be comparable since these
species are stable.
Statistical differences were found for all trace elements (Si, K, Ca, Fe, and Zn) at most
locations for the MetOne sampler as expected due to the inlet problem. The other samplers
had mixed results. However, even though the samplers were not equivalent (at alpha = 0.05)
to the FRM, in all cases, the differences on the average are usually small in absolute
concentrations (see Table n-8a) (except for the MetOne) and are reasonable based on the
precision obtained from collocated samplers (Table n-2c).
3. PM2 5 sulfate concentrations on the FRM quartz-fiber filter are compared to sulfate
concentrations on the Teflon (RAAS, SASS), quartz-fiber (MASS, VAPS), or nylon (IMPROVE)
filters in the speciation samplers. The hypothesis is that there is no difference between the
concentrations from each of the speciation samplers to that from the FRM. These
concentrations should be comparable since sulfate is a stable species.
Statistical differences were found for sulfate for all samplers at one or two of the sites; although
there is no consistent pattern among the sites and samplers. However, the differences on the
average are usually small in absolute concentrations (see Table n-8a) and relative to the FRM
(see Table n-8b) and are reasonable based on the precision obtained from collocated samplers
(Table II-2c).
4. PM2.5 ammonium concentrations on the FRM quartz-fiber filter are compared to
ammonium ion concentrations on the Teflon (RAAS, SASS), quartz-fiber (MASS, VAPS), or nylon
(IMPROVE) filters in the speciation samplers. The hypothesis is that there is no difference
between the concentrations from each of the speciation samplers to that from the FRM. These
concentrations may not be comparable since ammonium is volatile when in the form of
ammonium nitrate.
While manufacturers went to great trouble to develop samplers that minimized sampling artifacts
for nitrate, no designs included collecting ammonium with minimal artifacts. Therefore, one
would expect ammonium to agree with the FRM. However, the t-test (alpha = 0.05) indicated
significant differences at most sites for the Andersen, IMPROVE, and URG; the MetOne
sampler is statistically different only at Phoenix. For the collection of ammonium, based on the
averages (see Table n-8a) and the data presented in Tables U-lOa-e, there appears to be a
trend with quartz-fiber filters reporting the highest ammonium concentrations, then Teflon, with
nylon filters reporting the lowest averages and having the largest differences relative to the
FRM. In some cases, these differences are as great as 50% relative to the FRM ammonium
concentrations; however, relative to PM2 5 mass the differences are small. It is possible that the
IMPROVE sampler reports lower ammonium concentrations than the others due to enhanced
volatilization of ammonium from the basic nylon filter, i.e., NFySTOs that volatilizes after
collection on the nylon filter efficiently collects the HNO3, but not the NH^
Part II, Page 35
-------
5. PM2.5 nitrate concentrations on the FRM quartz-fiber filter are compared to nitrate
concentrations on the Teflon (RAAS, SASS), quartz-fiber (MASS, VAPS), or nylon (IMPROVE)
filters in the speciation samplers. The hypotheses is that there is no difference between the
concentrations from each of the speciation samplers to that from the FRM. These
concentrations should be comparable since the collection of nitrate on these filters should all be
biased by the loss of nitrate due to volatilization, with the exception of the IMPROVE.
The data presented in Table II-13 a, indicate a difference, on average, of up to 1.0 jig/m3 for
nitrate measured on the Teflon filter relative to the quartz-fiber filter with the latter being
consistently higher, under the conditions observed during this study. The difference is likely due
to a positive artifact associated with the quartz-fiber filter, which was pre-treated for carbon
analysis, i.e., heated to reduce carbon levels on the filer. This assumes that quartz-fiber and
Teflon filter media have similar losses due to volatilization, which is apparently true for quartz-
fiber filters that have not been pre-heated (Chow 1995). Nitrate concentrations obtained from
the nylon filter on the IMPROVE sampler were not included in the analysis.
6. PM2 5 nitrate concentrations on the FRM quartz-fiber filter are compared to nitrate
concentrations on the nylon (RAAS, SASS, IMPROVE), or quartz+nylon (MASS, VAPS) filters in
the speciation samplers. The hypothesis is that the nitrate from each of the speciation samplers
is greater than or equal to the nitrate on the FRM quartz-fiber filter, due to negative artifacts
anticipated collecting nitrate on a quartz-fiber filter in the FRM. This addresses questions about
the amount of nitrate volatilized from the FRM Teflon filter.
The paired t-test indicated statistically equivalent results for particle nitrate between the
MetOne sampler and the FRM, while mixed results were observed at the other sites for the
other samplers. The URG samplers reported consistently high nitrate concentrations relative to
the FRM as well as the VAPS sampler (see Tables n-10a - II-10e), mentioned here because
the VAPS and URG used similar methods for measuring nitrate. Table II-13b compares the
two methods for particulate nitrate. On the average, the direct method (nitrate measured on a
nylon filter directly behind a denuder) is consistently lower then the indirect method (requires
the sum of two filters, a nylon behind a denuder and Teflon filter and a quartz-fiber filter in
parallel) for particulate nitrate. The absolute differences are larger than the difference due to
using the quartz-fiber filter, suggesting another reason for the difference, other then just the
positive artifact observed on the quartz-fiber filter for nitrate.
Volatilized nitrate was a low fraction (on average < 10%) of the total nitrate at all sites except
Phoenix, where it represented about 50% of the total nitrate on average. Absolute values of
volatilized nitrate were less than 0.5 |ig/m3 at Rubidoux, Philadelphia, and RTF, and around 1.3
jig/m3 at Phoenix. On the average, differences between the FRM, Andersen, IMPROVE, and
MetOne were typically about 10%, with larger differences observed between these samplers
and the URG and VAPS. Differences between the direct and indirect methods for determining
volatilized nitrate were on the average, small (< 0.2 |ig/m3) (see Table II-13c); although, on
average, the direct measurement was higher than the indirect at all sites.
Part II, Page 36
-------
7. PM2,5 elemental carbon (EC) concentrations on the FRM quartz-fiber filter are compared
to EC concentrations on the quartz-fiber filters from the chemical speciation samplers. The
hypothesis is that there is no difference between the concentrations from each of the speciation
samplers to that from the FRM. These concentrations should be comparable since elemental
carbon is stable.
Mixed t-test results were observed for EC across the sites and samplers as seen in Tables II-
lOa-e. In general, the FRM reported higher EC concentrations then the speciation samplers as
indicated in Tables II-10a-10e and Table n-8b. However, the differences were usually small
(10-20%) on an absolute basis and reasonable based on the precision results obtained by
collocated sampling at Rubidoux (see Table H-2c).
8. PM2,5 organic carbon (OC) concentrations on the FRM quartz-fiber filter are compared
to OC concentrations on the quartz-fiber (MASS, RAAS, SASS, IMPROVE) filters in the
speciation samplers. The hypothesis is that there is no difference between the concentrations
from each of the speciation samplers to that of the FRM. These concentrations should be
comparable since as designed they all potentially experience similar negative or positive
artifacts.
Statistically different OC concentrations relative to the FRM were reported for the Andersen,
IMPROVE, and URG sampler at nearly all sites. The MetOne sampler was only different from
the FRM at RTF. Differences and variability were larger for the OC data then for many of the
other species. As discussed earlier, the measurement of OC by the speciation samplers also is
dependent on the face velocity of air being pulled through the filter (see Organic and Elemental
Carbon; Denuded vs Non-Denuded Organic Carbon Results). This resulted in larger (10-
30%) differences between the averages for OC measured by the speciation samplers and the
FRM then for the other major species. The scatter is illustrated in Figures II-3e and H-4e. The
face velocity effects can be seen in Figures n-7a and n-7b.
9. PM2.5 OC concentrations on the FRM quartz-fiber filter are compared to OC
concentrations on the quartz filter following an oversized XAD denuderfrom the VAPS
speciation sampler and from channel 5 on the SASS sampler. The hypothesis is that OC on the
FRM will be greater than OC collected by the VAPS or SASS denuded channels if there are
positive artifacts or less than the VAPS and SASS denuded channels if there are negative
artifacts. No difference would be inconclusive.
The SASS denuder was never installed in the samplers, as it was found to be to brittle. A
modified honeycomb denuder will be used in Atlanta. Paired t-test analysis was not performed
for this comparison. However, it is evident from the other analyses that lower OC
concentrations were measured using the VAPS with the XAD diffusion denuder then without
the denuder for OC concentrations measured on all other samplers. Results suggest (see Figure
n-6) that the VAPS denuder is at least eliminating partially the positive OC artifact OC
positive artifacts, relative to the FRM appear to account for up to 3 jig/m3 under the conditions
observed during this study. No information is available on OC negative artifacts from the
experiments conducted here.
Part II, Page 37
-------
Q3. How well do the measured concentrations from the various speciation samplers agree?
On the average, the chemical speciation samplers produced results for mass and the chemical
components of mass that agree within reasonable expectations based on the precision estimates
obtained from collocated samplers, that is, to within 10-30 percent depending on the species.
However, real differences exist among the samplers for organic carbon and nitrate. Removing the
biases associated with those measurements will provide even better agreement for those two species.
Paired t-test analysis supports this finding on the average. Differences greater then 30 percent were
observed in some cases for the trace elements, which may be important for receptor modeling, but not
for reconstructing PM2 5 mass. A number of exceptions have been noted and reasons for those
exceptions are described in the previous sections.
Q4. What are the causes of the differences among the speciation samplers for measured
concentrations of mass and the components of mass if they exist.
This has been explained in the previous sections, with regard to mass, nitrate, and organic carbon.
However, there are subtle differences among the samplers that would be beneficial to explain. As well,
the samplers are all designed slightly different and it would be useful to examine more carefully those
differences on their ability to obtain reliable and consistent chemical speciation data.
Specific Hypotheses Related to Questions Q3 and Q4
The following hypotheses compare the concentrations of the chemical components measured on the
chemical speciation samplers to each other to address why there are differences among measured
concentrations, if they exist. For each of these hypotheses, only the concentrations from the speciation
samplers are compared, excluding the FRM, except for the first hypothesis, which will include the FRM
in the analysis.
The following hypotheses were addressed in the above section when comparing the speciation samplers
to the FRM. There are subtle differences, i.e., speciation samplers show more variability among
themselves for nitrate and carbon, than relative to the FRM. This is typically due to the samplers
showing a consistent bias rather than a random bias relative to the FRM, i.e., one sampler is usually
high relative to the FRM, while another is usually low. Once again, from a practical standpoint and for
most species, the differences are usually small and reasonable given the measured precision estimates
obtained from collocated sampling. However, there are real differences observed for organic carbon
and nitrate that need to be addressed to reduce the variability among the samplers as differences as high
as 1-3 |ig/m3 are observed on average for nitrate and 1-4 |ig/m3 are observed for organic carbon.
1. PM2 5 mass and chemical composition as determined according to the manufacturer's
guidelines. The hypothesis is that all species of interest are comparable among the chemical
speciation samplers when concentrations are determined according to manufacturer's
guidelines.
As noted above, on the average the concentrations of the species measured by the chemical
speciation samplers when operated according to the manufacturer's specifications agree within
reasonably expectations for most species, based on the observed precision estimates obtained
from collocated sampling. Organic carbon showed more variability among the samplers, and
relative to the FRM, two samplers tended to be high while two tended to report lower OC
concentrations. Thus, differences among the samplers for OC is somewhat larger then when
Part II, Page 38
-------
comparing their values to the FRM. This variation can be seen in Figure II-3e and II-4e.
These differences are in part due to a combination of positive and negative artifacts that have
been shown to occur during sampling and perhaps sample storage (Eatough et al. 1989;
McDow and Huntzicker 1990; Turpin et al. 1994).
Results shown in Figure U-7a and II-7b indicate the significant impact that face velocity through
the filter or pressure drop across the filter has on the measured OC concentrations. Differences
among the chemical speciation samplers also were noted for aerosol nitrate and were most
pronounced when volatilized nitrate was a large fraction of the total nitrate (i.e., about 50% vs
10% in this study). These differences resulted in a consistent bias among the speciation
samplers relative to the FRM and to each other with the URG and VAPS sampler typically
reporting nitrate concentrations higher then the FRM and the Andersen and MetOne typically
reporting concentrations lower then the FRM. It is believed these differences are due to the
methods used to collect aerosol nitrate. Two factor have been identified. First, the use of the
pre-heated quartz-fiber filter in the URG, VAPS, and FRM samplers and its associated positive
bias, and secondly, the direct vs indirect methods for collecting particulate nitrate.
However, based on the data presented here and the fact that there are no reference standards
for OC or nitrate, we are unable to indicated which sampler is providing more accurate results.
However, the data do indicate real differences between organic carbon and nitrate
concentrations among the samplers, both most likely due to positive artifacts for their respective
species.
2. PM2,5 mass concentrations by gravimetric method on Teflon filters are compared among
the samplers. The hypothesis is that there is no difference in these masses.
On average, when the samplers are operating correctly, they appear to be reporting similar
PM2 5 mass concentrations. For MetOne this condition is based on results obtained for that
sampler using the sharp cut cyclone, rather then its original design tested here using the spiral
inlet (see Appendix H). The one exception to this is the URG sampler at Rubidoux.
Examination of the composition data failed to reveal the cause of this difference.
3. PM2 5 elements (individually or in groups) from the Teflon filters as measured by XRF are
compared among the samplers. The hypothesis is that there is no difference in these
concentrations as these elements are stable during sampling and analysis.
In general, there was good agreement among the samplers for trace elements, with the
exception of MetOne, due to the spiral inlet apparently allowing coarse particles to penetrate to
the filter. More information is given above in No. 2 under Specific Hypotheses Related to
Questions Ql and Q2.
4. PM2 5 nitrate concentrations from the nylon (RAAS, SASS, IMPROVE) or Teflon+nylon
(MASS, VAPS) filters are compared among the samplers. The hypothesis is that there is no
difference in these concentrations.
Part II, Page 39
-------
Real differences are seen among the speciation samplers for particle nitrate. The differences
appear to be due to the use of pre-heated quartz-fiber filters, used for organic carbon analysis,
and differences among how the denuder/filter pack arrangements are used in the samplers.
More detail is given above in No. 6 under Specific Hypotheses Related to Questions Ql and
Q2 and No. 1 under Specific Hypotheses Related to Questions Q3 and Q4.
5. PM2_S sulfatefrom the Teflon (RAAS, SASS), quartz (MASS, VAPS), or nylon (IMPROVE)
filters are compared among the samplers. The hypothesis is that there is no difference in sulfate
concentrations as sulfate is stable during sampling and analysis. We will also look at sulfate
estimated from XRF sulfur (SX 3 equal to sulfate) to see how well XRF sulfur estimates sulfur
determined by 1C from both Teflon or quartz filters.
As discussed above, no differences are observed for measured concentrations of sulfate among
the samplers. More details are given above in No. 3 under Specific Hypotheses Related to
Questions Ql and Q2.
Comparison of sulfur by XRF times three (S*3) to sulfate by 1C showed excellent agreement
among all samplers. Ratios of S*3/Sulfate were 1.08 ± 0.13; 1.06 ± 0.10; 0.99 ± 0.08; and
1.06 ± 0.05 at Rubidoux, Phoenix, Philadelphia, and RTF, respectively. These data are
indistinguishable from a ratio of 1, thus, XRF sulfur could be used to estimate sulfate, under the
conditions that occurred during this study.
6. PM2 5 organic and elemental carbon from the quartz-fiber filters where no denuder is
used will be examined relative to each other. Since channel 5 of the VAPS uses a denuder, it will
not be included in this hypothesis. The hypothesis is that the concentrations ofOC and EC are
the same provided no denuder is employed.
As discussed above, differences in face velocity across the filter (or pressure drop through the
filter) resulted in real differences (from 1-4 |ig/m3 on average) among the measured OC
concentrations for these samplers. More details are given in Nos. 7 and 8 under Specific
Hypotheses Related to Questions Ql and Q2 and No. 1 under Specific Hypotheses Related to
Questions Q3 and Q4.
7. PM2.5 organic and elemental carbon from quartz filters including speciation samplers
where a denuder is used is compared among the samplers. The hypothesis is that samplers that
use denuders (i.e., the VAPS) will have lower OC concentrations since the potential for positive
artifact due to organic vapors has been minimized. Nothing can be said about negative
artifacts. EC should not be affected.
As described above, the OC measured by the VAPS sampler behind the XAD denuder was
consistently lower then the other samplers indicating removal of positive artifact from the air
stream. Estimates of this positive artifact were obtained relative to PM2 5 mass measured by
the FRM (also see Tolocka et al. 2000). Positive artifact for OC ranged from about none at
RTF to about 3 |ig/m3 at Phoenix. Artifact at Rubidoux could not be estimated due to the
operational problem with the VAPS sampler at that site. More information is given above in
Part II, Page 40
-------
No. 9 under Specific Hypotheses Related to Questions Ql and Q2 and No. 1 under Specific
Hypotheses Related to Questions Q3 and Q4.
8. PM2,5 ammonium from the Teflon (RAAS, SASS), quartz (MASS, VAPS), or nylon
(IMPROVE) filters are compared among the samplers. The hypothesis is that the concentrations
of ammonium are the same for all speciation samplers as all potentially suffer from negative
artifacts.
Measurement of ammonium ion in the speciation samplers did not include a method to minimize
artifact due to volatilization of ammonium nitrate. Variability among the samplers was on the
order of 0.5 |ig/m3 or less. In general, the IMPROVE sampler was slightly lower then the other
samplers, likely due to the use of a nylon (basic pH) filter to collect a basic gas (NHs after
dissociation of NH^NOs), while HNO3 would be collected efficiency by the nylon filter. More
detail is given above in No. 4 under Specific Hypotheses Related to Questions Qla and Qlb.
9. Ammonium nitrate equilibrium is a function of temperature and relative humidity and
thus, nitrate andHNO3 concentrations will be compared as a function of temperature andRH to
determine if these factors bias sample collection. The hypothesis is that there is not difference in
the samplers as a function of temperature or RH.
Due to resource limitations, HNO3 data were not obtained from the VAPS Na2CO3 denuder.
Therefore, this hypothesis can not be fully evaluated. However a partial response is given in
No. 6 under Specific Hypotheses Related to Questions Ql and Q2 and No. 1 under Specific
Hypotheses Related to Questions Q3 and Q4.
10. Crustal related elements are typically associated w ith particles greater than 2.5 um AD,
however, differences in the inlet efficiency or slope of the collection efficiency curves may result
in different concentrations ofcrustal related material being measured by these samplers. The
hypotheses is that the crustal related material as determined from summing the oxides ofFe, Ca,
and Si are not statistically different among the samplers. If differences are observed relate these
differences to the coarse particle mass as measured by the VAPS or by other collocated PM10
monitors. Wind speed may also be a variable of interest in this analysis.
Significantly higher mass concentrations are reported by the MetOne sampler using the spiral
inlet then the other samplers, including the FRM. As stated above, this is due to the spiral inlet
allowing particles greater than 2.5 um to penetrate, as well as an efficiency curve that is not as
steep as the FRM or Andersen cyclone (Peters et al. 2000; Appendix H; Appendix I)
Secondly, the VAPS sampler at sites where high coarse particle loadings were observed, also
had higher measured concentrations of crustal related elements relative to the other samplers,
except the MetOne. At Rubidoux this is likely due to the operator leaving out the collection jet
after the first day of sampling, thus the VAPS collected PM10 on both its fine and coarse
sampling legs. However, high coarse particle related species also were observed at Phoenix.
This suggests that the collection efficiency of the VAPS sampler is not as efficient as the other
speciation samplers. This is expected as the slope of the efficiency curve for the VAPS is not
as sharp as, for example, the WINS impactor.
Part II, Page 41
-------
In general, the Andersen and IMPROVE samplers consistently reported slightly higher
concentrations of Fe, Ca, and Si then did the FRM, while the URG reported slightly lower
concentrations. While the differences among the Andersen, IMPROVE, and URG are usually
small, and reasonable based on the precision obtained with collocated samplers, they are
consistent biases and may be truly representative of differences among the samplers. For
example, the Andersen and IMPROVE use cyclones and the efficiency curve for the cyclone is
likely not quite as sharp as the WINS. The difference between the FRM and URG may be due
to the location of the denuder in the URG sampler, which is placed before the WINS impactor,
while the FRM does not have a denuder, but a hollow down tube. Additional evaluations of
these fractionators are needed to more thoroughly address this hypothesis.
Part II, Page 42
-------
SUMMARY AND CONCLUSIONS
The purpose of this study was to evaluate, prior to their implementation in EPA's PM2 5 Chemical
Speciation Network (EPA, 1999), the three chemical speciation samplers now available on the
National Sampler Contract. These samplers were compared to two historical samplers, the
IMPROVE sampler and the VAPS, as well as to two PM2 5 FRM, one operating with a Teflon filter
and one with a quartz-fiber filter to allow for determination of the same species as in the chemical
speciation monitors. Samplers were operated for up to 20 days during January and February 1999 at
four sites around the county: Rubidoux, CA; Phoenix, AZ; Philadelphia, PA; and Research Triangle
Park, NC. These sites were chosen to represent different chemical atmospheres and environmental
(meteorological) conditions. Rubidoux experiences high nitrate and moderate organic carbon, but low
sulfate, Phoenix experiences high coarse particle crustal material, high organic carbon, moderate nitrate,
and low sulfate, Philadelphia was chosen since it has high sulfate and low nitrate, while RTF was chosen
because it would represent an area near the sampler's limit of detection.
Key findings from the study, separated into Site Characteristics, Operations, Measurement
Performance, and Implications include:
Site Characteristics
• Chemical analysis indicated that the four cities chosen for the study met the criteria for different
chemical atmospheres (Tolocka et al. 2000).
Meteorological conditions also were varied with average high temperatures around 70 F in
Phoenix and Rubidoux and around 38 F in Philadelphia, where it snowed during parts of the
study.
Operations
• All samplers encountered operational or design problems that increased variability among
sampler results. Most operational and design issues have been resolved by the manufacturers.
The MetOne spiral inlet allowed particles greater then 2.5 m to penetrate and has been
replaced with a sharp cut PM 2.5 cyclone. Both fractionators have been evaluated in the
laboratory (Peters et al. 2000) against the WINS impactor in the FRM.
Flow indicator problems in the Andersen sampler identified through the audits, invalidated a
series of OC and EC data at Rubidoux. Fortunately, replicate data were obtained at that site.
This problem has been rectified by the manufacturer.
• Insufficient pump capacity in the R&P 400 sampler invalidated data for the first two days at
Phoenix and Rubidoux. This problem was rectified by the manufacturer during the study.
Tradeoffs exist among the samplers for ease of use, flexibility for sampling, and cost.
Measurement Performance
• Samplers tend to agree well (10-15%on average) for stable species. There were a few
exceptions due to differences in inlet collection characteristics.
XRF sulfur times 3 was statistically equal to sulfate by ion chromatography.
Part II, Page 43
-------
Differences of-1-3 jig/m3 are observed among the samplers for nitrate. Differences appear to
be due partially to the use of pre-heated quartz-fiber filters for nitrate vs Teflon filters. Results
also suggest differences due to the method for collecting particle nitrate, one of which includes
adding the nitrate collected on the quartz-fiber filter to a nylon filter behind a denuder and
Teflon filter (i.e., what is termed in this report the indirect method).
Sulfate measured on Teflon filters agreed well (with 5% or so) to sulfate measured on quartz-
fiber filters, suggesting that the differences mentioned above for nitrate between the two filter
types was not due to the ability of the samplers to collect fine particles, but to their ability to
collect ammonium nitrate.
Differences of up to 3.5 |ig/m3, on average, are observed among samplers for organic carbon.
Differences appear to be due to a positive sampling artifact and are a function of the face
velocity across the collection filters as well as other variables. Positive OC artifact was
estimated relative to a quartz-fiber filter behind an XAD denuder. Reactive backup filters were
not used in this study so no information is available on negative artifacts. Results from Phase n
of the study, where XAD and carbon impregnated filters and denuders are being evaluated for
efficiency and capacity, should yield considerable insight for better understanding positive and
negative artifacts associated with collecting OC on quartz-fiber filters.
While not definitive, the data suggest a low bias for the collection of ammonium on nylon filters.
It is postulated that the ammonia produced from ammonium nitrate volatilization is not efficiently
collected by the basic (pH) nylon filter; although it is appropriate for collecting nitrate with
minimal bias if a suitable denuder is located up stream of the nylon filter. Further investigation
should be conducted to evaluate the efficacy of using a nylon filter to collect ammonium
aerosols.
Results indicate that vacuum XRF can reduce the amount of nitrate on the filter by up to 40%.
Therefore, a Teflon filter analyzed for trace elements by XRF should not be used for nitrate
analysis; however, sulfate is not affected.
Implications
Nitrate
Direct measurement of nitrate on a reactive filter (nylon or Na2CO3 impregnated) directly
proceeded by a diffusion denuder will likely provide the most reliable measure of total particle
nitrate. However, denuder coatings and reactive coating should be appropriately matched.
This implication is supported by the following two results:
• Determination of nitrate by a sampling protocol that requires addition of nitrate
collected on a pre-heated quartz-fiber filter will result in nitrate data biased high relative
to the direct denuder/filter method.
Determination of nitrate by a sampling protocol that requires addition of nitrate
collected on a Teflon filter that has been previously analyzed by vacuum XRF may be
biased low by at least 40%, based on initial tests. Use of atmospheric pressure XRF
may provide a suitable solution. Further tests are needed in this area.
Part II, Page 44
-------
Carbon
Samplers should operate the OC channel at 16. 71/min or an equivalent face velocity of other
flow rates or filter sizes are used to provide OC data that are consistent to the FRM, that is, to
minimize bias between OC associated with FRM mass measured on a Teflon filter to that
measured on a quartz-fiber filter. However, the FRM flow rate (face velocity) may not be the
ideal face velocity for minimizing OC artifacts on quartz-fiber filters, it is only suggested here for
obtaining equivalent OC concentrations (assuming face velocity is the major cause for
differences observed). Use of organic denuders and reactive backup filters, similar to that used
for nitrate, will be needed to obtain the least bias OC data, relative to OC concentrations
observed in air, as long as filter based technology with retrospective chemical analysis is the
method of choice. These implications are supported by the following results:
Positive artifact was observed for OC collected on quartz-fiber filters and accounted
for up to 3.5 |ig/m3 of the observed OC concentrations.
• OC concentrations measured on quartz-fiber filters were dependent on face velocity of
the air stream through the filter, with lower face velocities reporting the highest OC
concentrations.
Additional understanding of the differences between the IMPROVE and MOSH methods for
OC and EC determination are required to allow comparable results to be obtained for EC and
OC by the two, now commercially available methods. The Office of Research and
Development is in the process of conducting tests to establish the equivalency of these two
methods.
Part II, Page 45
-------
REFERENCES
Chow, J.C. and R.T. Egami. 1997. San Joaquin Valley 1995 Integrated Monitoring Study:
Documentation, Evaluation, and Descriptive Data Analysis of PmlO, PM2 5, and Precursor Gas
Measurements. Technical Support Studies No. 4 and No. 8. Final Report, Dri Document No.
5460. If 1. Prepared for the California Air Resources Board, Technical Support Division, Sacramento,
Ca.
Chow, J.C.; Watson, J.G.; Lu, Z.; Lowenthal, D.H.; Frazier, A.; Solomon, P.A.; Thuillier, R.H.;
Magliano, K. 1996. Descriptive Analysis of PM2 5 and PM10 at Regionally representative locations
during SJAQS/AUSPEX. Atmospheric Environment, 30(12), 2079-2112.
Eatough, D.L., B. Sedar, E.A. Lewis, E.A. Hansen, and RJ. Farber. 1989. Determination of Semi-
volatile Organic Compounds in Particles in the Grand Canyon Area. Aerosol Science and Technology,
10, 438-456.
Eldering, A.; Solomon, P.A.; Salmon, L.G; Fall, T.; Cass, GR 1991. Hydrochloric acid: A
Regional Perspecitive on Concentrations and Formation in the Atmosphere of Southern California.
Atmospheric Environment, 25A, 2091-2102.
Eldred, R.A., PJ. Feeney, and P.K. Wakabayashi. 1998. The Major Components of PM2 5 at
Remote Sites Across the United States. In Proceedings of the A&WMA International Speciality
Conference, PM2 5 A Fine Particle Standard, Eds. J.C. Chow and P. Kourtrakis. Air & Waste
Management Association, Pittsburgh, PA.
Fitz D., Chan M., Cass G, Larson D., and Ashbaugh L. A Multi-Channel, Multi-Component Size-
Classifying Aerosol And Gas Sampler, CARB Report; AV-TP-89/6033 Revised 5/89.
Gundel, L.A. andD.A. Lane. 1999. Sorbent-Coated Denuders for Direct Measurement of Gas/Particle
Partitioning by Semi-Volatile Organic Compounds, in Gas and Particle Partition Measurements of
Atmospheric Semivolatile Organic Compounds. D.A. Lane, ed., Gordon and Reach.
Gundel, L.A., V.C. Lee, K.R.R. Mahanama, R.K. Stevens, and J.M. Daisey. 1995. Direct Determination
of the Phase Distributions of Semi-volatile Poly cyclic Aromatic Hydrocarbons Using Annular Denuders,
Atmos. Environ. 29, 1719-1733.
Hering, S. and Cass, G. 1999. The Magnitude of Bias in the Measurement of PM25 Arising from
Volatilization of Particulate Nitrate from Teflon Filter. JAWMA, 49, 725-733.
John W., and Reischi G., JAPCA, vol. 3, No. 8, (1980) 872-876.
Kenny, L.C., Gussman, R.A., Meyer, M. 1999. Development of a Sharp-cut Cyclone for Ambient
Aerosol Monitoring Applications. Aerosol Science and Technology (in press).
Part II, Page 46
-------
Koutrakis, P. 1998. Recommendations of the Expert Panel On the EPA Speciati on Network. Prepared
for Office of Air and Radiation, OAQPS, Research Triangle Park, NC.
http://www.epa.gov/ttn/amtic/pmspec.html. file dated 07-22-98.
Koutrakis, P. 1999. Recommendations of the Expert Panel on the EPA Speciati on Network. Final
version. Prepared for Office of Air and Radiation, OAQPS, Research Triangle Park, NC.
http://www.epa.gov/ttn/amtic/pmspec.html. file dated 8-4-99.
Lippmann M., and Chan T.L., Ame. Ind. Hyg. Assoc. 1, 31, 133 (1970).
McDow, S.R. and Huntzicker, J. J. 1990. Vapor Adsorption Artifact in the Sampling of Organic Aerosol:
Face Velocity Effects. Atmospheric Environment, 24A, 2563-2571.
Peters, T.M. and Vanderpool, R.W.. 1996. Modification and Evaluation of the WINS Impactor. Final
Report, prepared by Research Triangle Institute, Research Triangle Park, NC for the US EPA, NERL,
Atmospheric Processes Division, Research Triangle Park, NC.
Peters, T.M.; Gussman, R.A.; Kenny, L.C. 2000. Evaluation of PM25 Size Selectors Used in Speciation
Samplers. Aerosol Science and Technology, Special Issue for PM2000, submitted for publication, Jan.
RTI. 1999. Quality Assurance Proj ect Plan for Four-City PM2.5 Chemical Speciation Sampler Evaluation
Study, RTI Project Number 07263-030. RTI, RTF, NC.
Russell, A.G.; McRae, G.J.; Cass, G.R. 1983. Mathematical modeling of the formation and transport of
ammonium nitrate aerosol. Atmospheric Environment, 17(5), 949-964.
Solomon, P.A., S.M. Larson, T.Fall, and G.R. Cass. 1988. Basinwide Nitric Acid and Related Species
Concentrations Observed during the Claremont Nitrogen Species Comparison Study. Atmospheric
Environment, 22(8): 1587-1594.
Solomon, P.A., T. Fall, L. Salmon, G.R. Cass, H.A. Gray, and A. Davidson. 1989. "Chemical
Characteristics of PM-10 Aerosols Collected in the Los Angeles Area." J. Air Pollut. Control Assn.
39(2):154-163.
Solomon, P.A.; Salmon, L.; Fall; T.; Cass. G.R. 1992. The Spatial and Temporal Distribution of
Atmospheric Nitric Acid and Particulate Nitrate Concentrations in Los Angeles. Environ. Sci. Technol.
26(8):1594-1601.
Stevens R.K. et al., Measurement of FDSTO3, NH3, SO2 and particulate nitrate with Annular Denuder
System, in Proceedings of Fifth Annual National Symposium on Present Advances in the Measurement of
Air Pollution, Document #EPA/600/9-85-029, USEPA, Research Triangle Park, NC, pp55-71 (1985).
Part II, Page 47
-------
Tolocka, M.P.; Solomon, P.A.;, Mitchell, W.; Norris, G.; Gemmill, D.; Vanderpool, R.; Homolya, J.
2000. East vs. West in the US: Chemical Characteristics of PM2.5 during the Winter of 1999. Aerosol
Science and Technology, Submitted for publication.
Turpin, B.J., JJ. Huntzicker, S.V. Hering. 1994. Investigation of Organic Aerosol Sampling Artifacts in
the Los Angeles Basin. Atmospheric Environment, 28(19), 3061-3071.
C:\WINDOWS\TEMP\04 Part 2, RESULTS, v8-f .wpd
Part II, Page 48
-------
Table n-1. Summary of Flow Audit Results Outside ± 10%*, Values in Percent (Indicated - Audit
Flow); Data Shown are [Sampling Line] and Bias for Each Value Exceeding the Criteria.
(I = Initial, M = Middle; F = Final Flow Audits).
Andersen #1
Andersen #2
Met One #1
URG#1
URG#2
IMPROVE #1
IMPROVE #2
FRM#1
FRM#2
VAPS
IMPROVE
25 mm
I
M
F
I
M
F
I
M
F
I
M
F
I
M
F
I
M
F
I
M
F
I
M
F
I
M
F
I
M
F
I
M
F
Rubidoux
[4] -43
[1] +17
[4] +167
[R] -13
Phoenix
[E-B]+11; [W-B]+12
Philadelphia
[3] +11
[3] +29
RTP
[3] -13
* There were 9 values outside the range of < -10% and > +10% and; 19 values between the ranges of +5% to 10%
and (-5%) to (-10%) of which 8 were associated with the IMPROVE sampler that used 25 mm filters, leaving 11 in this
range over all other sampling lines. The total number of sampling lines where flow rates were audited across all sites
and audits was about 315.
Part II Tables, Page 1
-------
Table II-2a. Limits of Detection in ng/m3.
Species
Mass1
Sulfate2
Nitrate2
Ammonium2
OC3
EC3
S4
Si4
K4
Ca4
Mn4
Fe4
Cu4
Zn4
Pb4
As4
Concentration
(ng/m3)
2000*
19
14
14
<300
<500
12
30
5.4
7.8
2.1
5.4
2.4
2.1
4.2
2.4
Method
Gravimetric Analysis
Ion Chromatography
Ion Chromatography
Ion Chromatography
Thermal Optical Reflectance
Thermal Optical Reflectance
EDXRF
EDXRF
EDXRF
EDXRF
EDXRF
EDXRF
EDXRF
EDXRF
EDXRF
EDXRF
By Federal Reference Method: Using field blanks, mass detection limit of the FRM is defined as the
absolute value of measured mean plus 10 times the standard deviation. From past studies, this resulted
in approximately 46 micrograms. For a 16.7 Lpm sampler, this equates to a detection limit of approx. 2
micrograms per cubic meter.
2 LOD based on 3 times the baseline noise.
3' Detection limits calculated as 3 times the standard deviation of the field blanks, averaged across all
field blanks collected during the study.
Detection limits calculated as 3 times the propagated uncertainly in the XRF method.
Part II Tables, Page 2
-------
Table II-2b. Average Field Blank Data for All Species and Samplers Averaged Across All Sites In
Atmospheric Concentrations (n=10).
Sampler
Species
PM2.5
SO4=-T
S04=-Q
NO3-T
NO3-N
N03-Q
NH,+-T
NH,+-Q
OC
EC
S
Si
K
Ca
Mi
Fe
Cu
Zn
FRM
AND
IMP
MET
TJRG
VAPS
Detection
Limit*
Detection
Limit**
Average ± Standard Deviation in [ig/nf
0.2 ±0.2
0.05 ±0.02
0.02 ±0.01
0.02 ±0.02
0.5 ±0.2
0.0 ±0.1
0.2 ±0.1
0.0 ±0.0
0.0 ±0.0
0.0 ±0.0
0.0 ±0.0
1.5 ±0.3
0.0 ±0.1
0.1 ±0.2
0.0 ±0.0
0.0 ±0.0
0.0 ±0.0
0.0 ±0.2
0.0 ±0.0
0.2 ±0.5
0.0 ±0.1
0.0 ±0.0
0.0 ±0.0
0.0 ±0.0
0.7 ±0.6
0.0 ±0.1
0.1 ±0.2
0.1 ±0.0
0.0 ±0.0
0.0 ±0.0
0.4 ±0.01
0.0 ±0.0
0.2 ±0.2
0.0 ±0.0
0.0 ±0.0
0.0 ±0.0
0.0 ±0.0
0.4 ±0.1
0.0 ±0.0
2
0.01
0.01
0.01
0.01
0.02
0.01
0.01
0.4
0.09
5
0.03
0.03
0.04
0.04
0.04
0.03
0.03
1
0.2
Average ± Standard Deviation in ng/nf
1.9±3.1
10 ±14
0.6 ±1.0
0.0 ±0.0
0.2 ±0.3
2.1 ±3.0
0.1 ±0.3
0.1 ±0.2
3.2 ±3.6
5.1 ±8.4
0.2 ±0.6
0.0 ±0.0
0.2 ±0.3
0.7 ±1.4
0.1 ±0.3
0.3 ±0.5
2.7 ±3.0
9.3 ±8.0
0.0 ±0.0
0.3 ±1.4
0.0 ±0.0
0.3 ±0.88
0.1 ±0.3
0.0 ±0.0
18.8 ±22.3
35 ±38
1.7±4.1
16±31
0.3 ±0.7
5.5 ±13.8
0.8 ±1.9
2.4 ±4.5
1.5±3.3
6.4 ±8.2
0.0 ±0.0
0.0 ±0.0
0.0 ±0.0
0.6 ±1.7
0.0 ±0.0
0.2 ±0.5
4.4 ±4.1
7.1 ±8.2
0.0 ±0.0
0.0 ±0.0
0.3 ±0.6
0.3 ±0.7
0.2 ±0.5
0.2 ±0.5
3.5
8.5
1.5
2.1
0.5
1.4
0.7
0.5
8.7
21.2
3.7
5.3
1.4
3.7
1.7
1.4
* Based on 16.7 L/min
* * Based on 6.7 L/min
Part II Tables, Page 3
-------
Table II-2c. Precision (as % CV) Achieved by FRM and Speciation Samplers Based on the Results
from the Collocated Samplers at Rubidoux.*
Sampler
Species
PM2.5
SO4=
NO3-T
NCV-P
NH4+
OC
EC
S as SO4=
Si
K
Ca
Mn
Fe
Cu
Zn
Pb**
As**
PM10*
FRM
AND
IMP
MET
URG
In Percent
5.8
11.0
See Note 1
11.1
12.8
8.4
12.7
3.3
23.1
6.7
15.1
16.6
8.7
54.9
10.4
31
140
3.2
2.9
4.0
15.3
2.4
See Note 2
See Note 2
o o
J.J
24.4
4.4
12.4
37.8
8.9
30.6
6.8
14
120
5.4
6.1
See Note 1
6.1
7.8
22.7
38.4
8.7
17.9
10.5
11.3
17.9
9.3
38.0
16.0
36
120
4.7
3.1
3.1
3.9
o o
J.J
9.7
7.0
4.3
19.5
11.3
18.3
46.1
11.9
39.3
10.6
40
130
7.6
6.2
See Note 1
13.5
6.4
8.3
7.6
2.4
19.9
5.6
10.2
17.1
8.2
29.7
10.7
19
140
* Only a single VAPS sampler was located at Rubidoux.
* * Values were at or near the limit of detection for that species and these species have been excluded in
the remainder of the comparisons for that reason.
1. Nitrate was not measured on the Teflon filter for these samplers.
2. Precision for OC and EC could not be calculated for the Andersen Sampler because of an incorrect
indicated flow rate on Channel 4 for the primary sampler at Rubidoux. Therefore, replicate data for these
species were used for statistical analyses performed in this report.
Part II Tables, Page 4
-------
Table II-3. Valid Data Capture in Percent by Sampler and Major Species.*
Rubidoux
Phoenix
Philadelphia
RTF
Rubidoux
Phoenix
Philadelphia
RTF
Rubidoux
Phoenix
Philadelphia
RTF
Maximum
Possible
32
20
20
20
32
20
20
20
32
20
20
20
AND
FRM
IMP
MET
URG
VAPS
PM25Mass and Trace Elements
97%
100%
100%
100%
94%
95%
95%
90%
97%
95%
90%
95%
91%
90%
90%
90%
75%
70%
100%
95%
75%
65%
75%
75%
Sulfate, Nitrate, Ammonium Ions
94%
100%
100%
100%
84%
100%
100%
100%
100%
85%
95%
85%
97%
90%
90%
90%
100%
100%
95%
100%
100%
85%
95%
85%
Organic and Elemental Carbon
97%
100%
100%
100%
88%
100%
100%
95%
97%
100%
95%
95%
97%
95%
90%
85%
100%
100%
95%
100%
94%
85%
95%
80%
* Shaded values represent data capture less than 90%.
Part II Tables, Page 5
-------
Table II-4. Summary of Problems Encountered In the Field During Operations Of Sampler Evaluated in this Study.
Sampler
Andersen RAAS
MetOne SASS
Category
Sampler Installation
Initial Setup
Audits and Calibration
Programming
Filter Holder Exchange
Data Retrieval
Sampler Maintenance
Operating Manual
Other operational problems
Sampler Installation
Initial Setup
Audits and Calibration
Programming
Filter Holder Exchange
Data Retrieval
Sampler Maintenance
Site1
PA
RU
RU,PX,PA,RT
RU,PA,RT
RU
RU
RU
RTF lab
Noted Problem or Deficiency
Sampler weight noted to be heavier than average
Calibrations were lost on two different occasions
O-rings fell out during field filter exchange
Fluctuations in indicated flow rates
Drifts in cabinet temperature noted
Slight water intrusion in back panel
Difficult to diagnose leaks in sampler manifold
Laboratory disassembly of filter canister noted as difficult without damaging or
contaminating collected filter sample
Part H Tables, Page 7
-------
Table II-4. Summary of Problems Encountered In the Field During Operations Of Sampler Evaluated in this Study.
Sampler
MetOne SASS (cont)
TJRGMASS
Category
Operating Manual
Other operational problems
Sampler Installation
Initial Setup
Audits and Calibration
Programming
Filter Holder Exchange
Data Retrieval
Sampler Maintenance
Operating Manual
Other operational problems
Site1
RU
PXRT
PA
RT
RT
RU
PA
PA
RU
RU
RU
PX
RU,PX,PA,RT
PX,PA
Noted Problem or Deficiency
Sampler shut down after 10 hours on one occasion
On one occasion, sampler indicated a 1 sec run time
On two occasions, sampler did not start as programmed.
Initially, the sampler indicated incorrect flow rates due to use of improper flow
transducers during the sampler's construction. Replacement of transducers
corrected the problem.
Pump failed and required replacement.
Recommended changing the manner in which information is listed on the
Sharp edges were encountered on inlet head assembly
Screen not easily read in bright sunlight
screen
Sampler could not provide sufficient flow rate. Higher capacity pumping
installed.
Problems encountered with LED screens.
Filter temperature flags frequently encountered.
Filter temperature flags frequently encountered.
Sampler could not provide sufficient flow rate. Higher capacity pumping
installed
Filter holder subject to cross-threading
system
system
Part II Tables, Page 8
-------
Table II-4. Summary of Problems Encountered In the Field During Operations Of Sampler Evaluated in this Study.
Sampler
IMPROVE
VAPS
Category
Sampler Installation
Initial Setup
Audits and Calibration
Programming
Filter Holder Exchange
Data Retrieval
Sampler Maintenance
Operating Manual
Other operational problems
Sampler Installation
Initial Setup
Audits and Calibration
Programming
Filter Holder Exchange
Site1
RU,PX,PA,RT
RU
RT
RT
RU
RU
RU
PX
PX
RT
RT
RT
PX
PA
RU
RT
RU
Noted Problem or Deficiency
Setup of sampler noted to be cumbersome
Faulty pumps, timers, and valves
Transformer had to be replaced due to water intrusion into
case
A condensed operating manual was suggested to be of value
Faulty pump timer and required replacement
Pump failed and required replacement.
Denuders difficult to change
Problems encountered with magnahelics.
On one occasion, timer did not display correct elapsed time.
On two occasions, water was noted inside sampling case.
On one occasion, solenoid valve did not function.
On one occasion, elapsed timer failed.
Filter holders difficult to change
Connection from solenoid to filter holder was unreliable
Problems encountered with flow calibration requiring tube replacement.
Sampler did not start due to operator error
Hoses connecting filter holders would frequently work loose
Part II Tables, Page 9
-------
Table II-4. Summary of Problems Encountered In the Field During Operations Of Sampler Evaluated in this Study.
Sampler
VAPS (cont)
FRM-BGIPQ200
Category
Data Retrieval
Sampler Maintenance
Operating Manual
Other operational problems
Sampler Installation
Initial Setup
Audits and Calibration
Programming
Filter Holder Exchange
Data Retrieval
Sampler Maintenance
Operating Manual
Site1
PX
PA
RT
RU,PX,PA
RU
PX
RU,PX,PA,RT
PA,RT
PA
RU
RU
RU
Noted Problem or Deficiency
On one occasion, sampler did not automatically shut down as programmed.
On one occasion, sampler did not automatically shut down as programmed.
On two occasions, sampler did not automatically shut down as programmed.
O-rings routinely fell out during filter exchange
Sampler provided no information except total flow rate and elapsed time
Problems encountered with timers
Lower virtual impactor nozzle routinely fell out during filter exchange
Simpler, more reliable form of timer was suggested
Unit needs direct readout of sampling flow rate
Sampling pump required replacement
Entire sampler had to be replaced on one occasion
Recommended sampler redesign to prevent impactor from falling out during
sample exchange
Part II Tables, Page 10
-------
Table II-4. Summary of Problems Encountered In the Field During Operations Of Sampler Evaluated in this Study.
Sampler
FRM-BGIPQ200
(cont.)
FRM-R&P
Category
Other operational problems
Sampler Installation
Initial Setup
Audits and Calibration
Programming
Filter Holder Exchange
Data Retrieval
Sampler Maintenance
Operating Manual
Other operational problems
Site1
RU
PA
PA
RT
Noted Problem or Deficiency
On one occasion, some water intrusion noted on impactor shelf and bottom of
cabinet
Sampler would not operate - firmware upgrades installed
On several occasions, some water intrusion noted within cabinet
Rotating handle mechanism needs to be redesigned
Filter exchange mechanism did not operate properly on three occasions
1. Ru = Rubidoux, PX = Phoenix, PA = Philadelphia, RT = Research Triangle Park
Part II Tables, Page 11
-------
Table n-5. Species Concentration Data for the FRM at Each Location of the 4 City Study.
PM-2.5
S04Q
N03P
NH4
OC
EC
S
Si
K
Ca
Fe
Cu
Zn
Pb
As
Rubidoux
AVG*
Std
Dev
Max
Min
Valid
Data
Phoenix
AVG*
Std
Dev
Max
Min
Valid
Data
Philadelphia
AVG*
Std
Dev
Max
Min
Valid
Data
RTP
AVG*
Std
Dev
Max
Min
Valid
Data
Concentrations in (ug/m3)
26.7
1.7
11.8
4.0
5.7
3.3
20.2
1.6
10.4
3.6
2.1
1.9
74.3
6.0
38.0
13.2
10.0
7.7
2.2
0.3
0.1
0.1
2.1
0.6
14
13
13
13
13
13
14.9
0.9
3.1
1.2
7.6
3.3
6.8
0.4
2.2
0.7
2.5
1.3
25.3
1.8
7.4
2.5
12.5
5.7
3.9
0.2
0.2
0.1
4.0
1.5
17
18
18
18
18
18
17.4
4.1
3.8
2.6
4.3
2.5
9.0
2.1
2.1
1.3
2.0
1.3
37.6
8.4
8.6
5.8
9.6
5.7
5.0
1.5
0.9
0.8
2.3
0.8
17
18
18
18
18
18
11.0
3.3
0.7
1.3
3.4
1.5
5.4
1.6
0.6
0.6
1.9
0.9
23.8
6.2
2.3
2.3
8.5
3.7
4.4
0.7
0.1
0.2
1.5
0.6
16
18
18
18
17
17
Concentrations \n(ng/n?)
600
160
80
160
170
7
70
14
0.3
560
100
35
110
100
8
78
130
0.6
1930
378
125
360
386
31
255
49
1.4
110
16
24
34
55
0
4
1.8
0.0
12
12
12
12
12
12
12
12
12
300
280
140
110
210
9
18
6.6
1.1
150
150
70
62
110
9
10
5.2
1.3
640
480
250
220
420
38
48
22
3.5
71
50
30
26
57
2
2
0.0
0.0
17
17
17
17
17
17
17
17
17
1400
47
55
35
100
5
32
12
0.7
730
30
25
24
82
4
22
11
1.1
280
130
120
110
300
13
83
39
3.6
390
19
29
13
18
1
5
2.6
0.0
17
17
17
17
17
17
17
17
17
1100
72
67
32
52
8
16
4.3
0.5
560
61
42
36
32
15
14
3.5
0.8
210
230
180
150
120
64
55
16
2.9
260
19
32
13
11
0
5
0.0
0.0
16
16
16
16
16
16
16
16
16
Sampling dates included in averages:
Rubidoux - 1/15/99 to 2/26/99
Phoenix-1/14/99 to 2/21/99
Philadelphia - 1/21/00 to 2/26/99
RTP-1/19/99 to 2/24/99
Part II Tables, Page 12
-------
Table II-6. Estimated PM2.5 Mass Balance of Species versus Measured PM2.5 Mass (ug/m3) for the
FRM at Each Site.
Species+
SO4Q
NO3P
NH4
Organic
Material
EC
Crustal
Material
Sum of
Components
Measured
Mass
Percent
Measured
Mass
Coarse PM**
%NO3*
%SO4*
%Organics*
% Crustal*
Fine/PMlO in %
RUB
1.7
11.8
4.0
8.0
3.3
0.8
29.6
26.7
111
21.6
44.2
6.4
30.1
3.0
55
PHO
0.9
3.1
1.2
10.7
3.3
1.1
20.1
14.9
135
17.2
20.5
5.8
71.5
7.0
46
PHI
4.1
3.8
2.6
6.0
2.5
0.3
19.2
17.4
110
4.2
21.6
23.3
34.2
1.7
81
RTF
3.3
0.7
1.3
4.8
1.5
0.3
11.9
11.0
108
4.0
6.7
30.1
43.8
2.5
73
* Relative to the measured mass.
** Rubidoux coarse mass estimated from difference between VAPS PM2 5 mass minus PM2 5 mass measured on the
Andersen, assumes VAPS sample is PM10.
+ Species
SO4Q is sulfate measured on the quartz-fiber filter by 1C
NO3P is nitrate measured on the quartz-fiber filter by 1C and represents the best measurement of nitrate by
that sampler and is referred to in this report as particle nitrate
Organic Material = OC *1.4
Crustal Material Estimate (ng/irf) = (2.14*Si + 1.43*Fe+1.4*Ca)/1000
Part II Tables, Page 13
-------
Table II-7. Average Volatilized Nitrate (NO3V) in jig/m3 Observed for Each Sampler at Each City.
Volatilized Nitrate
Andersen
MetOne
URG
VAPS
IMP-FRM
Average NO3P2
All Samplers
Average PM2.5
Mass, All Samplers2
n
NO3V/Avg NO3P
(Average ±6) (%)
NO3V / PM2.5 Mass
(Average ±6) (%)
Average
Avg. Max.
Avg. Min.
Rubidoux
Phoenix
Philadelphia
RTF
(ug/m3)
Flow Rate
T or Q Denuder
(Lpm)
-0.36
1.01
0.99
0.64
0.12
11.5
32.4
65
12 ± 19
1.5± 1.8
1.46
1.14
1.57
1.41
0.87
3.3
17.1
77
37 ±20
7.6 ± 1.7
0.22
0.41
0.37
0.78
0.26
3.7
18.4
83
13 ± 16
2.2 ± 1.2
0.48
0.48
0.56
0.58
0.47
1.0
11.8
82
51 ±26
4.4 ±0.4
Temp. (°F) During Study Days
57
70
45
56
70
44
41
50
33
47
60
34
16.7 7.3
6.7 6.7
16.7 16.7
16.7 16.7
16.7 22.7
Nitrate measured on the nylon filter on the IMPROVE sampler minus nitrate measured on the Teflon filter of
the Andersen sampler.
Averaged over all samplers except the FRM at a given site; these PM2.5 mass and nitrate values are given
to allow for general comparisons amongst sites of the amount of nitrate volatilized from the Teflon filter
under the conditions of this experiment.
Part II Tables, Page 15
-------
Table II-8a. Mean Analyte Concentrations for Each Sampler at All Sites.1
Analyte
Site
PM25
Sulfate
Sulfur as Sulfate
Particulate Nitrate
Ammonium
Organic Carbon
Elemental Carbon
Rubidoux
Phoenix
Philadelphia
RTF
Rubidoux
Phoenix
Philadelphia
RTF
Rubidoux
Phoenix
Philadelphia
RTF
Rubidoux
Phoenix
Philadelphia
RTF
Rubidoux
Phoenix
Philadelphia
RTF
Rubidoux
Phoenix
Philadelphia
RTF
Rubidoux
Phoenix
Philadelphia
RTF
Mean Concentrations
N
9
7
13
10
13
13
16
13
4
7
11
10
8
9
15
10
13
13
16
13
12
14
16
12
12
14
16
12
Silicon
Potassium
Calcium
Iron
Zinc
Rubidoux
Phoenix
Philadelphia
RTF
Rubidoux
Phoenix
Philadelphia
RTF
Rubidoux
Phoenix
Philadelphia
RTF
Rubidoux
Phoenix
Philadelphia
RTF
Rubidoux
Phoenix
Philadelphia
RTF
4
7
11
10
4
7
11
10
4
7
11
10
4
7
11
10
4
7
11
10
FRM
AND
IMP
MET
URG
VAPS
(ug/m3)
17.9
15.1
18.9
11.4
1.7
0.8
4.0
3.4
1.6
0.9
4.1
3.1
14.7
2.7
3.7
1.0
4.0
1.2
2.6
1.3
5.6
7.6
4.4
3.6
3.2
3.3
2.6
1.5
17.7
15.4
19.0
11.3
1.7
0.7
3.8
3.3
1.7
0.9
4.2
3.1
13.8
2.7
3.5
1.1
3.7
0.6
2.4
1.3
6.9
9.1
5.6
4.1
2.8
2.8
2.4
1.4
17.4
13.3
18.8
10.9
1.6
0.7
3.8
3.3
1.6
1.0
3.7
2.9
14.4
2.2
3.5
1.1
3.2
0.6
2.2
1.2
4.1
6.5
4.0
3.0
1.8
2.7
2.3
1.2
23.5
27.8
19.7
12.1
1.8
0.9
3.9
3.3
1.9
1.1
4.5
3.2
16.0
2.9
3.7
1.1
4.0
0.8
2.6
1.2
6.2
8.1
4.0
4.1
3.0
3.0
2.2
1.4
24.1
14.7
19.2
11.7
1.8
0.8
3.9
3.4
1.8
0.9
4.4
3.3
16.8
4.0
4.0
1.7
4.4
1.0
2.6
1.5
5.3
6.5
3.8
2.9
3.1
3.2
2.4
1.5
38.4
16.1
22.7
13.2
1.9
0.8
3.5
3.3
1.9
0.9
4.3
3.0
17.0
4.0
3.9
1.7
3.8
1.1
2.2
1.5
6.4
4.7
2.8
3.1
3.0
2.8
2.2
1.5
(ng/m3)
103
293
54
89
61
141
61
75
79
119
38
41
111
223
122
54
12.3
18.1
36.2
19.3
159
356
59
82
70
144
62
73
126
152
44
40
153
262
135
52
13.4
18.4
38.2
14.9
134
364
64
80
66
148
61
69
118
157
42
38
141
264
128
49
12.0
18.8
34.6
14.5
553
1502
119
105
140
328
73
82
478
671
84
51
339
669
177
70
25.4
29.6
38.5
14.4
106
232
57
79
63
134
64
77
81
98
35
35
117
193
122
49
12.9
17.6
38.9
15.6
821
383
61
166
188
150
64
82
577
166
39
56
503
276
126
70
18.5
17.3
38.7
14.8
Shaded data indicate values inconsistent with other samplers. Reasons for differences are explained in the
text for most data points. Questionable data, with no direct explanation, remain in the data base and are
included in the statistical analyses presented in this report.
Part II Tables, Page 16
-------
Table n-8b. Ratio of Speciation Sampler to FRM for Chemical Components by Site5'2.
PM25
Sulfate
P-nitrate
Ammonium
Organic
Carbon6
Elemental
Carbon
Sulfur
Silicon
Potassium
Calcium
Iron
Zinc
Site
iubidoux
Dhoenix
Dhiladelphia
}TP
iubidoux
Dhoenix
Dhiladelphia
iTP
iubidoux
Dhoenix
Dhiladelphia
iTP
iubidoux
Dhoenix
Dhiladelphia
iTP
iubidoux
Dhoenix
Dhiladelphia
iTP
iubidoux
Dhoenix
Dhiladelphia
iTP
iubidoux
Dhoenix
Dhiladelphia
iTP
iubidoux
Dhoenix
Dhiladelphia
iTP
iubidoux
Dhoenix
Dhiladelphia
iTP
iubidoux
Dhoenix
Dhiladelphia
iTP
iubidoux
Dhoenix
Dhiladelphia
iTP
iubidoux
Dhoenix
Dhiladelphia
}TP
Andersen
0.99
1.02
1.01
099
1.00
0.88
0.95
0.97
0.94
1.00
0.95
1.10
0.93
0.50
0.92
1.00
1.08
1.94
2.00
1.32
0.88
0.85
0.92
0.93
1.05
0.98
1.02
1.00
1.54
1.21
1.09
0.91
1.15
1.02
1.02
0.98
1.59
1.27
1.15
0.98
1.39
1.17
1.10
0.96
1.09
1.02
1.06
077
IMPROVE
0.97
0.88
0.99
0.96
0.94
0.88
0.95
0.97
0.98
0.81
0.95
1.10
0.80
0.50
0.85
0.92
0.64
1.38
1.43
0.97
0.56
0.82
0.88
0.80
1.00
1.03
0.90
0.94
1.30
1.24
1.17
0.89
1.08
1.05
1.01
0.93
1.49
1.31
1.10
0.93
1.28
1.19
1.05
0.90
0.98
1.04
0.96
075
URG
1.35
0.97
1.02
1.03
1.06
1.00
0.98
1.00
1.14
1.48
1.08
1.70
1.10
0.83
1.00
1.15
0.83
1.38
1.36
0.94
0.97
0.97
0.92
1.00
1.10
0.99
1.06
1.08
1.03
0.79
1.04
0.88
1.03
0.95
1.06
1.04
1.02
0.82
0.92
0.86
1.06
0.86
1.00
0.90
1.05
0.97
1.07
081
MetOne
1.31
1.84
1.04
1.06
1.06
1.13
0.98
0.97
1.09
1.07
1.00
1.10
1.00
0.67
1.00
0.92
0.97
1.72
1.43
1.32
0.94
0.91
0.85
0.93
1.13
1.17
1.08
1.04
5.35
5.13
2.19
1.18
2.29
2.33
1.20
1.10
6.04
5.62
2.20
1.25
3.06
3.00
1.45
1.28
2.07
1.64
1.06
075
Average
1.16
1.18
1.01
1.01
1.01
0.97
0.96
0.98
1.04
1.09
0.99
1.25
0.96
0.63
0.94
1.00
0.88
1.61
1.55
1.14
0.84
0.89
0.89
0.92
1.07
1.04
1.02
1.01
2.31
2.09
1.37
0.97
1.39
1.34
1.07
1.01
2.53
2.26
1.34
1.00
1.70
1.56
1.15
1.01
1.29
1.17
1.04
077
StdDev
0.20
0.45
0.02
0.05
0.06
0.12
0.01
0.01
0.09
0.28
0.06
0.30
0.13
0.16
0.07
0.11
0.19
0.27
0.30
0.21
0.19
0.07
0.04
0.08
0.06
0.09
0.08
0.06
2.04
2.03
0.54
0.14
0.60
0.66
0.09
0.07
2.35
2.25
0.58
0.17
0.92
0.97
0.20
0.18
0.52
0.31
0.05
003
Shaded cells represent values exceeding the PM Expert Panel's performance criteria: Slope = 1 ± 0.1 for
mass, nitrate, and ammonium, and 1 ± 0.05 for sulfate
Organic carbon is relative to OC measured on the quartz-fiber filter behind the XAD denuder located in
the VAPS sampler.
Part II Tables, Page 17
-------
Table II-9.
Regression Statistics of FRM (x-axis) versus Speciation Samplers (y-axis) for All Sites, Samplers, and Major Species.
Site
Rubidoux
Phoenix
Philadelphia
RTF
PM2.5 Mass*
Sampler
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Slope
1.10
1.11
1.15
1.43
1.90
0.98
0 76
1.67
0.94
1.19
1.01
1.00
1.08
1.02
1.33
0 97
0 99
1.09
1.05
1.12
Std Err.
Slone
0 05
0.07
0.14
0.04
0.23
0.02
0 05
0.21
0.04
0 10
0.01
0.02
0 05
0 03
0.06
0 03
0 03
0.04
0 02
0 05
Intercept
-2.94
-3.61
4.16
-2.18
1.88
0.29
1.56
1.52
0.32
-1.26
-0.11
0.00
-059
0 01
-2.45
0 09
-020
-0.28
-002
0.57
Std. Err.
Intercept
1.55
2.30
4.61
1.35
5.11
0.26
0.81
3.44
0.68
1.65
0 19
0.38
1.02
0 54
1.31
0 42
0 35
0.46
0 30
0.67
R2
0.979
0.955
0.851
0992
0.883
0.996
0.940
0.826
0.979
0936
0999
0.995
0.970
0989
0.976
0983
0.988
0.985
0993
0981
Rubidoux
Phoenix
Philadelphia
RTF
Nitrate*
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
0 92
0.93
1.06
1.05
1.09
0.95
0.73
1 00
1.41
1.21
0.88
0.96
1.02
1 09
1.03
0.96
0.92
0.85
1.54
1.39
0 03
0.04
0.03
0 06
0 07
0.04
0.04
0 06
0.07
0.08
0 08
0.04
0.05
0 06
0.09
0.06
0.08
0.19
0.11
0.22
-0.13
0.23
0.06
1.15
0 55
0.14
0 20
0.23
0.26
0.46
0.21
-0.05
-0.21
-0.14
0.04
0.15
0.16
0.27
0.11
0.28
0 49
0.56
0.51
1.13
1 09
0.14
0 16
0.21
0.24
0.30
0 34
0.18
0.24
0.27
0.39
0.06
0.08
0.18
0.11
0.23
0987
0.984
0.989
0.975
0.967
0.977
0.951
0959
0.970
0.951
0.884
0.972
0.959
0949
0.905
0.937
0.903
0.570
0.926
0.792
Highlighted values are slopes out side the range from 0.8 to 1.2 or R less than 0.8
Units for Intercept and Standard Error of the Intercept are * = ug/m ; and ** = ng/m .
Sulfate*
Sampler
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Slope
1.03
0.93
0.97
1.10
1.08
0.89
0.89
0.96
0.95
0.87
1.04
0.99
1.02
1.01
0.90
1.04
1.04
1.05
1.02
1.04
Std Err.
Slone
0 04
0.04
0.04
0 03
0.05
0.04
0 04
0.04
0.05
0 06
0.04
0.03
0 03
0 02
0.02
0 02
0 02
0.03
0 03
0.04
Intercept
-0 06
0.00
0.12
-0 09
0.05
-0.02
-0.01
0.13
0.03
0 03
-0 40
-0.14
-0.17
-0.16
-0.15
-0.18
-0.22
-0.21
-0 03
-0.17
Std. Err.
Intercept
0.10
0.09
0.10
0.08
0.11
0.04
0 04
0.04
0.05
0.06
0.16
0.13
0.12
0.10
0.09
0 07
0 09
0.12
0.10
0.14
R2
0.982
0.982
0.977
0990
0.979
0.971
0969
0.971
0.957
0.947
0981
0.988
0990
0.992
0.993
0995
0993
0.986
0.988
0982
Ammonium*
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
1.01
0.83
1.03
1.06
0 99
0.57
0.49
0.73
0.89
0.89
1.03
0.92
1.04
1 09
0.93
0.97
0.92
1.00
1.07
1.17
0 02
0.03
0.02
0 03
0 05
0.06
0 05
0 06
0.07
0.08
0 03
0.04
0.05
0.04
0.03
0.05
0.07
0.07
0.05
0.07
-0 32
-0.12
-0.16
0.14
-0.22
0.00
-0.01
-0 01
0.00
0.05
-0.25
-0.22
-0.10
-0.27
-0.25
0.03
-0.04
-0.05
0.07
-0.05
0 09
0.16
0.09
0.14
0.26
0.08
0.07
0 09
0.10
0.10
0 09
0.13
0.13
0.11
0.08
0.07
0.10
0.10
0.06
0.10
0997
0.986
0.997
0993
0.975
0.872
0.861
0902
0.911
0.915
0985
0.966
0.972
0979
0.985
0.958
0.931
0.933
0.971
0.952
-------
Table II-9. Regression Statistics of FRM (x-axis) versus Speciation Samplers (y-axis) for All Sites, Samplers, and Major Species (cont.)
Site
Rubidoux
Phoenix
Philadelphia
RTF
Organic Carbon*
Sampler
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
TIRG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Slope
1.39
0.39
1 08
1 06
1.76
1 13
0.82
1.03
0.79
0.77
099
085
0.99
089
0.66
1 11
0.88
1.10
0.95
092
Std Err.
Slooe
0.15
019
0 12
012
0.30
005
0.06
029
0.10
009
Oil
006
0.13
005
0.06
007
0.04
006
0.06
0.13
Intercept
-1.05
1.76
012
-0.72
-3.40
061
0.10
-0.17
0.31
-1.16
122
0.26
-0.35
-0 13
-0.09
-002
-0.28
0.06
-0.41
-0.23
Std. Err.
Intercept
092
1.13
073
0.71
1.77
039
0.50
2.38
0.82
068
052
0.26
0.60
023
0.27
027
0.15
0.25
0.25
053
R2
0.886
0.289
0 882
0.884
0.776
0972
0.916
0453
0.789
0 852
0 833
0936
0.802
0 954
0.902
0944
0.975
0.960
0.936
0.821
Elemental Carbon*
Sulfur**
Rubidoux
Phoenix
Philadelphia
RTF
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
1.14
1.08
1.11
1.17
1 19
0.97
1.17
1 09
1.01
0.85
1.03
0.92
1 08
1.08
1.07
1 00
1.01
1.03
1 06
1 06
004
0.04
0.04
0.06
008
0.02
Oil
007
002
0.06
003
0.08
003
0.03
0.05
002
004
002
001
0 15
-47.3
-52.6
26.3
-44.0
-9 9
3.6
-56.0
192
-10.3
28.4
-20.4
-19.6
20
-29.9
-63.8
-8 9
-57.9
4.2
6.6
-96.5
30.7
29.1
34.6
62.0
400
7.0
353
230
7.5
18.7
41.1
125
553
52.3
70.5
277
460
25.8
17.6
185
0991
0.989
0.985
0.986
0969
0.993
0.888
0949
0995
0.962
0990
0.913
0 986
0.986
0.982
0993
0.981
0995
0997
0 853
Sampler
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
TTRG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Slope
0.78
0.23
085
0.82
0.78
094
0.82
0.74
0.91
0.89
0.29
0.77
1.01
094
0.95
080
0.64
0.87
0.90
1.13
Std Err.
Slone
008
016
008
007
0.07
010
0.06
024
0.03
004
023
005
0.12
005
0.05
005
0.05
004
0.06
016
Intercept
030
1 00
023
0.46
0.47
-023
0.02
0.31
0.11
-0 16
1 59
0.26
-0.43
-002
-0.31
012
0.20
005
0.10
-0.22
Std. Err.
Intercept
030
059
029
028
0.27
035
0.20
0.85
0.12
0.12
063
0.14
0.34
0 13
0.14
010
0.09
006
0.10
0.29
R2
0 895
0.168
0 919
0.917
0.918
0 852
0.927
0.399
0.977
0979
0.096
0934
0.827
0 963
0.960
0934
0.916
0979
0.941
0 819
Silicon**
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
1 03
1.46
5.61
0.82
750
1.08
1.03
4.51
0.76
0.76
0.79
0.85
208
0.90
1.14
080
0.81
1.16
0.69
0.43
014
0.14
0.95
0.15
081
0.06
0.12
077
005
2.59
0.13
0.16
048
0.15
0.15
008
010
018
008
046
333
-35.9
-111
15.6
159
27.5
50.8
775
59
473
18.3
17.4
22 2
6.8
-1.8
109
4.7
7.4
17.7
120
237
26.5
177
30.6
111
19.6
38.4
252
17.3
707
7.3
9.6
264
8.5
9.1
7 1
92
17.3
7.8
489
0 866
0.913
0.776
0.854
0994
0.953
0.827
0.708
0952
0.009
0.707
0.675
0 576
0.699
0.854
0 887
0.846
0.757
0.846
0.091
Highlighted values are slopes out side the range from 0.8 to 1.2 or R less than 0.8
Units for Intercept and Standard Error of the Intercept are * = ug/m ; and ** = ng/m .
-------
Table II-9. Regression Statistics of FRM (x-axis) versus Speciation Samplers (y-axis) for All Sites, Samplers, and Major Species (cont).
Site
Rubidoux
Phoenix
Philadelphia
RTF
Calcium* *
Sampler
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Slope
0.97
1.16
5.40
0.60
6.93
1.10
1.03
4.82
076
0.96
1.09
1 08
2.49
0.88
1.18
1.03
1.12
1.17
0.92
1.10
Std Err.
Slniv
0.15
0.17
1.21
0.16
0.51
0.05
0.12
0.86
004
2.83
0.09
009
0.37
0.07
0.06
0.04
0.07
0.07
0.03
0.26
Intercept
39.0
95
-26.4
34.1
1.6
15.3
27.1
43.4
6 3
189
0.6
0.9
-12.7
0.5
-5.9
-1.6
-6.2
4.0
-1.8
10.1
Std. Err.
Tntprrpnt
26.8
325
229.3
332
69.9
6.7
15.0
111.6
5 2
312
3.8
3 5
16.0
•7 2
2.6
2.0
3 3
3.4
1.6
14.4
R2
0.831
0.821
0.667
0.729
0.964
0.966
0834
0.691
0973
0.013
0.905
0925
0.761
0.902
0.977
0978
0953
0.955
0.983
0.666
Iron* *
Rubidoux
Phoenix
Philadelphia
RTF
Zinc**
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
1.01
0.92
1 02
1.10
0.59
0.69
0.65
099
0.69
085
1.10
1.08
1.12
1.13
1.19
0.46
0.19
0.43
0.39
0.33
0.10
0.11
0 10
0.16
0.14
0.08
0.13
023
0.10
029
0.07
0.08
0.06
0.07
0.08
0.15
0.20
0.15
0.15
0.23
6.2
4.5
144
10.0
14.2
5.2
5.1
8 9
4.1
4.3
-2.4
-3.9
-29
-2.4
-4.1
6.3
10.0
6.1
8.5
7.9
9.4
11.2
9 8
16.2
10.5
1.7
2.8
4 8
2.1
4.8
2.7
•7 2
2.3
2.6
3 5
•7 2
4.1
3.2
3.1
5.4
0.922
0.878
0919
0.908
0.719
0.820
0.612
0.576
0.825
0.495
0.942
0.932
0.961
0.949
0952
0.390
0.070
0.391
0.352
0.195
Sampler
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Slope
1.22
1.18
3.16
092
3.62
1.04
1 00
2.73
083
0.95
1.02
1.01
1.48
0.98
1.05
093
096
1.28
0.90
0.98
Std Err.
Slnnp
0.09
0.13
0.50
006
0.43
0.04
0.11
0.40
003
1.42
0.05
0.05
0.10
0.04
0.08
007
006
0.08
0.05
0.20
Intercept
2.9
-5.0
-72.3
13.8
65.1
24.9
340
12.1
4 1
185
8.5
3.2
-3.6
2.1
-1.7
1.0
-29
-1.4
-1.9
15.4
Std. Err.
Tntprrent
17.5
24.3
96.9
13.2
63.9
8.7
26.7
96.7
8 5
303
6.6
7.0
13.5
4.8
11.8
4.2
3.6
4.7
3.1
12.3
R2
0.948
0898
0.800
0.977
0.911
0.982
0836
0.768
0981
0.047
0.964
0966
0.938
0.979
0.943
0.927
0954
0.956
0.963
0.719
Potassium**
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
Andersen
IMPROVE
MetOne
URG
VAPS
1.06
0.96
2.07
1.20
1.51
0.97
0.95
2.14
0.94
0.66
1.02
0.89
1.16
1.05
1.17
0.98
0.96
1.04
1.05
1.10
0.13
0.19
082
0.15
0.87
0.02
0.11
030
0.02
0.71
0.04
0.08
010
0.05
0.06
0.03
0.05
005
0.02
0.06
5.4
6.3
228
-9.2
91 0
6.8
8.9
109
0.2
94.7
0.8
6.7
3.8
0.4
-7.4
-0.2
-1.4
4.8
-1.2
-0.2
11.1
16.5
708
14.3
63.5
3 3
17.0
467
2.5
97.7
2.1
4.7
6.2
3 2
4.1
2 ~i
4.0
3.7
1.9
4.8
0.884
0.718
0.391
0.924
0.299
0.993
0.827
0.784
0.997
0.088
0.982
0.912
0902
0.962
0972
0.987
0.964
0975
0.993
0978
Highlighted values are slopes out side the range from 0.8 to 1.2 or R less than 0.8
Units for Intercept and Standard Error of the Intercept are * = ug/m ; and ** = ng/m .
-------
Table n-10a Results from the Paired T-Tests Between the Andersen and the FRM Samplers for Each Analyte/Site.
SUE
PM25
S04=
s
P-N03
NH4
OC
EC
Si
K
Ca
Fe | 7n
P-values from the Paired T-tests
(Values less than 0.05 Mean That the Andersen and FRM Samplers Are Statistically Different)
Rubidoux
Phoenix
Philadelphia
RTF
0.8046
0.8567
0.2428
0.3181
0.9811
0.0001
0.0028
0.3618
0.1878
0.0992
0.2042
0.2995
0.0147
0.8751
0.1933
0.0043
0.0003
0.0001
0.0004
0.7789
0.0003
0.0001
0.0001
0.0083
0.0200
0.0042
0.5427
0.0070
0.0119
0.0001
0.0594
0.5253
0.0443
0.0521
0.0509
0.2029
0.0517
0.0001
0.1044
0.6908
0.0060
0.0001
0.0160
0.2760
0.3403
0.6960
0.6290
0.3710
Statistical Decision Based on P-values
("Y" Means That the Andersen and FRM Are Statistically the Same and "N" Means They Are Statistically Different)
Rubidoux
Phoenix
Philadelphia
RTF
Y
Y
Y
Y
Y
N
N
Y
Y
Y
Y
Y
N
Y
Y
N
N
N
N
Y
N
N
N
N
N
N
Y
N
N
N
Y
Y
N
Y
Y
Y
Y
N
Y
Y
N
N
N
Y
Y
Y
Y
Y
Number of the Differences (FRM - Andersen) That Were Positive/total Number of Differences Used in T-test1
Rubidoux
Phoenix
Philadelphia
RTF
9/12
8/17
8/17
12/16
7/13
18/18
15/18
11/18
5/13
12/17
8/17
8/16
11/13
16/18
16/18
4/18
13/13
18/18
16/18
6/18
1/13
0/18
1/18
4/17
2/13
14/18
12/18
12/17
2/11
14/18
12/18
12/17
1/11
7/17
5/17
11/16
1/11
0/17
3/17
10/16
1/11
0/17
3/17
9/16
6/11
7/17
9/17
8/16
1 Individual data are presented in Appendices G-l and G-2 as absolute differences and percent differences.
Part II Tables, Page 20
-------
Table n-10b. Results from the Paired T-Tests Between the MetOne and the FRM Samplers for Each Analyte/Site.
SUE
PM25
S04=
s
P-N03
NH4
OC
EC
Si
K
Ca | Fe
Zn
P-values from the Paired T-tests
(Values less than 0.05 Mean That the Met One and FRM Samplers Are Statistically Different)
Rubidoux
Phoenix
Philadelphia
RTF
0.0121
0.0001
0.1022
0.0085
0.3549
0.0001
0.0592
0.5417
0.0076
0.5387
0.0007
0.0031
0.0572
0.0521
0.3036
0.1805
0.4718
0.0002
0.9839
0.2622
0.0367
0.8109
0.8820
0.0038
0.1350
0.0772
0.3125
0.0049
0.0036
0.0001
0.0005
0.1007
0.0030
0.0001
0.0002
0.0013
0.0059
0.0001
0.0067
0.0048
0.0035
0.0001
0.0067
0.0048
0.0413
0.0008
0.5579
0.2977
Statistical Decision Based on P-values
("Y" Means That the Met One and FRM Are Statistically the Same and "N' Means They Are Statistically Different)
Rubidoux
Phoenix
Philadelphia
RTF
N
N
Y
N
Y
N
Y
Y
N
Y
N
N
Y
Y
Y
Y
Y
N
Y
Y
N
Y
Y
N
Y
Y
Y
N
N
N
N
Y
N
N
N
N
N
N
N
N
N
N
N
N
N
N
Y
Y
Number of Differences (FRM - Met One) that Were Positive/total Number of Differences Used in T-test1
Rubidoux
Phoenix
Philadelphia
RTF
2/12
0/16
8/16
2/15
5/13
1/16
14/17
10/17
0/12
2/16
0/16
3/15
1/13
4/16
10/17
2/17
10/13
16/16
11/17
10/17
4/13
5/17
12/17
3/14
8/13
11/17
13/17
12/15
0/12
0/16
1/16
2/15
1/12
0/16
1/16
1/15
0/12
0/16
2/16
1/15
1/12
0/17
1/16
1/15
1/12
2/17
8/16
9/15
1 Individual data are presented in Appendices G-l and G-2 as absolute differences and percent differences.
Part II Tables, Page 21
-------
Table H-lOc. Results from the Paired T-Tests Between the IMPROVE and the FRM Samplers for Each Analyte/Site.
SUE
PM25
S04=
s
P-N03
NH4
OC
EC
Si | K | Ca
Fe
Zn
P-values from the Paired T-tests
(Values less than 0.05 Mean That the IMPROVE and FRM Samplers Are Statistically Different)
Rubidoux
Phoenix
Philadelphia
RTF
0.8006
0.0011
0.7233
0.0326
0.0850
0.0855
0.0300
0.0440
0.3672
0.6851
0.0567
0.8762
0.1839
0.0017
0.0378
0.0957
0.0016
0.0001
0.0001
0.0041
0.0034
0.0001
0.0015
0.0001
0.0074
0.0001
0.0034
0.0025
0.0714
0.0041
0.0655
0.1696
0.3461
0.7233
0.0655
0.1696
0.0743
0.0004
0.0997
0.3837
0.2084
0.0104
0.2654
0.0135
0.4717
0.4664
0.5416
0.4598
Statistical Decision Based on P-values
("Y" Means That the IMPROVE and FRM Are Statistically the Same and "N" Means That They Are Statistically Different)
Rubidoux
Phoenix
Philadelphia
RTF
Y
N
Y
N
Y
Y
N
N
Y
Y
Y
Y
Y
N
N
Y
N
N
N
N
N
N
N
N
N
N
N
N
Y
N
Y
Y
Y
Y
Y
Y
Y
N
Y
Y
Y
N
Y
N
Y
Y
Y
Y
Number of Differences (FRM - IMPROVE) that Were Positive/total Number of Differences Used in T-test1
Rubidoux
Phoenix
Philadelphia
RTF
8/12
15/17
7/16
11/15
11/13
11/15
14/18
12/15
10/12
12/17
15/15
11/15
10/13
17/18
13/18
4/15
12/13
18/18
18/18
13/15
11/13
17/18
14/18
15/16
13/13
18/18
15/18
14/16
2/12
3/17
5/15
9/15
4/12
7/17
7/15
12/15
1/12
2/17
7/15
12/15
1/12
2/17
7/15
11/15
8/12
10/17
9/14
9/15
1 Individual data are presented in Appendices G-l and G-2 as absolute differences and percent differences.
Part II Tables, Page 22
-------
Table II-10d. Results from the Paired T-Tests Between the URG and the FRM Samplers for Each Analyte/Site.
SUE | PM25
S04=
s
P-N03
NH4
OC
EC
Si | K
Ca
Fe
Zn
P-values from the Paired T-tests
(Values less than 0.05 Mean That the URG and FRM Samplers Are Statistically Different)
Rubidoux
Phoenix
Philadelphia
RTF
0.0126
0.1028
0.0849
0.0035
0.2518
0.4804
0.0306
0.4797
0.1252
0.0312
0.0058
0.0001
0.0285
0.0001
0.1530
0.0001
0.0039
0.0194
0.7010
0.0001
0.1228
0.0001
0.0001
0.0001
0.3385
0.0023
0.0097
0.2117
0.3121
0.0004
0.6142
0.4226
0.2054
0.0003
0.0235
0.1004
0.2852
0.0007
0.0541
0.0059
0.9188
0.0005
0.9746
0.0007
0.2359
0.1797
0.3195
0.5242
Statistical Decision Based on P-values
("Y" Means That the URG and FRM Are Statistically the Same and "N" Means They Are Statistically Different)
Rubidoux
Phoenix
Philadelphia
RTF
N
Y
Y
N
Y
Y
N
Y
Y
N
N
N
N
N
Y
N
N
N
Y
N
Y
N
N
N
Y
N
N
Y
Y
N
Y
Y
Y
N
N
Y
Y
N
Y
N
Y
N
Y
N
Y
Y
Y
Y
Number of Differences (FRM - URG) that Were Positive/total Number of Differences Used in T-test1
Rubidoux
Phoenix
Philadelphia
RTF
0/7
9/13
7/16
3/15
3/13
11/18
12/18
7/18
0/7
10/13
2/17
0/14
1/9
0/15
7/17
0/17
1/13
13/18
11/18
1/18
9/13
15/18
18/18
15/17
9/13
14/18
12/18
10/17
3/7
13/13
7/17
6/15
1/7
12/13
4/17
4/15
5/7
13/13
13/17
12/15
6/7
12/13
9/17
13/15
2/7
8/13
6/17
3/15
1 Individual data are presented in Appendices G-l and G-2 as absolute differences and percent differences.
Part II Tables, Page 23
-------
Table H-lOe. Results from the Paired T-Tests Between the VAPS and the FRM Samplers for Each Analyte/Site.
SUE | PM25
S04=
s
P-N03
NH4
OC
EC
Si
K
Ca | Fe
Zn
P-values from the Paired T-tests
(Values less than 0.05 Mean That the VAPS and FRM Samplers Are Statistically Different)
Rubidoux
Phoenix
Philadelphia
RTF
0.0017
0.1079
0.0029
0.0003
0.0317
0.0283
0.0001
0.4910
0.0646
0.3578
0.2833
0.7570
0.0558
0.0004
0.3888
0.0021
0.1426
0.2232
0.0001
0.0063
0.3898
0.0001
0.0001
0.0330
0.1948
0.0001
0.0001
0.7603
0.0022
0.1281
0.2766
0.0438
0.0009
0.1685
0.1892
0.0291
0.2852
0.1297
0.7160
0.1869
0.0005
0.1180
0.5425
0.0526
0.5488
0.2688
0.3816
0.4045
Statistical Decision Based on P-values
("Y" Means That the VAPS and FRM Are Statistically the Same and "N' Means That They Are Statistically Different)
Rubidoux
Phoenix
Philadelphia
RTF
N
Y
N
N
N
N
N
Y
Y
Y
Y
Y
Y
N
Y
N
Y
Y
N
N
Y
N
N
N
Y
N
N
Y
N
Y
Y
N
N
Y
Y
N
N
Y
Y
Y
N
Y
Y
Y
Y
Y
Y
Y
Number of Differences (VAPS and FRM) that Were Positive/total Number of Differences Used in T-test1
Rubidoux
Phoenix
Philadelphia
RTF
0/9
3/11
1/13
0/11
1/13
15/15
17/17
8/15
1/9
7/11
7/12
3/10
1/11
1/12
6/16
0/13
10/13
11/15
17/17
2/15
5/12
15/15
16/17
9/13
7/12
15/15
16/17
9/13
0/9
0/11
5/12
1/11
0/9
0/11
5/12
1/11
0/9
0/11
6/12
4/11
0/9
0/11
4/12
3/11
2/8
6/13
8/12
6/11
1 Individual data are presented in Appendices G-l and G-2 as absolute differences and percent differences.
Part II Tables, Page 24
-------
Table II-11. Results from the ANOVA for Examining Equivalency Among the Samplers for Particle Nitrate.
SITE
Rubidoux
Phoenix
Philadelphia
RTF
Sampler Means (|ig/m3) and Number of Values (X) Used to Compile
Each Mean
Andersen
10.0(14)
3.1(18)
3.5(18)
0.86(18)
IMPROVE
10.5(14)
2.5(18)
3.6(18)
0.88(15)
MetOne
11.7(14)
3.4(16)
3.6(17)
0.96 (17)
URG
16(9)
4.3 (15)
4.0(18)
1.3(17)
Statistically Equivalent Samplers
(Andersen, IMPROVE)
Andersen, IMPROVE);
(Andersen, Met One)
(Andersen, IMPROVE, Met One);
(IMPROVE, Met One, URG)
(Andersen, IMPROVE, Met One);
(Met One, URG)
Part II Tables, Page 25
-------
Table II-12a. Nitrate Concentrations (|ig/m3) Measured on Teflon (T) or Quartz-Fiber (Q) Filters by
Sampler Type Averaged Over the Study Period .
Site
Rubidoux
Phoenix
Philadelphia
RTF
AND-T
10.4
1.6
3.3
0.4
MetOne-T
10.7
2.4
3.2
0.4
FRM-Q
11.8
3.1
3.8
0.7
URG-Q
11.4
2.8
3.6
0.7
VAPS-Q
11.7
2.5
3.1
0.8
Table II-12b. Total Particle Nitrate Concentrations (|ig/m3) Measured by Each Sampler Averaged
Over the Study Period.
Site
Sampling
Configuration*
Rubidoux
Phoenix
Philadelphia
RTF
AND
(MgO/N)
(direct)
10.0
3.0
3.5
0.86
MetOne
(MgO/N)
(direct)
11.7
3.4
3.6
0.89
URG
(Q+ Na2CO3/N)
(indirect)
12.7
4.3
4.0
1.3
VAPS
(Q+ Na2CO3/N)
(indirect)
12.8
3.9
3.9
1.4
IMP
(Na2CO3/N
)(direct)
10.5
2.5
3.6
0.9
FRM
(Q)
11.8
3.1
3.8
0.7
* MgO/N = The direct method for particle nitrate — MgO coated denuder followed by a nylon filter, total nitrate
is measured on the nylon filter; the indirect method for particle nitrate — Na/Q/N = Na2CO3 coated denuder with
total particulate nitrate equal to the sum of nitrate measured on the Quartz-fiber filter, also used for OC/EC
analysis, plus nitrate measured on the nylon filter behind the Na2CO3 coated denuder; Na/Q/N / MgO/N = ratio
of two methods. Also see Figures 1-1 to 1-5.
Part II Tables, Page 26
-------
Table II-12c. Volatilized Nitrate Concentrations (|ig/m3) Measured by Each Sampler Averaged Over
the Study Period.
Site
Sampling
Configuration
Rubidoux
Phoenix
Philadelphia
RTF
AND
(MgO/N)
(indirect)
-0.36
1.46
0.22
0.48
MetOne
(MgO/N)
(indirect)
1.01
1.14
0.41
0.48
URG
(Q+ Na2CO3/N)
(direct)
0.99
1.57
0.37
0.56
VAPS
(Q+ Na2CO3/N)
(direct)
0.64
1.41
0.78
0.58
IMP
(Na2CO3/N)
(indirect)
0.12
0.87
0.26
0.47
* MgO/N = The indirect method to measure volatilized nitrate — MgO coated denuder followed by a nylon filter,
volatilized nitrate is the difference between the nitrate on the nylon filter minus that on the Teflon filter
collected in parallel; The direct method to measure volatilized nitrate — Na/Q/N = Na2CO3 coated denuder with
parti culate nitrate equal to the sum of nitrate measured on the Quartz-fiber filter, also used for OC/EC analysis,
plus nitrate measured on the nylon filter behind the Na2CO3 coated denuder and volatilized nitrate is measured
directly on the nylon back-up filter; Na/Q/N / MgO/N = ratio of two methods. Also see Figures 1-1 to 1-5.
Table II-12d. Sulfate Concentrations (|ig/m3) Measured on Teflon (T) and Quartz-Fiber (Q) Filters
Averaged Over the Study Period.
Site
Rubidoux
Phoenix
Philadelphia
RTF
AND-T
1.6
0.8
3.8
3.3
MET-T
1.7
1.0
4.1
3.3
FRM-Q
1.7
0.9
4.1
3.3
URG-Q
1.7
0.8
4.0
3.3
VAPS-Q
1.8
0.8
3.4
3.2
FRM-Q
1.5
0.8
3.8
3.2
Part II Tables, Page 27
-------
Table II-13 a. Nitrate Concentrations (|ig/m3) Measured on Teflon or Quartz-Fiber Filters Averaged
Over the Study Period*.
Site
Rubidoux
Phoenix
Philadelphia
RTF
Teflon
10.5
1.9
3.2
0.4
Quartz
11.4
2.9
3.4
0.7
(Q-T)
0.9
1.0
0.2
0.3
Q/T**
1.3 ±0.7
1.7±0.8
1.2±0.3
1.8±0.6
* In Tables II-13a through II-14d, Rubidoux has 14 data points in the averages, while the other three sites
have 18.
** Average and standard deviation.
Figure II-13b. Total Particle Nitrate Concentrations (|ig/m3) Measured by Different Denuder-Filter
Pack Methods Averaged Over the Study Period.
Denuder-Filter Pack Configuration1"
Site
Rubidoux
Phoenix
Philadelphia
RTF
MgO/N
(direct)
10.9
3.2
3.6
0.9
Na/Q/N
(indirect)
12.5
4.4
4.0
1.3
(Na/Q/N) -
(MgO/N)
1.6
1.2
0.4
0.4
(Na/Q/N) /
(MgO/N)**
1.1 ±0.4
1.3 ±0.2
1.1±0.1
1.5 ±0.5
* MgO/N = The direct method for particle nitrate — MgO coated denuder followed by a nylon filter, total nitrate
is measured on the nylon filter; the indirect method for particle nitrate — Na/Q/N = Na2CO3 coated denuder with
total particulate nitrate equal to the sum of nitrate measured on the Quartz-fiber filter, also used for OC/EC
analysis, plus nitrate measured on the nylon filter behind the Na2CO3 coated denuder; Na/Q/N / MgO/N = ratio
of two methods. Also see Figures 1-1 to 1-5.
** Average and standard deviation.
Part II Tables, Page 28
-------
Table II-13 c. Volatilized Nitrate Concentrations (|ig/m3) Measured by Different Denuder-Filter Pack
Methods Averaged Over the Study Period.
Site
Rubidoux
Phoenix
Philadelphia
RTF
Direct*
0.51
1.35
0.45
0.52
Indirect**
0.32
1.28
0.30
0.48
(Direct - Indirect)
0.19
0.07
0.15
0.04
(Direct / Indirect)
1.9±2.7***
1.0±0.3
1.91 ±2.3
1.1±0.7
* Direct is volatilized nitrate measured directly on a Nylon Filter behind a denuder and a Teflon Filter (URG,
VAPS). Also see Figures 1-1 to 1-5.
** Indirect is nitrate measured by difference (Nylon - Teflon) (Andersen and MetOne). Also see
Figures I-1 to 1-5.
*** Average and standard deviation.
Table II-13d. Sulfate Concentrations (|ig/m3) Measured on Teflon and Quartz Filters Averaged Over
the Study Period.
Site
Rubidoux
Phoenix
Philadelphia
RTF
Teflon
1.6
0.8
3.9
o o
J.J
Quartz
1.7
0.8
3.9
3.3
(Q-T)
0.1
0.00
0.00
0.00
Q/T
1.1
1.0
1.0
1.0
Part II Tables, Page 29
-------
Table n-14. Loss of Nitrate Resulting from Analysis of Teflon Filter by Vacuum XRF.
Average
Standard
Deviation
n
Standard Error
Nitrate Lost from Filter
Loss NO3" (ug)
73.7
47.7
26.0
9.3
NO3 (%)
38.8
15.7
26.0
3.1
Sulfate Lost from Filter
SO42
0.4
2.6
26.0
0.5
SO42- (%)
-1.4
6.1
26.0
1.2
Part II Tables, Page 30
-------
Table 11-15. Summary of Site Operators Surveys Regarding Speciation Sampler Setup and Operation.
Evaluation Criteria
Sampler Installation
Was more than one person required for field installation?
Were the operating manual and schematics of value during
assembly?
Number of sampler components involved during
installation
Were sampler components easily accessible during
installation?
Relative sampler installation time
Ease of adjusting sampler's initial date, time, flow rate, etc.
Overall rating pertaining to ease of sampler field installation
Sampler Audits and Calibrations
Ease of temperature and pressure audit procedures
Ease of leak check procedure
Ease of flow audit procedure
Were specialized tools or adaptors required for flow audit
procedure?
Can the sampler be temperature calibrated in the field?
Can the sampler be pressure audited in the field?
Andersen
RAAS
Met One
SASS
TJRG
MASS
IMPROVE
VAPS
FRM
Yes
Partially
Moderate
Yes
Long
Good
Good
No
Yes
Moderate
Yes
Quick
Good
Good to
Excellent
Yes
Yes
Moderate
Yes
Moderate
to Quick
Good
Good
Yes
Yes
Too many
Yes
Long
Fair
Fair to
Average
Yes
Yes
Moderate
Yes
Moderate
Average
Average
Yes
Yes
Moderate
Yes
Moderate to Quick
Good
Good
Good to
Excellent
Good to
Excellent
Good
Yes
Yes
Yes
Good
Good to
Excellent
Good
Yes
No
No
Good
Good
Good
No
Yes
Yes
Good
Good
Average
No
No
No
-
Average
Average
Yes
No
No
Good
Good
Good
No
Yes
Yes
Part II Tables, Page 31
-------
Table 11-15. Summary of Site Operators Surveys Regarding Speciation Sampler Setup and Operation.
Evaluation Criteria
Can the sampler be flow calibrated in the field?
Ease of temperature calibration procedure
Ease of pressure calibration procedures
Ease of flow calibration procedure
Sampler Programming
Ease of programming sampler start and stop times
Reliability of the programming
Filter Holder Exchange Between Sampling Events
Would parts such as o-rings, fractionator components, etc.
fall during filter exchange?
Ease of assembling and disassembling filter holders and
denuders in the laboratory
Overall rating pertaining to ease of filter holders and
denuders in the field
Time required to change filter holders and denuders in the
field
Andersen
RAAS
Yes
Average to
Good
Average to
Good
Average
Met One
SASS
No
-
-
-
TJRG
MASS
Yes
Average
Average
Average to
Good
IMPROVE
Yes
-
-
Average to
Good
VAPS
Yes
-
-
Fair to
Average
FRM
Yes
Average to Good
Average to Good
Good
Good to
Excellent
Good
Excellent
Excellent
Good to
Excellent
Good to
Excellent
Average
Average
Fair to
Average
Average
Good to Excellent
Good to Excellent
Always
Good to
Excellent
Fair to
Average
Moderate
No
Fair to
Average
Good to
Excellent
Quick
No
Good
Good
Quick
No
Good
Average to
Good
Moderate
to Long
Yes
Fair to
Average
Fair to
Average
Moderate
to Long
No
Good
Good
Quick to Moderate
Part II Tables, Page 32
-------
Table 11-15. Summary of Site Operators Surveys Regarding Speciation Sampler Setup and Operation.
Evaluation Criteria
Data Retrieval
Did the sampler provide the necessary post-sampling
information?
Was the data easily retrievable from the sampler?
Overall rating pertaining to the ease of data retrieval in the
field
Operating Manuals
Were the contents of the operating manual clear?
Were the contents of the field SOP clear?
Andersen
RAAS
Met One
SASS
TJRG
MASS
IMPROVE
VAPS
FRM
Yes
Yes
Good
Yes
Yes
Good to
Excellent
Yes
Yes
Good
No
Yes
Average
Yes
Yes
Fair to
Average
Yes
Yes
Good
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Part II Tables, Page 33
-------
Table 11-16. Recommended Spare Parts and Supplies for Use of Chemical Speciation Samplers and FRM Used in the Chemical Speciation
Evaluation Study
Sampler
FRM
Andersen RAAS
TJRG MASS 400,
450
MetOne SASS
VAPS
Improve
Item Description
Filter Cassettes
WINS Oil
Filter Cassettes
O-ring kit
Denuder - for MgO coating
Two stage filter holder - male threads (top), & quick connect
(bottom)
Single stage filter holder - male threads(top), & female threads
(bottom)
76 mm flow straightners - male threads
WINS oil
Denuder (down tube) - for Na2CO3 coating
MetOne filter holders
Filter Cassettes
Carbon denuders
MgO denuder
Sharp cut cyclone
Two stage filter holder - male threads (top), and quick connect
(bottom)
Single stage filter holder - male threads (top), & quick connect
(bottom)
Couplers
#30 Teflon Seal Ring
Glass denuder - for XAD coating
Denuder - for Na2CO3 coating
Improve filter cassettes
Denuder - for Na2CO3 coating
Part#
RAAS-CASS
RAAS-Oil
RAAS-CASS
RAAS-OR1
RAAS-S-DN
URG-2000-30F
URG-2000-30FB
URG-2000-30WN
URG-2000-30PB
URG-M130-80DD
8370
460128
8375
8374
8670
URG-2000-30F
URG-2000-30F
URG-2000-30BC2
URG-2000-30x24x3T
URG-2000-30CF
URG-2000-30x242-3CSS
-
-
Quantity
4
Two 0.5 ounce
bottles
16
2
2
4
4
2
Two 1 ounce bottles
2
10
20
4
4
5
4
8
2
6
4
4
12
4
Price/item1
35.00
10.00
35.00
5.00
700.00
334.15
333.13
102.20
15.30
321.28
190.00
28.00
100.00
100.00
250.00
334.15
281.11
46.82
16.93
527.75
311.92
5.00
75.00
1. Price estimate based on current manufacturers prices at the time of this study and not estimates for current day purchases.
C:\WINDOWS\TEMP\05 Part 2 Tables v8-f.wpd
Part II Tables, Page 34
-------
-
-
1
~
\
T
. _r_
•
r^ ~r
1 1 1
Mass _
.
-
=, :
i
10
8
i 6
«
1 4
0
O
2
0
-
.
-
•
I
j
Sulf ate ion
r T
1
.
•
• , T , -L [
L _L
1 ! ! !
Rubidoux, CA Phoenix, AZ Philadelphia, PA RTF, NC
Location
Rubidoux, CA Phoenix, AZ Philadelphia, PA RTF, NC
Location
-
-
1
J
1
L
T
•
I
1
T
1
Nitrate ion
•
1
ncentratiqig/m
(^, 0 IJ\
O
O
0
-
•
1
Ammonium i on
— ,
i
T • -r
1 • 1 • 1
i i i
Rubidoux, CA Phoenix, AZ Philadelphia, PA RTF, NC
L ocation
Rubidoux, CA Phoenix, AZ Philadelphia, PA RTF, NC
L oc ation
Figure Il-la. Frequency Distributions, Given as Box and Whisker Plots, of PM2.5 FRM Mass and Sulfate, Nitrate, and Ammonium ions at
Each of the Four Cites Studied.
Part II Figures, Page 1
-------
-
•
~
1
J
1
L
[
B
1
— i —
i
.
•
L
Org
1
ani c C arb on
1
1
—
8
"e 6
^
•£
03
0
O
0
Elemental Carbon
T
1
• • T
-r- ' T
i •
•
1 1 1 1
Rubidoux, CA Phoenix, AZ Philadelphia, PA RTF, NC
Location
Rubidoux, CA Phoenix, AZ Philadelphia, PA RTF, NC
Location
.2 300
T
Rubidoux, CA Phoenix, AZ Philadelphia, PA RTF, NC
Location
Calcium
_L
Rubidoux, CA Phoenix, AZ Philadelphia, PA RTF, NC
L oc ati on
Figure Il-lb. Frequency Distributions, Given as Box and Whisker Plots, of FRM PM2.5 OC, EC, Iron, and Calcium at Each of the Four
Cites Studied.
Part II Figures, Page 2
-------
35.0 -
30.0
25.0 '
ro"
~5fc 20.0 '
^
o
•- 15.0
-M
4*
U
O
^ 10.0 '
5.0 '
0.0 1
•
O
/\.
>/
1
•
vv
\/
1=1 Crustal
Material
1=1 EC
Material
• NH4
n NOIP
1=1 S04Q
* Measured
FRM
^ Coarse
RUB
PHO PHI
Location
RTF
Figure 11-2. Chemical Mass Balance of PM2 5 at Each City. Mass Balance is Relative to the Measured Mass by the FRM (Coarse Mass as
Measured by the VAPS). (See Table II-6 for Concentration Data and Percentages.)
Part II Figures, Page 3
-------
PM2.5 at Rubidoux, Jan. - Feb. 1999
PM2.5 Mass at Phoenix, Jan. - Feb. 1999
120
100
D)
3
CM
s
Q.
1 2 3 4 5 6 7 8 91011121314151617181920
Sample Number
PM2.5 Mass at Philadelphia, Jan. - Feb. 1999
PM2.5 Mass at RTP, Jan. - Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
Figure II-3a. Time Series for Mass.
Part II Figures, Page 4
-------
Sulfate at Rubidoux, Jan. - Feb. 1999
Sulfate at Phoenix, Jan. - Feb. 1999
234
7 8 9 10 11 12 13 14 15 16
Sample Number
1234567
9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
Sulfate at Philadelphia, Jan. - Feb. 1999
Sulfate at RTP, Jan. - Feb. 1999
n
^)
3
3
(/)
7 8 9 10 11 12 13 14 15 16 17 1819 20
Sample Number
10 11 12 13 14 15 16 17 18 19 20
Sample Number
Figure II-3b. Time Series for Sulfate.
Part II Figures, Page S
-------
Nitrate at Rubidoux, Jan.-Feb. 1999
2345
6 7 8 9 10 11 12 13 14 15 16
Sample Number
Nitrate at Phoenix, Jan - Feb 1999
n
~v
HI
"a
2 3 4 5 6 7
8 91011121314151617181920
Sample Number
Nitrate at Philadelphia, Jan.-Feb. 1999
Nitrate at Research Triangle Park, Jan.-Feb. 1999
1 2 3 4 5 6 7 8 9 10111213141516171819 20
Sample Number
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
Figure II-3c. Time Series for Nitrate.
Part II Figures, Page 6
-------
Ammonium at Rubidoux, Jan.-Feb. 1999
Ammonium at Phoenix, Jan.-Feb. 1999
1234567
9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
Ammonium at Philadelphia, Jan.-Feb. 1999
Ammonium at Research Triangle Park, Jan.-Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
Figure U-3d. Time Series for Ammonium.
Part II Figures, Page 7
-------
OC at Rubidoux, Jan.-Feb. 1999
OC at Phoenix, Jan.-Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Sampling Day
OC at Philadelphia, Jan.-Feb. 1999
OCatRTP, Jan.-Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Sampling Day
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Sampling Day
Figure II-3e. Time Series for OC.
Part II Figures, Page 8
-------
EC at Rubidoux, Jan.-Feb. 1999
EC at Phoenix, Jan.-Feb. 1999
O
LU
EC at Philadelphia, Jan.-Feb. 1999
1234567
10 11 12 13 14 15 16 17 18 19
Sampling Day
Figure II-3f. Time Series for EC.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Sampling Day
EC at RTP, Jan.-Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Sampling Day
Part II Figures, Page 9
-------
Si at Rubidoux, Jan. - Feb. 1999
Si at Phoenx, Jan. - Feb. 1999
3000
2500
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Si at Philadelphia Jan. - Feb. 1999
-AND
-FRM
-IMP
-MET
-URG
-VAPS
1 2 3 4 5 6
7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
Si at RTF Jan. -Feb. 1999
350
300
7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
Figure H-3g. Time Series for Silicon.
Part II Figures, Page 10
-------
Ca at Rubidoux, Jan. - Feb. 1999
Ca at Phoenx, Jan. - Feb. 1999
3000
1 2 3 4 5 6 7 8 9 10 11 12 13 14
350
Ca at Philadelphia Jan. - Feb. 1999
-AND
-FRM
-A-IMP
-MET
-URG
-VAPS
1 2 34 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
200
180
-AND
-FRM
-IMP
-MET
-URG
-VAPS
23456
7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
Ca at RTF Jan. - Feb. 1999
-AND
-FRM
-A-IMP
-MET
-URG
-X-VAPS
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
Figure II-3h. Time Series for Calcium.
Part II Figures, Page 11
-------
Fe at Rubidoux, Jan. - Feb. 1999
Fe at Phoenx, Jan. - Feb. 1999
1800
1600
1400
£• 1200
1000
800
600
400
200
-AND
-FRM
-A-IMP
-MET
-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
Fe at Philadelphia Jan. - Feb. 1999
Fe at RTF Jan. - Feb. 1999
600
100
-AND
-FRM
-IMP
-MET
-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
160
140
-AND
-FRM
-A-IMP
-MET
-URG
-VAPS
1 2 34 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
Figure II-3i. Time Series for Iron.
Part II Figures, Page 12
-------
FRM vs Speciation Samplers for Mass at Rubidoux
FRM vs Speciation Samplers for Mass at Phoenix
CO
"S
_3_
HI
Q.
E
re
(/)
c
0
"re
'o
&
CO
^)
_3_
to
HI
Q.
re
(/)
c
o
re
'o
HI
Q.
(/)
iaj •
20 •
(
FR
50-
45-
40-
35-
30-
25-
20-
15-
10-
5 •
0-
o
_D
.-••''' 4
n ...•••'' n^ff^^
^^ A
10 20 30 40 50 60 70 8
-)
» AND
A IMP
D MET
0 URG
X PTEMP
Linear (AND)
'Linear (MET)
Linear (URG)
Linear (PTEMP)
FRM (ug/m3)
.M vs Speciation Samplers for Mass at Philadelphia
z
z -
/u :?
y ,^
~/^£_
/*£r
~^-
^r
ay
g
* AND
A IMP
a MET
0 URG
X VAPS
Linear (AND)
Linear (IMP)
••- Linear (MET)
Linear (URG)
Linear (VAPS)
0 5 10 15 20 25 30 35 40
FRM (ug/m3)
Speciation Sampler (ug/m3)
IUU •
n -
n
g--"GD
..Q-'"^XGX
,^n ^-flZ^ll^
^•••"^ft^^3^^^
^^^
« AND
A IMP
D MET
0 URG
* VAPS
Li near (AND)
Linear (IMP)
Li near (MET)
Linear (URG)
— Linear (VAPS)
n
I,
3
8
Q.
30-
25-
v 20-
Q.
15-
10-
10 15 20
FRM (ug/m3)
25
30
FRM vs Speciation Samplers for Mass at Research
Triangle Park
« AND
A IMP
D MET
0 URG
X VAPS
Linear (AND)
— Linear (IMP)
•••Linear (MET)
Linear (URG)
— Linear (VAPS)
10 15
FRM (ug/m3)
20
25
Figure II-4a. Regression Analysis for Mass.
Part II Figures, Page 13
-------
FRM vs Speciation Sampler for Sulfate at Rubidoux
FRM vs Speciation Samples for Sulfate at Phoenix
E
I
_5
I 4
(/>
.2
15
8 2 -
Q. 2
o
ffl
'•'ii
x^x*
^^_
^"
>^
y
• AND
& IMP
D MET
0 URG
X PTEMP
Linear (AND)
Linear (IMP)
Linear (URG)
Linear (PTEMP)
01234567
FRM (mg/m3)
FRM vs Speciation Samplers for Sulfate in Philadelphia
co 7 •
C
01 c .
a. 5
E
,n 4-
c
o
a 3
'o
0)
Q. 9 .
(0 ^
i
J^K
j^x
X/
<*2—
jj&
J*
$ +
• AND
A IMP
D MET
0 URG
V. VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
0 2 4 6 8 10
FRM (ug/m3)
i
« AND
A IMP
D MET
O URG
^ VAPS
Linear (AND)
Linear (IMP)
••• Linear (MET)
Linear (URG)
Linear (VAPS)
FRM (ug/m3)
FRM vs Speciation Samplers for Sulfate at Research
Triangle Park
* AND
A IMP
D MET
0 URG
^ VAPS
Linear (AND)
— Linear (IMP)
Linear (MET)
Linear (URG)
-Linear (VAPS)
234
FRM(ug/m3)
Figure U-4b. Regression Analysis for Sulfate.
Part II Figures, Page 14
-------
FRM vs Speciation Samplers for Particluate Nitrate at
Rubidoux
E
^
0)
Q.
C
O
TO
'o
Q.
0)
(/)
n J
,•'
,-x"\Xxt
4*S—
x2_
" ^
2P_
o aX*
J*_
» AND
A IMP
D MET
0 URG
X PTEMP
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
— Linear (PTEMP)
10
15 20 25
FRM (ug/m3)
30
35
40
FRM vs Speciation Samples for Participate Nitrate at
Phoenix
345
FRM (ug/m3)
* AND
A IMP
D MET
o URG
X VAPS
Li near (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
— Linear (VAPS)
FRM vs Speciation Samplers for Particulate Nitrate at
Philadelphia
FRM vs Speciation Samplers for Particulate Nitrate at
RTP
10-
n
E« .
to
2 •
C
o X2
*• /••/^
/.•';/'''
nj*.''/'
v >«HX-
v X>Xf-§<
' ,^r
"?yr\ *
s$*^
Xp**
i
524681
FRM (ug/m3)
0
t AND
A IMP
D MET
° URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
— Li near (VAPS)
4"
3.5 •
p -5 .
O)
3
M 2.5 •
HI
g- 2.
E ^
re
1 1'5"
o 1 -
Q.
w
0.5
o-
*.
0 ^0
X ^/"^ .-'X
0 / •••'" ~
*y^ - -^
0 ^ .- ^-!S^
a^X^--^^"
S^,*®]^^ A
ic ^S~ J'^s.j*'*
?&>& *
&M' '•'
W
] 0.5 1 1.5 2 2
FRM (ug/m3)
5
* AND
A IMP
d MET
X VAPS
Linear (AND)
Li near (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
Figure II-4c. Regression Analysis for Nitrate.
Part II Figures, Page 15
-------
FRM vs Speciation Samplers for Ammonium at
Rubidoux
FRM vs Speciation Samplers for Participate Ammonium
at Phoenix
ug
a>
a.
ra
OT
c
o
u
HI
Q.
FRM (ug/m3)
o •
E
^>
2" 2'
"a.
E i K •
™ 1-5
(/)
c
o
ra 1
'o
HI
Q.
.X'o
.sy-
» v-" ,-°
X S
n&^
,. x^J .,-• D£
bx*».-V T^^*
^y^\^
^> * CL jrf^ "^"^ A
*^&** ~*
^f^^
* AND
A IMP
D MET
0 URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
— Linear (VAPS)
CM 1 1 1 1 1
0 0.5 1 1.5 2 2.5 3
FRM (ug/m3)
FRM vs Speciation Samplers for Particulate Ammonium
at Philadelphia
u>
3
Q.
E
ra
OT
c
o
FRM vs Speciation Samplers for Particulate Ammonium
atRTP
234
FRM (ug/m3)
0.5 1 1.5
FRM (ug/m3)
Figure U-4d. Regression Analysis for Ammonium.
Part II Figures, Page 16
-------
FRM vs Speciation Samplers for OC at Rubidoux
FRM vs Speciation Samplers for OC at Phoenix
« 12 •
E '^
"S
•3- m-
to
b
Q. a -
E B
re
(/)
0
!™
HI
Q.
c
x
/ f
S S
/ *
S s ,
^f * * ' ' n
4 j('*"
* -i^ "i *
4 ~^^^^^^^^_
v^T^l L '
/
1 23456789 10 1
FRM (ug/m3)
1
» AND
A IMP
n MET
0 URG
X VAPS
Linear (AND)
Linear (IMP)
-•••Li near (MET)
Linear (URG)
Li near (VAPS)
CO 14'
~5)
3, 12 •
to
Q.
E
" n
(/) 8 •
c
o
re 6 •
'o
(/) 4 •
2 •
C
+s
n >* •'*
a •^.•fi A ^
,'*•*" ^\^ *
^^f^^
^^
R
1 2 3 4 5 6 7 8 9 10 11 12 13 1
FRM (ug/m3)
4
0 URG
X VAPS
O MET
A IMP
Linear (AND)
Linear (URG)
Li near (VAPS)
Li near (MET)
FRM vs Speciation Samplers for OC at Philadelphia
12 '
10 •
FRM vs Speciation Samplers for OC at RTP
a>
a.
» AND
& IMP
0 URG
D MET
X VAPS
Linear (AND)
— Linear (IMP)
Linear (URG)
•--Li near (MET)
-Linear (VAPS)
4567
FRM (ug/m3)
10
345
FRM (ug/m3)
Figure II-4e. Regression Analysis for OC.
Part II Figures, Page 17
-------
FRM vs Speciation Samplers for EC at Rubidoux
FRM vs Speciation Samplers for EC at Phoenix
3456
FRM (ug/m3)
n
HI
Q.
E
o
&
23
FRM (ug/m3)
FRM vs Speciation Samplers for EC at Philadelphia
FRM vs Speciation Samplers for EC at RTP
Q.
o
HI
Q.
234
FRM (ug/m3)
FRM (ug/m3)
Figure II-4f. Regression Analysis for EC.
Part II Figures, Page 18
-------
FRM vs Speciation Samplers for Si at Rubidoux
FRM vs Speciation Samplers for Si at Phoenix
on Sampler (ug/m3;
Speciati
1500"
500 "
o-
I
D
n ..-•'''
,..-•'''
n ...••'
^^^T^^SSef^''1
I 50 100 150 200 250 300 350 4
» AND
A IMP
D MET
O URG
Linear (AND)
LillKdl (IMP)
'Linear (MET)
— Linear (URG)
)0
FRM (ug/m3)
FRM vs Speciation Samplers for Si at Philadelphia
CO
rn
c
to
HI
Q.
E
re
(/)
c
0
re
HI
(/)
J3U
250 '
200 •
100 •
D
D
n
D
D -~'
•'' ^«*****^^
^^^••^•^
.'' ^^"^^^^-^^
_,•- D rj2 ^^Jfc*^^ ?
2 j ^^f^^"^
* AND
A IMP
D MET
0 URG
X VAPS
Linear (AND)
Linear (MET)
Linear (URG)
— Linear (VAPS)
0 20 40 60 80 100 120 140
FRM (ng/m3)
n
~ut
£
to
HI
Q.
ra
(/)
c
o
ra
'o
HI
Q.
(/)
CO
^)
_c_
to
HI
Q.
TO
OT
c
o
TO
O
HI
Q.
(/)
3000 •
2500 •
2000-
1500-
1000-
500 •
0 •
150 "
VAPS = (212, 3958 ng/m3) on
1/22/99 removed as an outlier in
this plot. D
D
D.,"-""'D "
Q-''
^^v-*******^
* AND
A IMP
D MET
0 URG
X VAPS
Linear (AND)
Lineal (IMP)
•-- Linear (MET)
Linear (URG)
Linear (VAPS1
3 100 200 300 400 500
FRM (ng/m3)
FRM vs Speciation Samplers for Si at RTP
X
x n
x .-••'x
,..-rj"
r^^^^J^t
^T^ ,--^^^
^ i °
0 50 100 150 200 250
FRM (ng/m3)
* AND
A IMP
d MET
0 URG
X VAPS
Linear (AND)
Linear (IMP)
-- Linear (MET)
Linear (URG)
Linear (VAPS)
Figure H-4g. Regression Analysis for Si.
Part II Figures, Page 19
-------
FRM vs Speciation Samplers for Ca at Rubidoux
FRM vs Speciation Samplers for Ca at Phoenix
ouuu •
CO
E
O)
HI
Q.
re 1500-
c
o
n 1000-
u
&
OT
D
ft
,-'
a a -d'"' D
iii-^ — — ~~ ~~*°
» AND
A IMP
D MET
0 URG
Linear (AND)
Linear (IMP)
'Linear (MET)
Linear (URG)
0 50 100 150 200 250 300 350 400
FRM (ng/m3)
I4UU •
1200 "
CO
E
^ 1000 "
to
d)
o_ 800 '
E
re
c 600 "
0
v 400 •
200 "
0 "I
C
VAPS = (89, 1747 ng/m3) on
1/22/99 removed as an outlier
from this plot.
rj
,••' n
O n i1'
D
.--•
pQ' '"' n ii^ir^
~^~^^^r^ • °
50 100 150 200 2;
FRM (ng/m3)
50
* AND
A IMP
D MET
° URG
^ VAPS
Linear (AND)
Li near (IMP)
••- Linear (MET)
Linear (URG)
Linear (VAPS)
FRM vs Speciation Samplers for Ca at Philadelphia
n
c1 250 "
to
Q. 200
re
0
re
a, 100
50 '
D
,,.••'''
,,.••'''
••'° Jt
DBffx-"" j^^^^
s^^* *
* AND
A IMP
n MET
0 URG
^ VAPS
Linear (AND)
LillKdl (IMP)
Linear (MET)
Linear (URG)
— Linear (VAPS)
0 20 40 60 80 100 120
FRM (ng/m3)
FRM vs Speciation Samplers for Ca at RTP
Speciation Samplers (ng/m3
20 •
v
s<&
X^
^^ >?^
s^yS §
[x^x
JX^>*
pr
* AND
A IMP
D MET
0 URG
X VAPS
Linear (AND)
Linear (IMP)
--- Linear (MET)
Linear (URG)
Linear (VAPS)
0 20 40 60 80 100 120 1 40 1 60
FRM (ng/m3)
Figure H-4h. Regression Analysis for Calcium.
Part II Figures, Page 20
-------
FRM vs Speciation Samplers for Fe at Rubidoux
on Sampler (ng/m3)
Speciati
800 '
D
_..•'''
D ..'•'
,.•''
n, • •'" D ^
Dn ,••' n .-?£>•
D ^.^^^ °
^fc*'
100 200 300
FRM (ng/m3)
400
500
FRM vs Speciation Samplers for Fe at Phoenix
* AND
A IMP
D MET
0 URG
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
1400 •
,., 1200 '
E
&
& 1000 •
to
c 600'
o
re
$ 400"
(/)
200"
0 •!
t
VAPS = (174,1 689 ng/m3) on
1/22/99 removed as an outlier
from this plot.
D
n
,•' u
.-•"a D
DCb'"
. -b A ^^-^^
n xg ""' ° U^^^^^^
^^^
1 100 200 300 400 5t
FRM (ng/m3)
50
* AND
^ IMP
Q MET
° URG
x VAPS
Linear (AND)
Li near (IMP)
"" Linear (MET)
— Linear (URG)
1 inpar A/APR1
FRM vs Speciation Samplers for Fe at Philadelphia
FRM vs Speciation Samplers for Fe at RTF
J-.500-
E
to
5
Q.
re 300 '
c
0
15 200 •
'o
HI
n
.•••'n
^s*
Jf^^^o
...--" ^*^
..-a" ^^^^
.••^f^
a j^Mf
* AND
A IMP
a MET
0 URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
— Linear (VAPS)
lou •
160 '
CO
.E 140 •
O>
c
« 12°
Q. 100 •
TO
w 80 "
c
0
*i en -
o-
„''
.•'' S
TD ••' n ^^^ o
.•v ^^_ ^ ^^^
x ..•^X' ^^^
&S^ **-£4r^ O
Jlf^ ^t
jy^O^*
^-Sco£^»
/a^^gA
^.*'«
^
« AND
A IMP
n MET
0 URG
X VAPS
Linear (AND)
Linear (IMP)
-- Linear (MET)
Linear (URG)
Linear (VAPS1
50 100 150 200 250 300 350
FRM (ng/m3)
20 40 60 80 100 120 140
FRM (ng/m3)
Figure II-4i. Regression Analysis for Iron.
Part II Figures, Page 21
-------
50
45 -
40 -
Non-Denuded
Denuded
Non-Denuded C
Denuded C
12
- 10
o
Figure II-5a. Average of Denuded (5 samplers) and Non-denuded (1 sampler) OC Samples as a Function of the TOR Carbon Fractions.
C:\WINDOWS\TEMP\06 Part 2 FIGURES v8-f.wpd
Part II, Table II-5, 22
-------
Non-Denuded
Denuded
Non-Denuded C
Denuded C
Figure II-5b. Denuded (15 LPM, 1 sampler) and Non-denuded OC (16.7 LPM, 2 samplers) Samples as a Function of the TOR Carbon Fractions.
C:\WINDOWS\TEMP\06 Part 2 FIGURES v8-f.wpd
Part II, Table II-5, 23
-------
Rubidoux, CA
Phoenix, AZ
16
14
"E 12
o 10
ro s
O
o
'E 6
(0
D)
6 4
2
0
OCF
oc,
o
_g
ro
O
o
'c
ro
D)
OCF
oc,,
Mass jig m
Mass jig m"
Philadelphia, PA
Research Triangle Park, NC
03
O
I «
ro
D)
OC
FRM
VAPS
e-
o
_Q
ro
O
"
c
ro
D)
• OC,
oc>
M a s s jig m
Mass, jig m
Figure E-6. Organic Carbon (OC) Concentrations Obtained from Both the FRM and the VAPS (Using an XAD Denuder in Front of the Quartz-fiber
Filter to Remove SVOC) Compared to Total PM2 5 Mass a Measured by the FRM. Regression Statistics as (Slope, Intercept, r): Rubidoux - FRM
(0.082, 3.6, 0.728), VAPS (0.15, 2.0, 0.738); Phoenix-FRM (0.33, 2.5, 0.908), VAPS (0.25, 0.9, 0.875); Philadelphia-FRM (0.18, 1.4, 0.810),
VAPS (0.15, 0.3, 0.971); RTF - FRM (0.30, 0.22, 0.878), VAPS (0.28, -0.21, 0.899).
C:\WINDOWS\TEMP\06 Part 2 FIGURES v8-f.wpd
Part II, Table II-5, 24
-------
J
• OCl: 12
0 OC2: 25
T OC3: 45
v OC4: 55
can -
4
_!
^
0 °C
0 °C
0 °C
0 °C
395% CI
>
2
^
0
20
40 60 80
Face Velocity (cm/sec)
100
120
Figure II-7a. Organic Carbon as a Function of Face Velocity.
C:\WINDOWS\TEMP\06 Part 2 FIGURES v8-f.wpd
Part II, Table II-5, 25
-------
KS
CO
KS
CO
o
o
• i-H
+-»
cd
4 -
3 -
2 -
0
0
EC1: 550 °C
EC2: 700 °C
ECS: 800 °C
-Mean±95%CI
o
20
40 60 80
Face Velocity (cm/sec)
100
120
Figure II-7b. Elemental Carbon as a Function of Face Velocity.
C:\WINDOWS\TEMP\06 Part 2 FIGURES v8-f.wpd
Part II, Table II-5, 26
-------
200
SB 100
ii
o
J
-50
Excludes 3 Data Points > 500 ug/m3 Original Nitrate
y = 0.37x + 8.6
R2 = 0.725
r = 0.851
*
I .^. i
I \|/ I
100
200
300
400
500
600
700
Original Nitrate (ug/filter)
All Data
= 0.0555x + 60
R2 = 0.0662
800
1C 00
Figure II-8. Loss of Aerosol Nitrate from Teflon Filters Due to Vacuum XRF Analysis. Loss Is Difference Between Nitrate Concentrations on
Collocated Teflon Filters Before and After XRF Analysis. Error Bars are 5% of the X- Value and 7% of the Y- Value Plotted.
C:\WINDOWS\TEMP\06 Part 2 FIGURES v8-f.wpd
Part II, Table II-5, 27
-------
Evaluation of PM2.5 Chemical Speciation Samplers for Use in the EPA
National PM2.5 Chemical Speciation Network
Volume II - Appendices
Final Report
15 July 2000
Prepared By
Paul Solomon
William Mitchell
Michael Tolocka
Gary Norris
David Gemmill
Russell Wiener
US EPA, NERL
Research Triangle Park, NC 27711
Robert Vanderpool
Robert Murdoch
Sanjay Natarajan
Eva Hardison
Research Triangle Institute
Research Triangle Park, NC 27711
Prepared for
Richard Scheffe
James Homolya
Joann Rice
Office of Air Quality Planning and Standards
Research Triangle Park, NC
-------
List of Appendices
Appendix A: Sample Analysis Methods for Chemical Speciation
Appendix B. Standard Operating Procedures
Appendix C. Summary of Comments from Field and Laboratory Analysis Logbooks
Appendix D: Final Data Archive, By Species, Site, Date, and Sampler for the 4-City Study
Appendix Dl: Final Data Archive, By Species, Site, Date, and Sampler for Replicate No.2
Measurements at Rubidoux
Appendix E: Time Series Plots for All Species Measured in the 4-City Study
Appendix F: Regression Analysis Plots for All Species Measured in the 4-City Study
Appendix G: Absolute Differences Between the FRM, (Reference Sampler), and the Speciation
Samplers by Site and By Sampling Period
Appendix Gl: Percent Differences Between the FRM, (Reference Sampler), and the Speciation Samplers
by Site and By Sampling Period
Appendix H: Field Evaluation of a Spiral and Cyclonic PM2 5 Size Selective Separator for the Metone
Ambient Chemical Speciation Sampler-SASS
Appendix I: Evaluation of PM2 5 Size Selectors Used in Speciation Samplers (Peters et al. 2000)
Appendix J: Comparison of Paniculate Organic and Elemental Carbon Measurements Made with the
IMPROVE and NIOSH Method 5040 Protocols
Part I, Page i
-------
DRAFT 9 Nov. 1998
Appendix A
Sample Analysis Methods for Chemical Speciation
from
PARTICIPATE MATTER (PM2 5)
SPECIATION GUIDANCE
DOCUMENT
DRAFT
13 July 1998
US Environmental Protection Agency
Monitoring and Quality Assurance Group
Emissions, Monitoring, and Analysis Division
Office of Air Quality Planning and Standards
Research Triangle Park, NC 2771
-------
DRAFT 9 Nov. 1998
7.0 SAMPLE ANALYSIS METHODS
With the goal of ensuring data quality for several thousand filter analyses per year which may
utilize a range of analytical methods, it is critical to establish an analytical laboratory framework which is
consistent to support these needs. One key aspect of the framework is the establishment of a
standardized of SOPs for speciation analyses. The SOPs must be based on analytical methods with
proven application to the analysis of ambient particulate matter filter samples. In addition to guidance
on sampling and handling, the EPA will also develop guidance and documentation for SOPs on the
laboratory analysis of the target analytes given in Table 3-1 of the Guidance Document (EPA, 1999).
The guidance will include laboratory quality assurance guidelines specific to the methods of analysis and
guidelines on standardized data reduction, validation, sample handling, chain-of-custody, and reporting
formats. The EPA plans for the speciation data to be submitted to the Aerometric Information
Retrieval System (AIRS) data base. The U.S. EPA is developing laboratory SOPs for the NAMS
that are complementary to the techniques used by various agencies and research groups operating
ambient air paniculate matter speciation programs.
The operational schedule for providing the speciation laboratory support services begins with
development of a program team of EPA Office of Air Quality Planning and Standards (OAQPS) and
Regional Office personnel. The OAQPS will lead the development of the required guidance
information and the SOPs with input from the Office of Research and Development (ORD), EPA
Regional Offices, and the PM2 5 monitoring technical community. Initial deployment of approximately
50 of the speciation monitors is projected for the second quarter of calendar year 1999. Allowing for
time by site operators to be trained in the use of the monitors, we project that the laboratory services
support portion of the National PM2 5 Speciation Program will be in place by May of 1999. The
speciation analytical guidelines developed by EPA can also be used by State laboratories with
appropriate facilities to provide local support, if appropriate, for an individual State-level monitoring
network. In addition EPA is studying the process of evaluating alternative test methodology which may
be proposed by States. In such a process, an equivalency methodology guideline would have to be
developed that can be used by States to prove that their proposed method performs well as defined by
the guidelines.
The methods used for analyses of the filter media include gravimetry (electro-microbalance) for
mass and various instrumental methods for determining the chemical composition of the particles. In
addition to chemical analyses, special measurement needs may include determining particle size and
morphology through optical and/or electron microscopy.
-------
DRAFT 9 Nov. 1998
A national laboratory services support program consisting of contracted services with up to
three qualified laboratories capable of providing consistent laboratory service support for all of the
target analytes will be developed. The extent of the services will depend upon capacity needed, as
well as the level of participation by State and local government laboratories for providing analytical
services. Analytical support from the contract laboratories will be accessed through three EPA
Regional Project Officers or Delivery Order Project Officers (DOPOs) who will be located at
Regional Offices in the eastern (Region 1), Midwestern (Region 5), and western (Region 8) parts of the
country. The Project Officers will coordinate site needs for those States located in the three
geographic areas. Selected EPA Regional laboratories will operate in a quality assurance role for the
contractor laboratories involved in the speciation program. A flow diagram of the sample analysis
delivery order process is given in Figure A-l.
-------
DRAFT
9 Nov. 1998
Routine
Routine ^.
DOPO
Prepares
Delivery
Order
QA
f
Sample
Collection by
States
K
x Sample
\vType?^
S^
Figure A-l. Flow Diagram of the Sample Analysis Delivery Order Process
-------
DRAFT 9 Nov. 1998
The most commonly applied aerosol analyses methods can be divided into the following categories:
mass, elements, ions, and carbon. It is possible to obtain several different analyses from the same
substrate, but not possible to obtain all desired chemical species from a single substrate; therefore, the
appropriate filter media, sampling hardware, and analysis methods must be combined. Depending on
the study objectives and sources in an area, different chemical species may need to be added or
omitted. A flow diagram of filter processing and analysis activities for the NAMS is shown in Figure
A-2.
The following sections outline the filter analysis methods for the target chemical species
categories of elements, ions, and organic carbon.
7.1 Elemental Analysis
Energy dispersive X-ray fluorescence (EDXRF) is the analytical method of choice to
characterize the elemental composition of the aerosol deposits on PTFE filters for the PM2 5 NAMS
chemical speciation program. Interest in elemental composition is commonly derived from concerns
about health effects and the utility of these elements to trace the sources of suspended particles or
source characterization. Since sample filters often contain very small amounts of particle deposits,
preference is given to methods that can accommodate small sample sizes and require little or no sample
preparation or operator time after the samples are placed into the analyzer. X-ray fluorescence (XRF)
meets these needs and leaves the sample intact after analysis so it can be submitted for additional
examinations by other methods as needed. To obtain the greatest efficiency and sensitivity, XRF
typically places the filters in a vacuum which causes volatile compounds (nitrates and organic
compounds) to evaporate. Volatilization will not be an issue for the NAMS since PTFE filters will not
be subsequently analyzed for volatile species.
In XRF the filter deposit is irradiated by high energy X-rays which causes the ejection of inner
shell electrons from the atoms of each element in the sample. When a higher energy electron drops
into the vacant lower energy orbital, a fluorescent X-ray photon is released. The energy of this photon
is unique to each element, and the number of photons is proportional to the concentration of the
element. Concentrations are determined by comparing photon counts for a sample with those obtained
from thin-film standards of known concentration. XRF methods can be broadly divided into two
categories: wavelength dispersive XRF (WDXRF), which utilizes crystal diffraction for observation of
fluorescent x-rays, and energy dispersive x-ray fluorescence (EDXRF), which uses a silicon
semiconductor detector.
-------
DRAFT 9 Nov. 1998
The WDXRF method is characterized by high spectral resolution which minimizes peak
overlaps. WDXRF requires high power excitation to overcome low sensitivity which results in
excessive sample heating and potential degradation. Conversely, EDXRF features high sensitivity but
less spectral resolution, requiring complex spectral deconvolution procedures.
The type of filter is important and thin membrane filters (PTFE) are required so that the
background is low and penetration of particles into the matrix of the filter is small. The XRF provides
rapid, simultaneous, and nondestructive detection of the target elements from Na to Pb. Advantages of
using XRF are the quantitative analysis of bulk elemental composition, the ability to perform trace level
paniculate analysis with sensitivity to ppm levels, and the availability of instrumentation. The XRF
method of analysis employs an energy dispersive spectrometer. Analysis atmospheres are selectable
with choices of helium or air; helium is used for all target elements except Gd where air is employed
because it gives a lower background. Vacuum is not used, therefore, avoiding the loss of volatile
compounds.
Calibration is by far the most complicated task in the operation of the XRF. Re-calibration is
only performed when a change in fluorescors or x-ray tubes is made or a serious instrument malfunction
occurs. Three types of XRF standards are available for calibration: vacuum deposited thin film
elements; polymer films; and NIST thin-glass films. The vacuum deposited thin film standards are
available for almost all elements analyzed and are used to establish calibration curves. Some
standards have high inherent volatility and do not serve well as calibration standards. These are
selenium (Se), bromine (Br), mercury (Hg), and elemental arsenic (As). The same set of standards is
used every time the spectrometer is calibrated. These standards are sufficiently durable to last many
years. Polymer films contain two elements in the form of organo-metallic compounds dissolved in the
polymer as described in Dzubay et al, 1988. These standards are available for elements with atomic
numbers above 21 (titanium or heavier). The polymer films and NIST standards are typically used for
quality control measures. NIST produces reference materials for iron, lead, potassium, silicon,
titanium, and zinc (SRM 1833).
The sensitivity of this method is on the order of few ng/rr^ for 24-hour samples (flow rates of
10-20 liters per minute). Nonetheless, quite often environmental samples have elemental
measurements below the detection limit of this method. Thus, analytical uncertainties can have a
-------
DRAFT 9 Nov. 1998
significant impact on the quality of the data analysis such as for source apportionment studies.
It should be mentioned that, during the analysis using XRF, the sample is introduced into a chamber that
is evacuated and the sample is slightly heated due to the absorption of X-rays or protons. As a result,
species that can volatilize such as ammonium nitrate and certain organic compounds can be lost during
the analysis. This volatilization is important if the Teflon? filter is to be subjected to subsequent
analyses.
-------
DRAFT
9 Nov. 1998
Mass
PTFE (Teflon)
Carbon: Total,
Organ c,
Elemental,
Carbonate
1
Elements:
Aluminum - Lead
Quar z
1
1
Acceptance
Testing
Equilibration ^
Pre-sampling
Gravimetry
PTFE (Teflon)
Posf-samp//no
Mass
X-Rav
Fluorescence
Aluminum - Lead
Pre-fir ng
Acceptance
Testing
jXRefrigeratecrS.
^r Storage >^
Cations: Ammonium,
Sodium, Potassium
Sulfate
1 Nylon
Washing
Transfer to the
Field
Field Sampling
Transfer to the
Labora ory
Quartz
Thermal Ootical
Analysis
Total, Organic,
Elemental,
Carbonate Carbon
Data Validation
Acceptance
Nylon
Filter Extraction
Ion
Chromatoaraohv
Ammonium,
Sodium, Potassium,
Nitrate, Sulfate
.XRefrigeratecNv
^r Storage & X^
^r Archive for x^
>^ Re -analysis x.
f ^
I Data base,
I AIRS-AQS Input
)
Figure A-2. Flow Diagram of Filter Processing and Analysis Activities for the NAMS
-------
DRAFT 9 Nov. 1998
7.2 Ion Analysis
Aerosol ions refer to chemical compounds which are soluble in water. The water-soluble
portion of suspended particles associates itself with liquid water in the atmosphere when relative
humidity increases, thereby changing the light scattering properties of these particles. Different
emissions sources may also be distinguished by their soluble and non-soluble fractions as in the case of
soluble potassium. Gaseous precursors can also be converted to their ionic counterparts when they
interact with chemicals impregnated on the filter material.
Polyatomic ions such as sulfate, nitrate, ammonium, and phosphate are quantified by methods
such as ion chromatography (1C). Simple ions, such as chloride, and fluoride may also be measured
by 1C along with the Polyatomic ions. When the aerosol deposit is suspected of being acidic, its
hydrogen ion content can be determined by a pH electrode or by micro titration. It is important to
keep the filter away from ammonia sources, such as human breath, to minimize neutralization of the
acidic compounds.
Applied to aerosol samples, the anions and cations are most commonly analyzed by 1C. 1C
can be used for anions (fluoride, phosphate, chloride, nitrate, sulfate) and cations (potassium,
ammonium, sodium) by employing separate columns. All ion analysis methods require a fraction of
the filter to be extracted in deionized distilled water and then filtered to remove insoluble residues prior
to analysis. The extraction volume should be as small as possible to avoid over-diluting the solution
and inhibiting the detection of the desired constituents at levels typical of those found in ambient PM2 5
samples. 1C is the method of choice for the PM2 5 NAME speciation program for the analysis of the
target cations (ammonium, sodium, and potassium) and anions (nitrate and sulfate).
A major sampling requirement for analysis of water-soluble species is that the filter material be
hydrophilic, allowing the water to penetrate the filter and fully extract the desired chemical
components. A nylon filter is used for volatilized and particulate nitrate and sulfate anions and a pre-
fired quartz fiber filter is used for the ammonium, sodium, and potassium cations. The anions are
extracted with a sodium carbonate/sodium bicarbonate solution, and the cations are extracted with a
dilute hydrochloric acid/distilled water solution.
In 1C, the sample extract passes through an ion-exchange column which separates the ions for
individual quantification, usually by a electro-conductivity detector. The anions are separated when
passed through a resin consisting of polymer beads coated with quaternary ammonium active sites.
-------
DRAFT 9 Nov. 1998
The separation is a result of the different affinities of the anions for these sites. After separation and
prior to detection, the column effluent and anions enter a suppressor column where the cations are
exchanged for H+ ions. Species are then detected as their acids by a conductivity meter. The ions
are identified by their elution/retention times and are quantified by the conductivity peak area or peak
height. The 1C is especially desirable for particulate samples because it provides results for several
ions with a single analysis, low detection limits, and uses a small portion of the filter extract with low
detection limits.
The cations are analyzed in the same manner except the sample extract passes through a
surface-sulfonated ion exchange resin where separation occurs. After separation and prior to
detection, the cations enter a suppressor column where all the anions are exchanged for OH" ions.
The species are then detected as their bases (hydroxides) by a conductivity meter. Concentrations of
ions and cations are proportional to the conductivity changes.
-------
DRAFT 9 Nov. 1998
7.3 Particulate Organic Speciation
Organic compounds are important components of particulate matter, whether in urban, rural,
or remote areas. Most of the particulate organic carbon is believed to reside in the fine particle
fraction. PM2 5 samples were collected at four urban locations in southern California in 1982 to
quantify several individual organic compounds. This study identified and quantified over 80 individual
organic compounds in the PM2 5 fraction, including n-alkanes, n-alkanoic acid, one n-alkenoic acid,
one n-alkanal, aliphatic dicarboxylic acids, aromatic polycarboxylic acids, polycyclic aromatic
hydrocarbons, polycyclic aromatic ketones, polycyclic aromatic quinones, diterpenoid acids, and
some nitrogen-containing compounds. In general, the same type of organic compounds, although in
different proportions, are found in direct emissions from various sources such as diesel and gasoline
powered vehicle exhaust, charbroilers and meat cooking operations, cigarette smoke, biogenic
sources, etc. Thus, organic compounds are potentially valuable tracers for characterizing different
emission sources as well as for atmospheric transformation processes.
7.3.1 Carbon Analysis
Three classes of carbon are typically measured in ambient aerosol samples collected on pre-
fired quartz-fiber filters: 1) organic, volatilized, or non-light absorbing carbon; 2) elemental or
light-absorbing carbon; and 3) carbonate carbon. Carbonate-source carbon [i.e., potassium
carbonate (K2CO3), sodium carbonate (Na2CO3), magnesium carbonate (MgCO3), calcium
carbonate (CaCO3)] is determined from a separate filter section after acidification. Without
acidification, the determination of carbonate carbon is not specific and is detected as either organic or
elemental carbon.
Two thermal-optical methods currently are in use for the analysis of carbonaceous aerosols.
The measurement principle is fundamentally the same, but the methods differ with respect to
calibration, analysis time, temperature ramping and settings, types of carbon speciated, and pyrolysis
correction technique. The method=s pyrolysis correction feature allows correction for the char that
forms on the filter during analysis of some materials (e.g., cigarette smoke). Correction for pyrolysis
is made by continuously monitoring the filter transmittance (NIOSH Method 5040) or reflectance
(TOR) throughout the analysis.
For the NAMS chemical speciation program, total, organic, elemental, and carbonate carbon
will be determined by thermal/optical instrumentation specified in NIOSH Method 5040 (NIOSH,
-------
DRAFT
9 Nov. 1998
Hrtfc
Figure
A-3. Schematic of the Thermal-optical Analyzer Evaluated by NIOSH
-------
DRAFT 9 Nov. 1998
1996, 1998) and described in the literature (Birch and Gary, 1996). Method 5040 was developed
for occupational monitoring of diesel paniculate, but its evaluation also included a round-robin study
involving a variety of carbonaceous aerosols (Birch, 1998). The thermal-optical method is applicable
to nonvolatile, carbon-containing species only. Thermal-optical analyzers are practical, economical,
and are routinely used for environmental and occupational monitoring of carbonaceous aerosols.
Although the number of commercial laboratories currently is limited, an adequate number (3 in the
U.S.) are available to cover the analytical demands of this program.
Thermal-optical analyzers operate by liberating carbon compounds under different
temperature and oxidation environments. A small portion (or punch) is taken from a quartz-fiber filter
sample and placed in the sample oven. The oven is purged with helium and the temperature is then
stepped to a preset value. Volatilized compounds are converted to carbon dioxide (CO2) in an
oxidizer oven (MnO2 at 870C or higher), the CO2 is subsequently reduced to methane (CHt) in a
methanator (nickel-impregnated firebrick heated to -550C in a stream of hydrogen), and CHj is
quantified by a flame ionization detector (FID). In the second part of the analysis, an oxygen-helium
mix is introduced and the remaining carbon is removed through combustion and quantified in the same
manner. A schematic of the thermal-optical analyzer evaluated by NIOSH researchers is shown in
Figure A-3; an example of the instruments output, called a "thermogram," is given in Figure A-4.
In NIOSH Method 5040, the sample oven is purged with helium and the temperature is
stepped (to 250, 500, 650 and 850C) to volatilize the organic and carbonate-source carbon. It is
critical to ensure that trace oxygen is not present during the first part of the analysis. Potential sources
of oxygen include leaks and inadequate helium gas purification. If present, trace oxygen will cause
organic carbon to be overestimated and elemental carbon correspondingly underestimated. In the
second part of the analysis, the temperature is lowered, a 2% oxygen/98% helium mix is introduced,
and the temperature is then stepped to a maximum of 940C. At the end of the analysis, a calibration
gas standard (Cttj) is injected. Correction for pyrolysis is made by continuously monitoring the filter
transmittance throughout the analysis. The point at which the filter transmittance returns to its original
value is defined as the split between organic and elemental carbon. Designation of the split in this
manner allows assignment of pyrolyzed organic carbon to the organic fraction.
-------
DRAFT
9 Nov. 1998
DC-EC spit
Figure A-4.. Sample Output of Thermal-Optical Method Showing Different Forms of Carbon
In general, thermal-optical methods classify carbon as >organic= or >elemental.= Organic
carbon is non-light absorbing carbon that is volatilized in helium as the temperature is stepped to a
preset maximum (850C). Elemental carbon is light-absorbing carbon (elemental or organic) and any
non-light absorbing carbon evolved after pyrolysis correction. Depending on the sampling
environment, carbonates [e.g., potassium carbonate (K2CO3), sodium carbonate (Na2CO3),
magnesium carbonate (MgCO3), calcium carbonate (CaCO3)] also may be present in the sample.
Carbonate-source carbon is quantified as organic by NIOSH Method 5040, wherein the sample is
exposed to 850C during the first part of the analysis (i.e., in helium only). Under these conditions,
thermal decomposition of carbonate occurs. To quantify carbonate carbon, a second portion of the
filter sample is analyzed after its acidification. Carbonate is taken as the difference between the pre-
and post-acidification results (Note: the approach assumes a homogeneous filter deposit).
Alternatively, carbonate carbon in a simple carbonate (i.e., not a bicarbonate) can be estimated by
integrating the carbonate peak (typically the fourth peak in >thermogram). The carbonate peak can
be integrated separately within the instruments calculation software. This approach normally is taken
-------
DRAFT 9 Nov. 1998
when higher loadings of carbonate are present. In the case of the TOR method, carbonate in a filter
portion is determined through on-line measurement of the carbon dioxide (CO2) evolved upon
acidification. Acid (0.04 M HC1) is injected directly onto the filter portion through an injection port.
The sample oven is kept at ambient temperature while CO2 is evolved, reduced to methane and
quantified. (Note: This approach is subject to potential interference of adsorbed CO2). Typically,
carbonate carbon is not speciated in environmental samples because it has been found to constitute
less than 5 percent of the total carbon in most samples (Chow et al., 1993).
It is important to remember that elemental and organic carbon have meaning only in the
operational sense. That is, results reflect the method used and the appropriateness of a method
depends on its purpose. Operational methods differ from those used for specific, identifiable analytes
(e.g., sulfate or sulfur), where a well-defined entity is quantified and laboratory standards are available
for its determination. Because elemental and organic carbon are defined operationally, the details of
the measurement method must be rigorously prescribed.
Only one organic and one elemental carbon fraction are reported in NIOSH 5040 (total
carbon is the sum of these two). In contrast, four types of organic carbon and three types of
elemental are defined by the TOR technique. In both instances, different classes of carbon are
evolved from the sample during the analysis. In the case of Method 5040, the division into two
fractions reflects the purpose of the method (i.e., occupational monitoring of diesel particulate). For
other applications (e.g., source apportionment), additional fractions may be appropriate provided that
the applied temperature program is repeatable over time. Otherwise, relative information will not be
meaningful because non-constant analytical parameters can affect the classification of carbon types.
7.3.2 Semi-volatile Organic Aerosols
Semi-volatile organic aerosols should not be considered on a routine basis due to the non-
routine and research-oriented nature of measuring these species. Identification of the ideal denuder,
filter combination, and sorbents and development of routine sampling and analytical methods is
complicated due to the variety of semi-volatile organic aerosol compounds in the atmosphere and their
varying absorptive properties.
The collection of particulate organic matter can be accomplished using particulate sampling
instruments equipped with quartz fiber or TeflonT-impregnated glass fiber filters. However, since
many organic compounds are distributed between the gas and particle phases, additional sampling
techniques are required to measure the particle phase semi-volatile organic compounds. This
-------
DRAFT 9 Nov. 1998
methodology is susceptible to negative (desorption of semi-volatile compounds from the particles on
the filters) and positive (adsorption of gases by the filter material) artifacts. Considerable experimental
and theoretical effort has been expended to understand and correct for these vaporization and
condensation effects. Denuder technology has been employed to provide a less artifact-encumbered
approach for accurate determination of semi-volatile species because the gas phase is removed prior
to the particulate phase. A sorbent or denuder after the filter may also be employed to collect any
semi-volatile material desorbed from the filter.
For quantification of individual organic compounds, the denuder, filter, and sorbent is
extracted individually with a suitable organic solvent (or a combination of solvents). The extract is
then analyzed by gas chromatography (GC) combined with mass spectrometry (MS) or with other
specific detectors. Combined GC/Fourier transform infrared (FTIR)/MS techniques or high
performance liquid chromatography (FtPLC)/MS techniques are also used.
-------
Appendix B
Standard Operating Procedures
Chemical Speciation Sampler Intercomparison Study
Standard Operating Procedures for this study have not been
reproduced here to save paper, but can be found in the 4-City Study
QAPP. A listing of the SOPs can be found in Table 1-2.
Vanderpool, R. 1999. Quality Assurance Project Plan for the Four-
City PM2.5 Chemical Speciation Sampler Evaluation Study. Prepared
by Research Triangle Institute, RTI Project Number: 07263-030, EPA
Funding Number: 68-D5-0040, WAIH-30 and 111-33. RTI, Research
Triangle Park, NC.
-------
Appendix C
Summary of Comments from Field and
Laboratory Analysis Logbooks
Prepared by RTI
RTF, NC
F:\USER\BMILLAR\amticweb\fourcities\03 Appendix C, Comments(EPA12).wpd (14MaiOO)
-------
Date
Sampler Repeat Comments
01/06/99 AND 1 MgO denuders were not used in chl. Sampler displays high
flow rate ch4 (9.9 L/m).
01/06/99 AND 2 MgO denuders were not used in chl. Sampler displays high
flow rate ch4 (8.2 L/m).
01/09/99 AND 1 MgO denuders were not used in chl. Sampler displays high
Flow rate ch4 (9.9 L/m).
01/09/99 AND 2 Void-Sampler lost all calibrations.
01/15/99 AND 1 MgO denuders were used for remainder of study. Teflon filter (chl) has a hole near the ring-did not affect mass cone. Hole
occurred after sampling. Sampler displays high. Flow rate ch4 (8.9 L/m).
2 MgO denuders were used for remainder of study. Andersen sent rep to site to re-calibrate prior to this run.Sampler displays
high flow rate ch4 (7.6 L/m). XRF Data: "As" is below detection limits (BDL).
1 Sampler displays low flow rate ch4 (6.6 L/m). XRF Data: "As" is BDL.
2 Sampler displays low flow rate chl (6.0 L/m). Sampler
displays high flow rate ch4 (7.8 L/m).
1 Sampler displays low flow rate ch4 (4.1 L/m).
XRF Data: "As" is BDL.
2 Site op noted some water in the back near the top of the
back panel. Sampler displays high flow rate ch2 (17.7 L/m) and ch4 (7.9 L/m). XRF Data: "Mn" is BDL.
1 Sampler displays low flow rate ch4 (4.2 L/m).
2 Sampler displays low flow rate chl (6.1 L/m). Sampler
displays high flow rate ch2 (17.7 L/m) and ch 4 (7.9 L/m).
1 Sampler displays high flow rate chl (7.6 L/m). Sampler
displays low flow rate ch4 (4.2 L/m).
2 Sampler displays low flow rate chl (6.3 L/m). Sampler
displays high flow rate ch2 (17.9 L/m) and ch4 (8.0 L/m).
1 Sampler displays low flow rate ch4 (4.3 L/m).
2 Sampler displays low flow rate chl (5.8 L/m). Sampler
displays high flow rate ch2 (18.0 L/m) and ch4 (7.9 L/m).
1 Sampler displays low flow rate ch3 (15.6 L/m). Sampler
displays high flow rate ch4 (7.7 L/m).
2 Sampler displays low flow rate chl (6.4 L/m). Sampler
displays high flow rate ch4 (7.7 L/m).
1 Sampler displays low flow rate ch2 (16.0 L/m) and
ch3 (15.5 L/m) and ch4 (7.0 L/m). XRF Data: "As" is BDL.
2 Sampler displays low flow rate chl (6.1 L/m) and
ch3 (16.0 L/m). High flow rate ch4 (7.7 L/m).
1 Sampler displays high flow rate chl (7.8 L/m) and ch4 (7.9L/m). Sampler displays low flow rate ch3 (15.3 L/m).
XRF Data: "Pb"isBDL.
2 Sampler displays low flow rate chl (5.9 L/m). Sampler displays high flow rate ch4 (7.9 L/m). XRF Data: "Pb" and "As" are
BDL.
1 Samplerdisplayslowflowratech2(16.0L/m)andch3 (15.4 L/m). XRF Data: "As" isBDL.
2 Sampler displays low flow rate chl (6.3 L/m) and high flow flow rate ch4 (7.8 L/m).
1 Sampler displays low flow rate ch2 (15.5 L/m) and ch4 (7.0 L/m).
2 Sa in ch3-void SO4+2, NO3-, NH4+ data.
1 Sampler displays low flow rate ch3 (15.6 L/m) and ch4 (7.0 L/m). XRF Data: "As" is BDL.
2 Sampler displays low flow rate chl (6.2 L/m) and ch3 (16.0 L/m) and high flow rate ch4(7.8L/m). XRF Data: "As" is BDL.
1 Large tear and 1 small hole in Teflon (Chl) filter-did not
affect mass cone. Tear/hole occurred after sampling.Sampler displays low flow rate ch3 (15.7 L/m). XRF Data: Did not analyze
filter due to large tear.
2 Sampler displays low flow rate chl (5.9 L/m) and ch3(16.0 L/m) and high flow rate ch4 (7.7 L/m).
1 Sampler displays low flow rate chl (15.7 L/m) and ch4 (6.9 L/m). XRF Data: "As" is BDL.
2 Sampler displays low flow rate chl (6.3 L/m) and ch3 (15.9 L/m) and ch4 (14.3 L/m). Site op verified ch4 flow, sampler reading
13.2L/m with the pump off. XRF Data: "As" is BDL.
01/18/99 AND
01/18/99 AND
01/21/99 AND
01/21/99 AND
01/24/99 AND
01/24/99 AND
01/27/99 AND
01/27/99 AND
01/30/99 AND
01/30/99 AND
02/02/99 AND
02/02/99 AND
02/05/99 AND
02/05/99 AND
02/11/99 AND
02/11/99 AND
02/14/99 AND
02/14/99 AND
02/17/99 AND
02/17/99 AND
02/20/99 AND
02/20/99 AND
02/23/99 AND
02/23/99 AND
02/26/99 AND
02/26/99 AND
01/06/99 FRM-T
01/06/99 FRM-T
01/09/99 FRM-T
01/09/99 FRM-T
01/15/99
01/15/99
01/18/99
01/18/99
01/21/99
01/21/99
01/24/99
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
XRF Data: "As" is below detection limits (BDL).
Lab op noted 2 small holes—did not affect mass cone. Holes seems to have occurred subsequent to sampling. XRF Data:
"As" is BDL.
XRF Data: "As" is BDL.
T.Miguel arrived at the site @0840 during the sample and found the filter in the down position. XRF Data: "As" is BDL.
Small puddles of H2O found inside on the impactor shelf and bottom of the cabinet. XRF Data: "As" is BDL.
XRF Data: "As" is BDL.
No sample-sample did not run because during the firmware upgrade the sampler shut down and would not power back up.
F:\USER\BMILLAR\amticweb\fourcities\03 Appendix C, Comments(EPA12).wpd (14MaiOO)
-------
1 01/24/99 FRM-T
1 01/27/99 FRM-T
1 01/27/99 FRM-T
1 01/30/99 FRM-T
1 01/30/99 FRM-T
1 02/02/99 FRM-T
1 02/02/99 FRM-T
1 02/05/99 FRM-T
1 02/05/99 FRM-T
1 02/11/99 FRM-T
1 02/11/99 FRM-T
1 02/14/99 FRM-T
1 02/14/99 FRM-T
1 02/17/99 FRM-T
1 02/17/99 FRM-T
1 02/20/99 FRM-T
1 02/20/99 FRM-T
1 02/23/99 FRM-T
1 02/23/99 FRM-T
1 02/26/99 FRM-T
1 02/26/99 FRM-T
1 01/06/99 FRM-Q
1 01/06/99 FRM-Q
1 01/09/99 FRM-Q
1 01/09/99 FRM-Q
1 01/15/99 FRM-Q
1 01/15/99 FRM-Q
1 01/18/99 FRM-Q
1 01/18/99 FRM-Q
1 01/21/99 FRM-Q
1 01/21/99 FRM-Q
1 01/24/99 FRM-Q
1 01/24/99 FRM-Q
1 01/27/99 FRM-Q
1 01/27/99 FRM-Q
1 01/30/99 FRM-Q
1 01/30/99 FRM-Q
1 02/02/99 FRM-Q
1 02/02/99 FRM-Q
1 02/05/99 FRM-Q
1 02/05/99 FRM-Q
1 02/11/99 FRM-Q
1 02/11/99 FRM-Q
1 02/14/99 FRM-Q
1 02/14/99 FRM-Q
1 02/17/99 FRM-Q
1 02/17/99 FRM-Q
1 02/20/99 FRM-Q
1 02/20/99 FRM-Q
1 02/23/99 FRM-Q
1 02/23/99 FRM-Q
1 02/26/99 FRM-Q
1 02/26/99 FRM-Q
1 01/06/99 IMP
1 01/06/99 IMP
EC/OC.
1 01/09/99 IMP
1 01/09/99 IMP
1 01/15/99 IMP
1 01/15/99 IMP
1 01/18/99 IMP
1 01/18/99 IMP
1 01/21/99 IMP
1 01/21/99 IMP
1 01/24/99 IMP
2
1
2
1
2
i
2
1
2
i
2
1
2
i
2
1
2
i
2
1
2
i
2
1
2
i
2
1
2
i
2
1
2
i
2
1
2
i
2
1
2
i
2
1
2
i
2
1
2
i
2
1
2
i
2
i
2
1
2
i
2
i
2
1
Firmware upgraded on 1/22. XRF Data: "As" isBDL.
No sample—sampler still will not power up.
T.Miguel was able to get the sampler running—firmware has not be upgraded.
Teflon filter appears to have been scratched w/ forceps-did not affect the mass cone. XRF Data: "As" is BDL.
XRF Data: "Cu" and "As" are BDL.
XRF Data: "Cu", "Pb", and "As" are BDL.
XRF Data: "As" is BDL.
XRF Data: "As" is BDL.
XRF Data: "As" is BDL.
XRF Data: "As" is BDL.
XRF Data: "As" is BDL.
Installed clean wins impactors. XRF Data: "As" is BDL.
Installed clean wins impacotrs. XRF Data: "Cu" and "As" are BDL.
EC/OC Data: Non-white carbon punch after carbon analysis, indicative of mineral particles in deposit.
EC/OC Data: Non-white carbon punch after carbon analysis, indicative of mineral particles in deposit.
EC/OC Data: Non-white carbon punch after carbon analysis, indicative of mineral particles in deposit.
Sm all puddles of H2O found near the wins, near the locking m echanism and in the bottom of the cabinet.
Firmware upgraded on 1/22.
No Sample-following the firmware upgrade, the sampler would not power up.
Sampler ran 34 hours due to programming error.
No Sample-sampler ran 3 seconds.
Sampler running again-firmware has not been upgraded.
No Sample-sampler did not run, may need new pump.
BGI sent a leaner FRM to replace downed unit. Checked for leaks, cleaned wins. Sampler was not audited.
Average flow rate is low (16.4 L/m).
Installed clean wins impactor.
Installed clean wins impactor.
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is below detection limits (BDL).
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL. Quartz filter was not transferred to DRI for
Void-large hole in filter. Hole seems to have occurred before/during sampling.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules. EC/OC Data: Non-white carbon punch after carbon analysis, indicative
of mineral particles in deposit.
Used avg flow rates from the site audits for 3 modules. Pump timer appears to be stuck at the off trigger.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules. Timer for the pumps is broken. Site op is manually turning the sampler
off and on. XRF Data: "As" is BDL.
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL.
Used avg flow rates from the site audits for 3 modules.
-------
1 01/24/99 IMP
1 01/27/99 IMP
1 01/27/99 IMP
1 01/30/99 IMP
1 01/30/99 IMP
1 02/02/99 IMP
1 02/02/99 IMP
1 02/05/99 IMP
1 02/05/99 IMP
1 02/1 1/99 IMP
1 02/1 1/99 IMP
1 02/14/99 IMP
1 02/14/99 IMP
1 02/17/99 IMP
1 02/17/99 IMP
1 02/20/99 IMP
1 02/20/99 IMP
1 02/23/99 IMP
1 02/23/99 IMP
1 02/26/99 IMP
1 02/26/99 IMP
1 01/06/99 MET
2
1
2
1
2
i
2
1
2
i
2
1
2
i
2
1
2
i
2
i
2
1
1 01/06/99 MET
1 01/09/99 MET
1 01/09/99 MET
1 01/15/99 MET
1 01/15/99 MET
1 01/18/99 MET
1 01/18/99 MET
1 01/21/99 MET
1 01/21/99 MET
1 01/24/99 MET
1 01/24/99 MET
1 01/27/99 MET
1 01/27/99 MET
1 01/30/99 MET
1 01/30/99 MET
1 02/02/99 MET
1 02/02/99 MET
1 02/05/99 MET
1 02/05/99 MET
1 02/11/99 MET
1 02/11/99 MET
1 02/14/99 MET
1 02/14/99 MET
1 02/17/99 MET
1 02/17/99 MET
1 02/20/99 MET
1 02/20/99 MET
1 02/23/99 MET
2
1
2
1
2
1
2
i
2
1
2
1
2
i
2
1
2
i
2
i
2
1
2
i
2
1
2
i
02/26/99 MET
02/26/99 MET
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules. XRF Data: "Pb" and "As" are BDL. EC/OC Data: Non-white carbon
punch after carbon analysis, indicative of mineral particles in deposit.
Used avg flow rates from the site audits for 3 modules. XRF Data: "Cu", "Pb", and "As" are BDL.
Used avg flow rates from the site audits for 3 modules.XRF Data: "As" is BDL.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL.
Used avg flow rates from the site audits for 3 modules. EC/OC Data: Non-white carbon punch after carbon analysis,
indicative of mineral particles in deposit. XRF Data: "As" is BDL.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL.
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL.
Void-pin hole near ring of Teflon (chl). Hole seems to have occurred during/before sampling began. Sampler displays
high sample volume on ch3(10.214 m3) and ch4(l 0.605 m3). EC/OC Data: Non-white carbon punch after carbon analysis,
indicative of mineral particles in deposit.
Sampler displays low sample volume ch3 (8.588 m3) and ch4 (8.966 m3). XRF Data: "As" is below detection limits (BDL).
EC/OC Data: Non-white carbon punch after carbon analysis, indicative of mineral partilces in deposit.
Sampler displays high sample volume chl(10.791 m3) and ch2 (10.720 m3), ch4(10.676 m3). XRF Data: "Pb" is BDL; and
"As" is BDL. EC/OC Data: Non-white carbon punch after carbon analysis, indicative of mineral particles in deposit.
Sampler displays low sample volume ch3 (8.651m3) and ch4 (9.101 m3).XRFData: "As" is BDL. EC/OC Data: Non-white
carbon punch after carbon analysis, indicative of mineral particles in deposit.
Sampler displays low sample volume ch3 (8.651 m3). XRF Data: "As" is BDL. EC/OC Data: Non-white carbon punch
after carbon analysis,indicative of mineral particles in deposit.
Void-sampler ran 10.3 hours. (Sampler displays sample vol from the previous run.)
Teflon filter (CHI) has a tear and small hole that seems to have been caused by the forceps. Did not affect mass cone. Sampler
displays low sample vol ch3 (9.041 m3). XRFData: "As" is BDL.
Sampler displays low sample vol ch3 (8.810m3) and ch4 (9.059 m3). XRF Data: "As" is BDL.
Sampler displays low sample vol ch3 (9.110m3). XRFData: "Cu" is BDL; "As" is BDL.
Suspect data. Suspect an error in the recording of the filter weights. Sampler displays low sample vol ch3 (8.840 m3) and ch4
(9.070m3).
Sampler displays low sample vol ch3 (9.063 m3). XRF Data: "As" is BDL.
Sampler displays low sample vol ch3 (8.822 m3) and ch4 (9.067m3).
Sampler displays low sample vol ch3 (9.106 m3).
Sampler displays low sample vol ch3 (8.789 m3) and ch4 (9.054 m3).
Sampler displays low sample vol ch3 (9.055 m3).
Sampler displays low sample vol ch3 (8.730 m3) and ch4 (9.061 m3).
Large tear in Teflon (chl)-did not affect mass cone. Appears hole occurred subsequent to sampling. Sampler displays low
sample vol ch3 (9.098 m3). XRF Data: "As" is BDL.
Sampler displays low sample vol ch3 (8.739 m3) and ch4 (9.109 m3).
Large tear in Teflon (chl)-did not affect mass cone. Appears hole occurred subsequent to sampling. Sampler displays low
sample vol ch3 (8.972 m3). XRF Data: "Mn" is BDL.
Sampler displays low sample vol ch3 (8.734 m3) and ch4 (9.059 m3). XRF Data: "As" is BDL.
Sampler displays low sample vol ch3 (9.030 m3). XRF Data: "Cu" and "Pb" are BDL.
Sampler displays low sample vol ch3 (8.749 m3) and ch4 (9.131 m3). XRF Data: "As" is BDL.
Sampler displays low sample vol ch3 (9.068 m3).
Sampler displays low sample vol ch3 (8.752 m3) and ch4 (9.050 m3). XRF Data: "As" is BDL.
Tears @ ring Teflon (chl)-didnot affect mass cone. Appears hole occurred subsequent to sampling. Sampler displays
low sample vol ch3 (9.042 m3). XRF Data: "As" is BDL.
Sampler displays low sample vol ch3 (8.698 m3) and ch4 (9.036 m3). XRF Data: "As" is BDL.
Sampler displays low sample vol ch3 (9.089 m3). XRF Data: "As" is BDL.
Sampler displays low sample vol ch3 (8.473 m3) and ch4 (9.059 m3).
Sampler displays low sample vol ch3 (9.036 m3). EC/OC Data: Non-white carbon punch after carbon analysis, indicative of
mineral particles in deposit.
Sampler displays low sample vol ch3 (6.490 m3) and ch4 (9.097 m3). EC/OC Data: Non-white carbon punch after
carbon analysis, indicative of mineral particles in deposit.
Sampler displays low sample vol ch3 (9.007 m3). XRF Data: "Cu", "Pb", and "As" are BDL.
Fuzz on Teflon filter (chl)-didnot affect mass cone. Sampler displays low sample vol ch3 (8.690 m3) andch4 (9.056m3). XRF
Data: "Mn", "Cu", and "Pb" are BDL.
-------
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
deposit
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
01/06/99 URG
01/06/99 URG
01/09/99 URG
01/09/99 URG
01/15/99 URG
01/15/99 URG
01/18/99 URG
01/18/99 URG
01/21/99 URG
01/21/99 URG
01/24/99 URG
01/24/99 URG
01/27/99 URG
01/27/99 URG
01/30/99 URG
01/30/99 URG
02/02/99 URG
02/02/99 URG
02/05/99 URG
02/05/99 URG
02/11/99 URG
02/11/99 URG
02/14/99 URG
02/14/99 URG
02/17/99 URG
02/17/99 URG
02/20/99 URG
02/20/99 URG
02/23/99 URG
02/23/99 URG
02/26/99 URG
02/26/99 URG
01/06/99 VAPS
01/09/99 VAPS
01/15/99 VAPS
01/18/99 VAPS
01/21/99 VAPS
01/24/99 VAPS
01/27/99 VAPS
01/30/99 VAPS
02/02/99 VAPS
02/05/99 VAPS
02/11/99 VAPS
02/14/99 VAPS
02/17/99 VAPS
02/20/99 VAPS
02/23/99 VAPS
02/26/99 VAPS
01/06/99 AND
01/08/99 AND
01/14/99 AND
1
2
1
2
i
2
i
2
1
2
i
2
1
2
i
2
1
2
1
2
i
2
1
2
i
2
1
2
i
2
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
EC/OC Data: Non-white carbon punch after carbon analysis, indicative of mineral particles in deposit.
FTP Flag-did not affect mass cone. Low sample volume on URG400 (22.591 m3). EC/OC Data: Non-white carbon punch
after carbon analysis, indicative of mineral particles in deposit.
FTP Flag-did not affect mass cone. Low sample volume on URG400 (22.414 m3). XRF Data: "Cu" is below detection limits
(BDL).
XRF Data: "As" is BDL.
Low sample volume on URG400 (22.604 m3). XRF Data: "As" is BDL.
New Dry Gas Meter pumps installed and re-calibrated. Unable to locate Teflon filter. Filter was rec'd by RTI after
sampling; but was not post-weighed.
New Dry Gas Meter pumps installed and re-calibrated. XRF Data: "As" is BDL.
Wins impactors cleaned before this run. XRF Data: "As" is BDL.
Wins impactors cleaned before this run.
Void-sample ran 10.5 hrs (URG400). Mass cone and Particulate NO3- data is void. Outer ring had slight split.
No sample-did not run (URG400). Mass cone and Particulate NO3- data is void. Problem with the display screen.
Void-large tear in Teflon filter-affected mass cone. FTP Flag.
Void-T.Miguel was working on the sampler and lost all data from this sample run when sampler was re-booted (URG400).
Data was suspect anyway-sampler does not appear to have runned (Mass=1.00ug). Mass cone and particulate NO3- data
is void.
FTP Flag-did not affect mass cone. Medium-sized tear in Teflon filter-tear seems to have occurred after sampling.
Void-sample ran 1 second (URG400). Mass cone and particulate NO3- data is void.
Unable to locate Teflon filter. Filter was rec'd by RTI after sampling; but was not transferred for post-weight.
Replaced pump and cleaned wins impactor before this run. XRF Data: "As" is BDL.
XRF Data: "Cu" and "As" are BDL.
FTP Flag. XRF Data: "As" is BDL.
XRF Data: "As" is BDL.
FTP Flag.
XRF Data: "As" is BDL.
FTP Flag. XRF Data: "As" is BDL.
Void-sample ran 2 seconds. TmFlo flag. Bad pump plug connection (URG400). Mass cone and particulate NO3- data is void.
FTP Flag. XRF Data: "As" is BDL.
2 small holes near ring of Teflon filter-did not affect mass cone. Holes appear to have occurred after sampling.
Repaired pin on pump plug. Cleaned wins impactor before this run. XRF Data: "As" is BDL.
Cleaned wins impactor before this run.
Low sample volume chl (2.970 m3) and ch2 (19.058m3). Lab op dropped filter on floor. XRF Data: "As" is below
detection limits (BDL). EC/OC Data: Non-white carbon punch after carbon analysis, indicative of mineral particles in
Void-large hole in filter which affected mass cone—PM2.5 andPMlO. EC/OC Data: Non-white carbon punch after carbon
analysis, indicative of mineral particles in deposit.
Void-filter pack was improperly put together. (Teflon filter was loaded downstream of Nylon filter.)Mass cone and
NO3- data is void. The glass denuder (XAD) was received broken. EC/OC Data: Non-white carbon punch after carbon
analysis, indicative of mineral particles in deposit.
Void-ch2 (teflon)suspect that the weights were written down incorrectly. Ch3 low sample volume (17.031m3). XRF Data: "As" is
BDL.
Ch2 high sample volume (23.084m3). Ch3 low sample volume (19.412m3). XRF Data: "As" is BDL.
High sample vol chl (4.902 m3); Ch3 low sample volume (20.095m3).
Nylon filter was not transferred to CEMQA for 1C analyses. It appears that when an error occurred when unloading the
filterpack—nylon filter was not kept refridgerated. Particulate NO3- data is void. XRF Data: "As" is BDL.
Quartz filter was not transferred to DRI for EC/OC analysis.
XRF Data: "As" is BDL.
XRF Data: "As" is BDL. EC/OC Data: Non-white carbon punch after carbon analysi, indicative of mineral particles in deposit.
Void-PM2.5 data is suspect.
Suspect data-3 small holes appear to have caused by forceps. XRF Data: "As" is BDL.
XRF Data: "As" is BDL.
XRF Data: "As" is BDL.
MgO denuders were not installed. Sampler displays low sample vol chl-Nylon (9.504 m3) and ch2-Teflon (22.752 m3).
MgO denuders were not installed. Sampler displays low sample vol chl (9.504 m3);ch2 (22.752 m3);ch3(22.896 m3).
MgO denuders were installed. Sampler displays low sample vol chl (9.360 m3);ch2 (22.896 m3).
-------
01/16/99 AND
01/18/99 AND
01/20/99 AND
01/22/99 AND
01/24/99 AND
01/26/99 AND
01/28/99 AND
02/03/99 AND
02/05/99 AND
02/07/99 AND
02/09/99 AND
02/11/99 AND
02/13/99 AND
02/15/99 AND
02/17/99 AND
02/19/99 AND
02/21/99 AND
01/06/99 FRM-T 1
01/08/99 FRM-T 1
01/14/99 FRM-T 1
01/16/99 FRM-T 1
01/18/99 FRM-T 1
01/20/99 FRM-T 1
01/22/99 FRM-T 1
01/24/99 FRM-T 1
01/26/99 FRM-T 1
01/28/99 FRM-T 1
02/03/99 FRM-T 1
02/05/99 FRM-T 1
02/07/99 FRM-T 1
02/09/99 FRM-T 1
02/11/99 FRM-T 1
02/13/99 FRM-T 1
02/15/99 FRM-T 1
02/17/99 FRM-T 1
02/19/99 FRM-T 1
02/21/99 FRM-T 1
01/06/99 FRM-Q 1
01/08/99 FRM-Q 1
01/14/99 FRM-Q 1
01/16/99 FRM-Q 1
01/18/99 FRM-Q 1
01/20/99 FRM-Q 1
01/22/99 FRM-Q 1
01/24/99 FRM-Q 1
01/26/99 FRM-Q 1
01/28/99 FRM-Q 1
02/03/99 FRM-Q 1
02/05/99 FRM-Q 1
02/07/99 FRM-Q 1
02/09/99 FRM-Q 1
02/11/99 FRM-Q 1
02/13/99 FRM-Q 1
02/15/99 FRM-Q 1
02/17/99 FRM-Q 1
02/19/99 FRM-Q 1
02/21/99 FRM-Q 1
01/06/99 IMP
1 Teflon (ch2) filter had large tear @ ring. Hole seems to have occurred subsequent to sampling b/c mass cone is in
line with other samplers. Sampler displays low sample vol chl (9.648 m3);ch2 (22.896 m3);ch3 (22.896 m3).
1 Teflon (ch2) filter had pin hole. Hole seems to have occurred subsequent to sampling b/c mass cone is in line with other
samplers. Sampler displays low sample vol chl (9.648 m3).
1 Sampler displays low sample vol chl (9.504 m3).
1 Sampler displays low sample vol chl (9.648m3). XRFData: "As" is Below Detection Limit (BDL).
1 Sampler displays low sample vol chl (9.504 m3).
1 Sampler displays low sample vol chl (9.504m3); ch3 (22.896 m3).
1 Sampler displays low sample vol chl (9.504m3); ch3 (22.896 m3).
1 Sampler displays low sample vol chl (9.360m3). Sampler displays high sample vol ch4 (1 1.232 m3).
1 Sampler displays low sample vol chl (9.504 m3). Sampler displays high sample vol ch4 (1 1.232 m3). XRF Data: "As" is BDL.
1 Sampler displays low sample vol chl (9.504m3). Sampler displays high sample vol ch4 (1 1.232 m3).
1 Sampler displays low sample vol chl (9.504 m3). XRF Data: "As" is BDL.
1 Sampler displays low sample vol chl (9.792 m3). Sampler displays high sample vol ch4 (1 1.520 m3). XRF Data: "Mn'
are BDL.
1 Sampler displays low sample vol chl (9.648 m3). XRF Data: "As" is BDL.
1 Teflon (ch2) filter had a medium-sized tear. Site op noted pin hole on Teflon (ch3) filter. Hole seems to have occurred
subsequent to sampling b/c mass cone is in line with other samplers.
1 Sampler displays low sample vol chl (9.360 m3). XRF Data: "As" is BDL.
1 Teflon (ch2) filter had small tear under ring. Hole seems to have occurred subsequent to sampling b/c mass cone is i
other samplers. Sampler displays low sample vol chl (9.216 m3). XRF Data: "As" is BDL.
1 Sampler displays low sample vol chl (9.360 m3).
1 and "As"
in line with
Lab operator noted large tear in filter. Hole seems to have occurred subsequent to sampling b/c mass cone is in line with other
samplers.
XRF Data: "As" is below detection limit (BDL).
Pin hole near ring. Site op replaced wins impactor well. Hole seems to have occurred subsequent to sampling b/c mass cone is in
line with other samplers.
XRFData: "As" is BDL.
XRFData: "As" is BDL.
XRFData: "As" is BDL.
XRF Data: "Mn" and "Pb" are BDL.
XRFData: "As" is BDL.
XRFData: "As" is BDL.
XRFData: "Pb" is BDL.
XRFData: "As" is BDL.
XRFData: "As" is BDL.
XRFData: "As" is BDL.
Void—Large hole poked in filter during unpacking.
Site op replaced wins impactor well.
1 Used avg flow rates from the site audits for 3 modules. EC/OC Data: Non-white carbon punch after carbon analysis,
indicative of mineral particles in deposit.
-------
BDL.
01/08/99
01/14/99
01/16/99
01/18/99
01/20/99
01/22/99
01/24/99
01/26/99
01/28/99
02/03/99
02/05/99
02/07/99
02/09/99
02/11/99
02/13/99
02/15/99
02/17/99
02/19/99
02/21/99
01/06/99
01/08/99
01/14/99
01/16/99
01/18/99
01/20/99
01/22/99
01/24/99
01/26/99
01/28/99
02/03/99
02/05/99
02/07/99
02/09/99
02/11/99
02/13/99
02/15/99
02/17/99
02/19/99
02/21/99
01/06/99
01/08/99
01/14/99
01/16/99
01/18/99
01/20/99
01/22/99
01/24/99
01/26/99
01/28/99
02/03/99
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Dal
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Used avg flow rates from the site audits for 3 modules. EC/OC Data: Non-white carbon punch after carbon analysis,
indicative of mineral particles in deposit.
Used avg flow rates from the site audits for 3 modules. EC/OC Data: Non-white carbon punch after carbon analysis,
indicative of mineral particles in deposit. Filter support grid upside down-deposit had widely spaced stripes.
Used avg flow rates from the site audits for 3 modules. EC/OC Data: Non-white carbon punch after carbon analysis,
indicative of mineral particles in deposit.
Used avg flow rates from the site audits for 3 modules. EC/OC Data: Non-white carbon punch after carbon analysis,
indicative of mineral particles in deposit.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is below detection limits (BDL).
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules. When Ch2 (Nylon) turned on it was very high (.80/11.5), couldn't see
anything wrong. Filter pack looked ok—adj to .42. Ch3 (Quartz) hose on filter is loose, slip ring would not tighten. XRF
Data: "As" is BDL. EC/OC Data: Non-white carbon punch after carbon analysis, indicative of mineral particles in deposit.
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules. XRF Data: "Pb" and "As" are BDL.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL. EC/OC Data: Non-white carbon punch after
carbon analysis, indicative of mineral particles in deposit.
Used avg flow rates from the site audits for 3 modules. Elapsed timer did not record time on ch2 (Nylon). XRF Data: "As" is
Used avg flow rates from the site audits for 3 modules. Large hole in Teflon filter (chl)--void.
Sample vol is low ch3-nylon (8.939 m3) and ch4-quartz (9.084 m3). XRF Data: "As" is below detection limits (BDL). EC/OC
Data: Non-white carbon punch after carbon analysis, indicative of mineral particles in deposit.
Sample vol is low ch3 (8.939 m3) and ch4 (9.084 m3). EC/OC Data: Non-white carbon punch after carbon analysis,
indicative of mineral particles in deposit.
MgO honeycomb denuder installed onNylon filter channel. XRF Data: "As" is BDL.
XRF Data: "As" is BDL. EC/OC Data: Non-white carbon punch after carbon analysis, indicative of mineral particles in deposit.
EC/OC Data: Non-white carbon punch after carbon analysis, indicative of mineral particles in deposit.
NylonNO3-Results are void;cannister assemblied with backing screen upstream of nylon filter. XRF Data: "As" is BDL.
Sample vol is low ch3 (9.040 m3). XRF Data: "Pb" is BDL, "As" is BDL.
EC/OC Data: Non-white carbon punch after carbon analysis, indicative of mineral particles in deposit.
Void—Elapsed sample time = 1 second.
Sample vol is low ch3 (8.969 m3).
Lower shield left down for this run to help determine why Met One PM2.5 cone are so high. Sample vol is low ch3 (9.017 m3).
XRF Data: "As" is BDL.
Lower shield left down for this run to help determine why Met One PM2.5 cone are so high. Sample vol is low ch3
(8.994m3). XRF Data: "As" is BDL.
Lower shield left down for this run to help determine why Met One PM2.5 cone are so high. Sample vol is low ch3
(9.122m3). XRF Data: "As" is BDL.
Lower shield raised back up. XRF Data: "As" is BDL.
EC/OC Data: Non-white carbon punch after carbon analysis, indicative of mineral particles in deposit.
XRF Data: "As" is BDL.
Sample vol is low ch3 (9.023 m3). XRF Data: "As" is BDL. EC/OC Data: Non-white carbon punch after carbon analysis,
indicative of mineral particles in deposit.
XRF Data: "As" is BDL.
Teflon (Chi) Data is void due to very high mass cone (-600 ug/m3). XRF Data: "As" is BDL. Teflon (ch2) filter was not
transferred to CEMQA for 1C—appears to have been a mistake when unloading cannisters.
1 URG400: VOID-Sample ran~14hrs,Flags FTP, TM, F10. Mass cone andparticulateNO3- data is void.
1 URG400: VOID—Sample ran~10hrs, URG sent modification kit for pump. Mass cone andNO3- data is void.
1 URG400: Flag FTP
1 URG400: Flag FTP
1 URG400: Flag FTP,cleaned wins impactors
1 URG400: Flag FTP
1 URG400: Void-Nylon and Teflon filterpacks were cross- threaded. Mass cone and particulate NO3- data is void.
1 URG400: Flag FTP
1 URG400: Flag FTP
1 URG400: Flag FTP
1 URG400: Flag FTP, Double pump with DGM installed on 1/29. Bob M calibrated on 1/29. URG450: Bob M calibrated on
-------
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
02/05/99 URG
02/07/99 URG
02/09/99 URG
02/11/99 URG
02/13/99 URG
02/15/99 URG
02/17/99 URG
02/19/99 URG
02/21/99 URG
01/06/99 VAPS
01/08/99 VAPS
01/14/99 VAPS
01/16/99 VAPS
01/18/99 VAPS
01/20/99 VAPS
01/22/99 VAPS
01/24/99 VAPS
01/26/99 VAPS
01/28/99 VAPS
02/03/99 VAPS
02/05/99 VAPS
02/07/99 VAPS
02/09/99 VAPS
02/11/99 VAPS
02/13/99 VAPS
02/15/99 VAPS
02/17/99 VAPS
02/19/99 VAPS
02/21/99 VAPS
01/17/99 AND 1
01/19/99 AND 1
01/21/99 AND 1
01/23/99 AND 1
01/25/99 AND 1
01/27/99 AND 1
01/29/99 AND 1
01/31/99 AND 1
02/02/99 AND 1
02/04/99 AND 1
02/08/99 AND 1
02/10/99 AND 1
02/12/99 AND 1
02/14/99 AND 1
02/16/99 AND 1
02/18/99 AND 1
02/20/99 AND 1
02/22/99 AND 1
02/24/99 AND 1
02/26/99 AND 1
01/17/99 FRM-T 1
01/19/99 FRM-T 1
01/21/99 FRM-T
01/23/99 FRM-T
01/25/99 FRM-T
01/27/99 FRM-T
01/29/99 FRM-T
01/31/99 FRM-T
02/02/99 FRM-T
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
^
<;
I
-\
(
02/04/99 FRM-T 1
XRFData: "As" isBDL.
URG400: Flag FTP
URG400: VOID—Flag FTP. Filter pack assemblied incorrectly- Teflon in bottom position and Nylon in top position. Mass
cone and particulate NO3- data is void.
URG400: VOID-Sampler ran 16 hrs 41 mm, Flag FTP, Tm, Flo. Mass cone and particulate NO3- data is void.
XRFData: "As" isBDL.
URG400: Void (Teflon filter)—appears filter weights were recorded incorrectly.
URG400: Flag FTP. XRF Data: "As" is BDL.
URG400: Flag FTP. XRF Data: "As" is BDL.
URG400: Flag FTP. XRF Data: "As" is BDL.
URG400: Flag FTP
High sample volume on all channels. (Flow rates appear to be off.) Teflon (ch2) filter has started to tear @ ring.
Void—confusion w/ sample id#s. Ran Sample Day 4 filters on day 3
Void—confusion w/ sample id#s. Sent additional filters renamed filters Day 4 (too)
Sample vol high ch2-teflon (22.971 m3).
Void—Sample ran 36.4 hrs.
Void—Teflon (ch2-fine) filter weights appear to have been
recorded incorrectly.
XRFData: "As" is BDL.
XRF Data: "Pb" and "As" are BDL.
Void—Ch2 filterpack loaded incorrectly. Teflon loaded downstream of Nylon. Mass cone and particulate NO3- data is void.
XRFData: "As" is BDL.
Void—Teflon (chl-coarse)large hole in filter.
Void—Ch2 filterpack loaded incorrectly. Nylon loaded upstream of Teflon. Mass cone and particulate NO3- data is void.
Void—Teflon (ch2-fine) filter had a pin hole. Chi (PM10)— filter was not post-weighed-error occurred when filterpack
was unloaded.
Ch2:pinhole near ring.Hole seems to have occurred sub- sequent to sampling b/c mass cone looks ok.
XRF Data (from Teflon filter-ch2) "As" was BDL.
Sampler displays low sample vol ch3-Teflon (22.464 m3). XRF Data (from Teflon filter-ch2) "As" was BDL.
Filters (chl and ch3) were loaded incorrectly. Nylon (chl) was sampled in ch3. Sample volumes have been adjusted.
XRF Data (from Teflon filter-ch2) "Si" and "Mn" and "As" was BDL.
Sampler displays low sample vol low chl-Nylon (9.792 m3) and in ch3 (22.608 m3).
Sampler displays low sample vol chl (9.792 m3)
Sampler displays low sample vol chl (9.792 m3).
Sampler displays low sample vol ch3 (22.608 m3).
Ch2(Teflon):filter appears to have scratched with finger- nail or forceps. Did not appear to effect mass cone.
Sampler displays low sample vol chl (9.792 m3) and ch3
(22.608m3).XRFData: "As" is BDL.
XRF Data: "As" and "Cu" are BDL.
Sampler displays low sample vol chl (9.792 m3).
Ch4(Quartz):Void-site operator touched filter. Sampler displays low sample vol ch3 (22.608 m3).
Sampler displays low sample vol chl (9.792 m3) and ch3 (22.608 m3).
Sampler displays low sample vol ch3 (22.608 m3). XRF Data: "As" is BDL.
XRFData: "As" is BDL.
Sampler displays low sample vol chl (9.792 m3) and ch3 (21.744 m3).
XRFData: "As" was BDL
Void-H2O found on side of filter cassette and sides of impactor.When filter was unloaded there was moisture present.
Wins impactor cleaned and re-oiled before sample run.
Heavy rains. H2O droplets were observed on the sides of filter cassette. Filter observed to be ok. H2O seeped through the
intersection of the downtube to the enclosure. Sampler parts were cleaned and prepared for next run.
XRFData: "As" is BDL.
-------
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
02/08/99
02/10/99
02/12/99
02/14/99
02/16/99
02/18/99
02/20/99
02/22/99
02/24/99
02/26/99
01/17/99
01/19/99
01/21/99
01/23/99
01/25/99
01/27/99
01/29/99
01/31/99
02/02/99
02/04/99
02/08/99
02/10/99
02/12/99
02/14/99
02/16/99
02/18/99
02/20/99
02/22/99
02/24/99
02/26/99
01/17/99
01/19/99
01/21/99
01/23/99
01/25/99
01/27/99
01/29/99
01/31/99
02/02/99
02/04/99
02/08/99
02/10/99
02/12/99
02/14/99
02/16/99
02/18/99
02/20/99
02/22/99
02/24/99
02/26/99
01/17/99
01/19/99
01/21/99
01/23/99
01/25/99
01/27/99
01/29/99
01/31/99
02/02/99
02/04/99
02/08/99
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
XRFData: "As" isBDL.
XRFData: "As" isBDL.
XRFData: "As" isBDL.
XRFData: "As" isBDL.
XRFData: "As" isBDL.
Bob M had to change filter cassette, (filter had been loaded into the wrong filter cassette.)
1 Sample did not run due to short in electrical system .
1 Used avg flow rates from the site audits for 3 modules. Nylon/Quartz modules:Adj Mag Reading from 0.44" H2O
to 0.46" H2O after 10.4 hrs into run. XRF Data: "As" was BDL.
1 Used avg flow rates from the site audits for 3 modules. All channels flow was set to 0.52 b/c gauge was misread.
1 Used avg flow rates from the site audits for 3 modules. Nylon channel vacuum gauge is faulty. XRF Data: "Pb" was BDL.
1 Used avg flow rates from the site audits for 3 modules. Teflon filter has a hole which seems to have occurred
subsequent to sampling b/c mass cone is ok.
1 Used avg flow rates from the site audits for 3 modules. EC/OC Data: Non-white carbon punch after carbon analysis,
indicative of mineral particles in deposit.
1 Used avg flow rates from the site audits for 3 modules.
1 Used avg flow rates from the site audits for 3 modules.
1 Used avg flow rates from the site audits for 3 modules. EC/OC Data: Filter support grid upside down-deposit had
widely spaced stripes or grid pattern.
1 Used avg flow rates from the site audits for 3 modules.
1 Used avg flow rates from the site audits for 3 modules.
1 Used avg flow rates from the site audits for 3 modules.
1 Used avg flow rates from the site audits for 3 modules. Suspect Data Teflon channel-sampler appears to have
run ok-no reason to invalidate. XRF Data: "Si", "Pb", and "As" are BDL.
1 Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL.
1 Used avg flow rates from the site audits for 3 modules. EC/OC Data: Non-white carbon punch after carbon analysis,
indicative of mineral particles in deposit.
1 Used avg flow rates from the site audits for 3 modules.
1 Used avg flow rates from the site audits for 3 modules.
1 Used avg flow rates from the site audits for 3 modules.
1 Used avg flow rates from the site audits for 3 modules.
1 Used avg flow rates from the site audits for 3 modules.
1 No sample vol recorded for any channels.
1 Low sample vol ch3-nylon (8.933 m3). XRF Data: "As" was Below detection limits (BDL).
1 Quartz (Ch5): Mask shin was not installed. Analysis results corrected numerically to adjust for the actual deposition area. Low
sample vol ch3 (8.948 m3).
1 Low sample vol ch3 (8. 970m3). XRF Data: "Si" was BDL. "Cu" was BDL. "As" was BDL.
1 Low sample vol ch3 (9.042m3).
1 Low sample vol ch3 (8.908 m3).
1 Low sample vol ch3 (8.913 m3).
1 Low sample vol ch3 (8.882 m3). Backing screen found in petri dish of Teflon (chl)
1 Low sample vol ch3 (8.921 m3).
1 Low sample vol ch3 (8.935 m3).
1 Low sample vol ch3 (9.095m3). XRF Data: "Cu" and "As" are BDL.
-------
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
02/10/99
02/12/99
02/14/99
02/16/99
02/18/99
02/20/99
02/22/99
02/24/99
02/26/99
01/17/99
01/19/99
01/21/99
01/23/99
01/25/99
01/27/99
01/29/99
01/31/99
02/02/99
02/04/99
02/08/99
02/10/99
02/12/99
02/14/99
02/16/99
02/18/99
02/20/99
02/22/99
02/24/99
02/26/99
01/17/99
01/19/99
01/21/99
01/23/99
01/25/99
01/27/99
01/29/99
01/31/99
02/02/99
02/04/99
02/08/99
02/10/99
02/12/99
02/14/99
02/16/99
02/18/99
02/20/99
02/22/99
02/24/99
02/26/99
01/15/99
01/17/99
01/19/99
01/21/99
01/23/99
MET
MET
MET
MET
MET
MET
MET
MET
MET
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
AND
AND
AND
AND
AND
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Low sample vol ch3 (9.033 m3).
Lowsamplevolch3(8.919m3).XRFData: "As" isBDL.
Low sample vol ch3 (8.949 m3). XRP Data: "Mn" and "As" are BDL.
Low sample vol ch3 (8.930 m3). EC/OC Data: Non-white carbon punch after carbon analysis, indicative of mineral particles in
deposit.
XRFData: "Cu" is BDL.
Low sample vol ch3 (8.819m3). EC/OC Data: Void due to suspect data. XRP Data: "As" is BDL.
Small hole near ring-did not appear to effect mass cone. Hole seems to have occurred subsequent to sampling. Low
sample vol ch3 (8.878 m3). XRP Data: "Cu" and "As" are BDL.
Void—Sample ran 15 hrs 4 min.
Low sample vol ch3 (8.919 m3). XRF Data: "Mn" and "As" are BDL.
Quartz filter holder observed to be too loose—sample is invalid. Particulate NO3- data is void.
XRP Data: "As" was below detection limits (BDL).
XRFData: "As" was BDL.
Teflon filter has large tear—did not appear to effect mass cone. Tear seems to have occurred subsequent to sampling. XRP Data:
"As" was BDL.
Teflon filter has a hole-did not appear to effect mass cone. Hole seems to have occurred subsequent to
sampling. XRP Data: "Mn" and "As" are BDL.
XRF Data: "As" is BDL.
1 XRF Data: "As" is BDL.
1 Teflon filter has a pin hole-did not appear to effect mass cone. Hole seems to have occurred subsequent to
sampling. XRF Data: "As" is BDL.
1 XRF Data: "Si", "Cu" and "As" are BDL.
1 XRF Data: "Mn" and "As" are BDL.
1 XRF Data: "As" is BDL.
1 XRF Data: "As" is BDL.
Low sample vol ch2 (20.128 m3).XRF Data: "As" is Below Detection Limits (BDL).
Low sample vol ch2 (19.853 m3).
Low sample vol ch2 (19.814m3). XRF Data: "As" is BDL.
Ch2 (Teflon) is void-appears filter weights were recorded incorrectly.
Void-Sample Duration = 33.2 hours. XAD denuder was found to contain a very small piece of glass while unpacking the
cooler. A small crack on one of the ends of the denuder is suspected.
XRFData: "Cu" is BDL.
Ch3 (quartz) has high sample volume (29.367 m3). XRF Data: "Mn" and "As" were BDL.
Void-Ch2 (Teflon-fine) filter has several pin holes. Appears pin holes were present before sampling.
Ch2 (Teflon-fine) filter has 1 small hole present which seems to have occurred subsequent to sampling b/c mass
cone looks ok. XRF Data: "Mn", "Cu", and "As" are BDL.
Nylon filter was not transferred to CEMQA for analyses— believe an error occurred when unloading filter packs.
Particulate NO3- data is void.
XRFData: "Mn" is BDL.
Void-Ch2 filter has 1 pin hole. Appears pin hole was present before sampling. Low sample vol ch3-quartz
(20.237m3).Chi (PM10): filter weights were recorded incorrectly.
Void-Negative mass (PM2.5). Believe that lap operator wrote down the initial weights incorrectly.
XRF Data: "As" is BDL.
XRF Data: "Mn" is below detection limits (BDL); "As" is BDL.
XRF Data: "Mn" is BDL; "Cu" is BDL.
XRF Data: "As" is BDL.
Sampler displays low sample vol chl-Nylon (9.648 m3).
Site op noted a problem with the o-rings. Sampler displays low sample vol chl (9.792 m3). XRF Data: "Cu" is BDL;
"As" is BDL.
01/25/99 AND
10
-------
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
01/27/99
01/29/99
01/31/99
02/02/99
02/06/99
02/08/99
02/10/99
02/12/99
02/14/99
02/16/99
02/18/99
02/20/99
02/22/99
02/24/99
01/15/99
01/17/99
01/19/99
01/21/99
01/23/99
01/25/99
01/27/99
01/29/99
01/31/99
02/02/99
02/06/99
02/08/99
02/10/99
02/12/99
02/14/99
02/16/99
02/18/99
02/20/99
02/22/99
02/24/99
01/15/99
01/17/99
01/19/99
01/21/99
01/23/99
01/25/99
01/27/99
01/29/99
01/31/99
02/02/99
02/06/99
02/08/99
02/10/99
02/12/99
02/14/99
02/16/99
02/18/99
02/20/99
02/22/99
02/24/99
01/15/99
01/17/99
01/19/99
01/21/99
01/23/99
01/25/99
01/27/99
AND
AND
AND
AND
AND
AND
AND
AND
AND
AND
AND
AND
AND
AND
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-T
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
FRM-Q
IMP
IMP
IMP
IMP
IMP
IMP
IMP
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Suspect data(Teflon fliter-ch2)-fuzz and other unidentified particles present on the filter-affected the mass cone.
XRFData: "Mn" isBDL.
Sampler displays low sample vol chl (9.648 m3). XRF Data: "Mn" and "As" are BDL.
XRFData: "As" is BDL.
XRF Data: "As" is BDL.
Sampler displays low sample vol chl (9.792 m3).
XRFData: "As" is BDL.
XRFData: "As" is BDL.
Fuzz present on filter(Teflon filter-ch2)-did not appear to affect mass cone. Sampler displays low sample vol chl
(9.792 m3). XRF Data: "As" is BDL.
XRFData: "As" is BDL.
Average flow rates were recorded 2/25/99 @ 1200.
XRF Data: "As" is below detection limits (BDL).
XRFData: "As" is BDL.
XRFData: "As" is BDL.
2 pin holes in Teflon filter-did not affect mass cone. Appears hole occurred subsequent to sampling. XRF
Data: "Cu" is BDL; "Pb" is BDL; "As" is BDL.
No sample-did not run due to malfunctioning filter exchange mechanism.
XRFData: "Mn" is BDL.
No sample-did not run due to malfunctioning filter exchange mechanism.
XRFData: "As" is BDL.
XRF Data: "Cu" and "As" are BDL.
Sampler's Avg Temperature readings are incorrect. XRF Data: "Cu" and "As" are BDL.
XRFData: "As" is BDL.
Fuzz present on Teflon filter-did not appear to affect mass cone. XRF Data: "As" is BDL.
Small tear @ ring-did not affect mass cone. Appears hole occurred subsequent to sampling.
EC/OC Data: Quartz filter was not transferred to DRI.
Sampler's Avg Temperature readings are incorrect.
XRF Data:"Mn" is below the detection limits(BDL)."As" is BDL Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules.
XRF Data" "As" is BDL. Used avg flow rates from the site audits for 3 modules
Used avg flow rates from the site audits for 3 modules.
Slight tear in Teflon filter-did not affect mass cone. Appears tear occurred after sampling. Ch2 is void-moisture
found inside this channel. Filter and housing saturated withH2O. Valve would not turn on. Used avg flow rates from the site
audits for 3 modules. XRF Data: "Mn" is BDL; "Cu" is BDL; "As" is BDL.
Void-Water found inside Ch2. Used avg flow rates from the site audits for 3 modules.
No sample-filters were not installed in cartridges.
11
-------
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
01/29/99
01/31/99
02/02/99
02/06/99
02/08/99
02/10/99
02/12/99
02/14/99
02/16/99
02/18/99
02/20/99
02/22/99
02/24/99
01/15/99
01/17/99
01/19/99
01/21/99
01/23/99
01/25/99
01/27/99
01/29/99
01/31/99
02/02/99
02/06/99
02/08/99
02/10/99
02/12/99
02/14/99
02/16/99
02/18/99
02/20/99
02/22/99
02/24/99
01/15/99
01/17/99
01/19/99
01/21/99
01/23/99
01/25/99
01/27/99
01/29/99
01/31/99
02/02/99
02/06/99
02/08/99
02/10/99
02/12/99
02/14/99
02/16/99
02/18/99
02/20/99
02/22/99
02/24/99
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
IMP
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
MET
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
URG
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Used avg flow rates from the site audits for 3 modules.
Used avg flow rates from the site audits for 3 modules. XRF Data: "Mn" is BDL. EC/OC Data: Quartz filter was not
transferred to DRI.
Used avg flow rates from the site audits for 3 modules. XRF Data: "Mn" is BDL.
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL.
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL.
Used avg flow rates from the site audits for 3 modules.
Ch2 timer did not record any elapsed time. Though it appears to have run as programmed. Pressure drop was approx double
all previous runs. Used avg flow rates from the site audits for 3 modules. XRF Data: "Pb" is BDL.
Used avg flow rates from the site audits for 3 modules. XRF Data: "Cu" is BDL.
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL.
Used avg flow rates from the site audits for 3 modules. EC/OC Data: Filter support grid upside down-deposit has
widely spaced stripes or grid pattern. XRF Data: "As" is BDL.
Used avg flow rates from the site audits for 3 modules. EC/OC Data: Filter support grid upside down-deposit has
widely spaced stripes or grid pattern.
Used avg flow rates from the site audits for 3 modules. XRF Data: "As" is BDL.
Medium-sized tear in Teflon filer-did not affect mass cone. Appears tear occurred after sampling. Ch2 timer displays
0.0 hrs for elapsed time. Site op noted that it appears to have run as programmed. Used avg flow rates from the site audits for
3 modules. XRF Data: "As" is BDL.
Void-due to sample duration (sample began @ 1400.)
Sampler displaying incorrect sample volume. (Sampler reading 1/2 the acutal volume.) Site op verified volume. In the
database, the mass cone has been calculated using-9.0 m3. XRF Data: "Cu" is below detection limits (BDL).
Sampler displaying incorrect sample volume. (Sampler reading 1/2 the acutal volume.) Site op verified volume. In the
database, the mass cone has been calculated using -9.0 m3 XRF Data: "Pb" is BDL; "As" is BDL.
Sample volume was calculated using the flow measured by site op. Nylon filter was crimped-possible leak or
improperly put together. XRF Data: "As" is BDL.
New pump installed on 1/22/99. Low sample vol ch3-nylon (8.586 m3). XRF Data: "Si" is BDL; "Mn" is BDL; "Cu" is BDL;
"As" is BDL.
Flow check performed by site operator. Low sample vol chl- Teflon (9.092m3) andch3 (8.568 m3).
Low sample vol ch3 (8.507 m3).
Low sample vol ch3 (8.544 m3). XRF Data: "Pb" is BDL.
Low sample vol ch3 (9.082 m3).
1 Low sample vol ch3 (8.468 m3). XRF Data: "Mn", "Cu", "Pb", and "As" are BDL.
Cooling fan was turned off to try to identify high PM cone as compared to the FRM. Was to determine if aspiration of PM2.5
is affected by cooling fan. Low sample vol ch3 (8.651 m3). XRF Data: "Cu" is BDL.
Low sample vol ch3 (8.765 m3). XRF Data: "Pb" and "As" are BDL.
Low sample vol ch3 (8.607 m3). XRF Data: "Pb" and "As" are BDL.
Low sample vol ch3 (8.550 m3). XRF Data: "Si" and "As" are BDL.
Low sample vol ch3 (8.531 m3). XRF Data: "Cu" and "As" are BDL.
Void-sample ran for 1 second. (However, the volumes reported by the sampler look as if the sampler ran as programmed.)
Low sample vol ch3 (8.492 m3). XRF Data: "As" is BDL.
Low sample vol ch3 (8.456 m3). XRF Data: "As" is BDL.
Low sample vol ch3 (8.479 m3). XRF Data: "As" is BDL.
XRF Data: "Mn" is below detection limits (BDL).
Both URG400 and URG450 are displaying temps ~4°C higher than the other samplers—did not affect the data.
XRF Data: "Pb" is BDL; "As" is BDL.
XRF Data: "As" is BDL.
XRF Data: "Mn" is BDL; "As" is BDL.
Max meter drop reading = 199.4 mm Hg;No average flow or temp recorded. Did not affect mass cone.
XRF Data: "Mn" and "As" are BDL.
XRF Data: "Mn", "Cu", and "As" are BDL.
XRF Data: "As" is BDL.
XRF Data: "As" is BDL.
URG400: Void—does not appear that sampler ran as programmed.
XRF Data: "Cu" and "As" are BDL.
Wins impactor cleaned before this sample run. XRF Data: "As" is BDL.
XRF Data: "As" is BDL.
XRF Data: "As" is BDL.
01/15/99 VAPS 1 XRFData: "Mn" is below detection limits (BD1); "As" isBDL.
12
-------
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
01/17/99
01/19/99
01/21/99
01/23/99
01/25/99
01/27/99
01/29/99
01/31/99
02/02/99
02/06/99
02/08/99
02/10/99
02/12/99
02/14/99
02/16/99
02/18/99
02/20/99
02/22/99
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
VAPS
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
02/24/99 VAPS
Low sample volume ch2 (20.512m3) and ch3 (19.890m3); XRFData: "Mn" isBDL.
XRFData: "As" isBDL.
Low sample volume ch3 (19.698 m3).
Low sample volume ch3 (19.430 m3).XRF Data: "Mn" is BDL; "Pb" is BDL; "As" isBDL.
Low sample volume ch2 (20.531 m3).XRF Data: "Mn" and "Cu" and "Pb" and "As" are BDL.
Void—Sample duration = 34.7hrs.
Low sample volume ch2 (19.989 m3) and ch3 (19.964 m3). EC/OC Data: Non-white carbon punch, after carbon analysis,
indicative of mineral particles in deposit.
Low sample volume ch2 (19.879m3) and ch3 (20.514m3). XRF Data: "Pb" is BDL.
Low sample volume ch2 (19.744m3) and ch3 (19.240m3). XRFData: "Mn" and "As" are BDL.
Low sample volume ch2 (20.021 m3) and ch3 (19.188 m3). XRFData: "As" is BDL.
No sample—did not run due to programming error.
Void-Teflon and Nylon filters. Filterpack was assemblied incorrectly (Teflon loaded downstream of Nylon filter.)
Low sample volume ch3 (18.991 m3). Mass cone and particulate NO3- data are void.
Low sample vol ch2 (20.208 m3) and ch3 (18.073 m3). XRF Data: "Cu" and "As" are BDL.
Void-Teflon and Nylon filters. Filterpack was assembled incorrectly (Teflon loaded downstream of Nylon filter.)
Low sample volume ch3 (20.108 m3). Mass cone and particulate NO3- data are void.
Low sample volume ch2 (19.717m3) and ch3 (19.389 m3). XRF Data: "Cu" is BDL.
Void-Sample duration = 38.4 hrs
Low sample volume ch2 (20.138 m3) and ch3 (20.199 m3).
Two small holes-did not affect mass cone. Holes occurred after sampling. Low sample vol ch2 (20.103 m3).
XRFData: "As" is BDL.
EC/OC Data: Quartz filter was not transferred to DRI. XRFData: "Mn" and "As" are BDL.
13
-------
Appendix D
Final Data Archive, By Species, Site, Date,
and Sampler for the 4-City Study
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of PM-2.5 (jig/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler1
AND
32.7
6.7
26.8
85.1
13.3
30.1
2.7
22.0
5.4
18.0
2 2
34.2
59.0
31.2
17.4
22.6
26.4
21.3
21.9
25.6
24.6
12.9
9.1
20.6
8.4
4.4
17.1
11.0
22 3
13.6
4.2
12.3
12.6
20.2
13.5
12.9
21.4
9.0
25.3
9.4
19.3
38.5
15.0
11.8
13.9
25.8
16.9
11.8
15.5
5.1
32.5
24.9
11.0
5.3
10.8
14.8
6.4
9.4
4.2
23.1
6.2
8.5
15.7
18.3
10.7
7.0
9.4
6.1
11.0
12.4
5.7
19.7
9.5
13.9
5.0
11.9
FRM
35.3
7.7
28.6
74.3
14.9
29.4
3.0
29.5
5.6
21.0
2 2
35.6
59.0
29.1
17.6
23.6
26.8
21.2
22.0
25.3
24.4
12.7
9.0
19.8
7.6
4.1
17.3
11.5
23.3
13.8
3.9
12.4
12.8
20.3
13.4
20.6
9.2
25.8
19.4
37.6
15.0
12.4
13.8
25.8
16.5
11.9
15.6
5.2
32.2
24.9
10.5
5.0
10.8
14.4
8.6
9.7
4.4
23.8
6.5
8.3
15.3
8.7
10.4
6.6
11.5
12.8
6.5
20.5
9.8
13.9
5.2
12.0
IMP
34.7
7.9
26.3
87.5
13.2
27.0
2.8
16.4
5.7
19.7
2.6
34.2
58.7
32.8
16.0
22.6
22.3
18.3
15.7
21.2
22.6
10.7
8.9
14.7
6.9
4.1
15.5
10.1
18.2
12.0
3.9
13.2
12.2
17.4
12.4
9.0
26.7
9.1
18.8
38.6
14.7
11.7
13.6
25.7
16.6
11.4
5.4
30.3
25.1
10.5
5.7
10.8
14.4
6.2
9.5
5.0
23.0
5.7
8.5
18.6
7.2
7.0
9.1
6.2
11.2
12.6
6.2
20.4
9.9
14.2
5.0
11.4
MET
71.26
21.6
66.2
93.0
13.3
33.4
3.0
33.4
14.2
19.3
9.5
41.6
65.7
38.9
32.0
25.2
91.5
45.0
38.0
46.8
43.7
26.7
20.6
38.0
9.8
34.2
12.1
28.7
21.1
6.4
25.2
26.0
36.2
28.9
26.7
11.5
25.0
7.9
20.4
40.7
15.5
10.9
13.3
25.1
16.4
12.8
15.6
5.5
38.6
24.4
12.3
6.5
15.9
10.0
5.0
24.4
6.0
9.6
15.9
18.8
8.3
7.1
11.0
6.8
12.9
13.6
6.7
23.7
15.4
5.6
12.3
URG
34.9
7.7
32.9
105.1
16.8
38.7
3.1
5.1
2.5
53.5
80.5
39.8
22.3
35.1
19.0
24.5
23.5
11.6
19.5
8.1
4.0
16.1
12.7
3.4
11.7
20.8
13.5
11.3
21.0
9.4
25.0
8.7
21.2
40.9
15.4
12.3
14.0
26.4
16.5
11.7
15.3
5.2
31.6
24.4
11.3
5.9
11.7
15.0
5.9
9.4
4.2
24.6
6.2
8.4
16.1
18.8
9.1
7.8
9.9
7.2
12.0
7.2
21.3
11.0
15.2
5.5
13.1
VAPS
36.9
32.5
57.5
4.9
49.7
12.6
26.2
9.1
74.8
54.2
36.4
63.4
29.8
33.4
26.3
9.1
21.7
10.2
4.1
17.5
32.3
13.6
3.8
12.2
21.5
24.4
9.8
33.2
47.4
16.6
13.5
16.2
34.6
18.6
14.9
3.5
41.5
25.4
13.2
17.0
10.7
4.6
25.4
8.2
9.4
20.9
10.5
8.4
12.5
16.3
24.5
16.1
5.4
15.1
1 Shaded values are from replicate sampler used to increase n for mass statistics at Rubidoux.
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of SO4-T (jig/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
0.6
0.2
0.5
6.1
1.6
1.8
0.2
0.5
0.3
0.7
0.3
1.7
3.8
2.2
0.5
2.1
0.6
0.9
0.7
0.9
1.7
1.1
0.2
0.6
0.3
0.4
0.6
0.4
0.6
1.0
0.2
0.6
1.0
1.2
1.0
1.1
3.0
1.5
4.0
1.3
6.7
8.2
4.0
1.7
2.9
5.4
4.1
2.4
4.2
1.2
6.3
7.1
3.6
1.1
1.7
2.6
1.9
1.3
0.7
5.3
1.8
3.4
5.5
6.1
3.1
2.8
1.8
2.5
2.3
5.6
2.5
2.0
2.8
6.2
1.9
2.8
FRM
IMP
0.6
0.2
0.5
5.4
1.6
1.7
0.2
0.5
0.3
0.7
0.3
1.6
3.6
2.0
0.4
1.9
0.6
0.89
0.7
0.91
1.64
1.06
0.18
0.54
0.29
0.39
0.59
0.38
0.63
1.1
0.22
0.62
1.02
1.15
1.05
1.15
1.6
4.2
2.8
6.7
8.1
4.0
1.7
3.0
5.4
4.1
2.5
4.2
1.2
6.1
6.9
3.6
1.1
1.7
2.5
1.9
1.3
0.7
5.3
6.2
3.3
2.7
1.8
2.6
2.4
5.3
2.4
1.9
2.7
6.1
1.6
2.8
MET
1.0
0.5
1.0
6.0
1.7
1.8
0.3
0.7
0.4
0.8
0.4
1.7
3.6
2.1
0.8
1.9
0.9
1.2
1.0
1.2
1.9
1.3
0.4
0.8
0.4
0.8
0.5
0.7
1.3
0.3
0.9
1.2
1.5
1.3
1.7
4.3
2.5
6.8
8.4
4.0
1.8
3.0
5.4
4.2
2.5
4.4
1.3
6.5
7.0
3.6
1.2
2.5
1.3
0.7
5.1
1.8
3.5
5.7
6.2
3.3
2.8
1.8
2.5
2.4
5.6
2.5
1.5
6.1
1.9
2.8
URG
VAPS
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of SO4Q (jig/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
FRM
0.7
0.3
0.6
6.0
1.6
1.9
0.6
0.3
1.2
0.4
1.8
3.7
1.6
0.4
2.0
0.7
1.0
0.8
1.0
1.8
1.2
0.2
0.6
0.3
0.5
0.7
0.4
0.7
1.2
0.3
0.7
1.1
1.6
1.2
1.2
3.1
1.8
4.0
2.6
6.7
8.4
4.4
2.0
3.0
5.7
4.0
2.7
4.1
1.5
6.5
7.3
3.7
1.5
2.0
2.7
1.9
1.3
0.7
5.2
1.7
3.5
5.4
6.2
3.3
2.8
2.0
2.6
2.3
5.5
2.6
2.2
2.8
5.9
2.0
2.9
IMP
MET
URG
0.7
0.4
0.5
6.6
1.7
2.0
0.3
0.7
0.4
0.7
0.4
1.9
3.8
1.9
0.5
2.0
1.0
0.7
1.0
1.8
1.1
0.2
0.6
0.4
0.4
0.7
0.4
0.7
1.3
0.3
0.7
1.1
1.2
1.2
1.3
1.6
4.1
2 3
6.8
8.4
4.2
1.8
2.8
5.9
3.9
2.6
4.0
1.4
6.6
6.8
3.9
1.4
2.0
2.5
2.0
1.4
0.7
5.0
1.8
3.7
5.6
6.4
3 2
2.7
1.9
2.5
2.4
5.5
2.7
2.4
2.9
6.1
2 3
2.6
VAPS
0.7
0.5
0.9
6.7
1.9
2.0
0.3
0.7
0.4
0.8
0.4
1.8
3.7
2.3
0.7
2.1
0.6
1.0
1.7
0.2
0.6
0.3
0.4
0.7
0.4
0.7
1.1
0.3
0.7
1.0
1.1
1.1
1.2
2.6
1.3
3.6
2.3
7.5
3.5
1.4
2.6
4.9
3.4
2.3
3.5
1.2
5.8
6.5
3.3
1.2
1.7
2.6
1.8
1.3
0.7
4.9
1.8
3.3
6.6
3.4
2.8
1.9
2.3
5.4
2.1
2.1
6.2
1.7
3.0
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of NO3T (jig/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND FRM IMP MET URG VAPS
10.5
0.8
8.3
38.0
3.1
11.6
0.2
7.7
0.2
7.4
0.1
15.3
26.2
12.3
5.6
9.1
4.3
3.8
2.8
3.8
1.7
1.6
0.6
0.6
0.9
0.1
2.5
1.2
4.5
1.2
0.1
0.2
0.9
3.7
1.7
0.6
3.7
1.4
5.4
0.6
4.3
8.2
2.8
2.9
2.6
6.2
3.2
1.8
1.9
0.6
5.9
3.6
1.1
0.5
3.0
5.4
0.1
0.2
0.1
0.4
0.3
0.1
0.3
0.4
0.9
0.3
0.1
0.1
0.2
0.1
0.2
0.6
0.4
0.5
0.3
1.7
10.8
0.9
9.5
38.8
3.1
11.6
0.2
8.9
0.4
7.4
0.1
15.7
26.1
13.2
6.0
8.9
5.3
4.6
3.1
4.7
2.8
2.7
0.4
1.1
1.3
3.4
1.4
5.4
1.8
0.1
0.3
1.4
4.8
3 2
1.2
5.3
1.2
4.3
8.1
2.7
3.0
2.5
5.6
3.1
1.6
1.7
0.6
5.9
3.3
1.2
0.5
4.6
0.2
0.1
0.4
0.3
0.1
0.4
0.5
0.8
0.2
0.2
0.1
0.2
0.2
0.3
0.7
0.5
0.3
1.7
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of NO3P (jig/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
10.24
0.88
8.28
34.65
3.22
10.92
0.61
7.98
0.39
6.95
0.10
14.24
24.14
14.04
5.14
9.14
5.55
4.73
4.18
5.06
3.50
3.61
0.40
1.29
1.93
0.41
4.08
2.14
7.02
4.18
0.16
0.28
2.36
6.44
5.39
2.51
3.78
1.71
5.31
3.99
4.32
7.95
2.90
2.80
2.64
5.90
3.03
1.97
2.60
0.65
6.32
3.50
1.36
0.53
2.85
5.13
0.23
0.46
0.14
1.10
0.44
0.27
1.33
1.07
1.54
0.66
0.44
0.16
0.57
0.35
0.52
1.60
0.76
1.67
0.53
2.30
FRM
11.91
1.05
9.64
38.03
3.06
12.13
9.69
0.44
9.52
0.13
16.71
25.83
12.22
6.20
9.70
5.81
4.98
3.46
5.33
2.55
3.71
0.45
1.37
1.99
0.44
4.42
2.23
7.43
4.24
0.20
0.29
2.36
6.81
5.47
2.42
4.11
1.90
5.46
1.63
4.62
8.64
3.39
3.46
2.71
6.53
3.14
2.18
2.62
0.92
6.14
3.86
1.59
1.07
3.84
6.09
0.22
0.43
0.13
1.00
0.25
0.23
1.01
1.07
1.47
0.67
0.25
0.17
0.29
0.11
0.34
1.75
0.73
1.14
0.40
2.25
IMP
10.68
0.90
7.76
34.58
3.51
11.62
0.65
8.37
0.41
7.88
0.08
15.39
25.85
14.83
6.03
9.68
4.36
4.03
2.65
4.42
2.74
2.90
0.41
1.25
1.91
0.37
3.52
2.10
6.28
3.51
0.14
0.25
1.83
4.43
3.51
1.99
1.93
5.91
1.91
4.78
8.30
3.12
3.06
2.79
6.30
3.33
2.07
2.68
0.68
5.51
3.74
1.42
0.54
3.03
5.32
0.26
0.43
0.12
1.08
1.09
1.62
0.69
0.40
0.14
0.59
0.33
0.55
1.28
0.72
1.59
0.58
2.40
MET
10.94
1.11
10.19
40.42
3.82
12.47
0.76
9.71
0.49
8.32
0.17
16.94
27.29
16.27
6.52
10.50
6.26
5.44
4.63
5.75
3.97
0.55
1.59
1.55
4.56
2.24
7.42
4.07
0.19
0.37
2.57
6.84
5.82
2.82
1.90
5.69
1.71
4.67
8.57
3.18
3.14
2.77
6.34
3.36
2.15
2.73
0.73
6.91
2.34
1.51
0.61
5.60
0.53
0.24
0.72
0.49
0.34
1.57
1.36
1.69
0.76
0.51
0.19
0.71
0.39
0.64
0.28
1.92
0.64
2.76
URG
12.22
1.14
9.53
41.78
6.52
12.59
0.15
17.86
26.71
17.91
10.53
6.00
7.34
4.46
5.27
1.86
3.09
0.67
5.95
3.07
0.24
0.52
3.78
9.30
8.73
4.34
2.27
5.73
1.94
5.03
8.84
3.67
3.12
2.58
6.97
3.23
2.61
3.07
0.74
8.28
3.75
1.84
0.73
3.35
6.02
0.25
0.56
0.17
1.58
0.31
0.39
2.07
1.62
2.30
0.95
0.70
0.21
0.61
0.73
3.13
1.01
2.49
0.76
3.08
VAPS
13.15
1.28
42.34
4.16
13.31
1.19
0.62
8.58
0.16
17.43
27.04
20.25
6.81
11.13
7.88
5.96
4.29
0.53
1.75
2.57
0.62
5.96
4.26
9.11
4.59
0.20
3.85
8.77
4.29
5.97
2.96
6.31
3.23
8.07
3.00
2.40
3.30
6.86
3.22
1.99
2.78
0.73
7.93
1.73
0.62
3.53
6.40
0.24
0.64
0.20
1.34
0.42
0.40
1.76
2.68
1.02
0.80
0.41
2.82
3.25
0.86
2.68
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of NH4 (jig/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
3.21
0.28
2.45
13.20
1.13
4.00
0.10
2.45
0.11
2.44
0.12
4.90
9.00
4.18
1.69
2.88
1.38
1.40
0.98
1.37
1.02
0.79
0.06
0.24
0.33
0.16
0.89
0.47
1.48
0.74
0.05
0.21
0.57
1.42
0.77
0.52
2 22
1.00
3 22
1.18
3.76
5.53
2.35
1.56
1.96
3.82
2.55
1.50
2.09
0.59
3.97
3.68
1.72
0.55
1.47
2.52
0.65
0.51
0.24
1.82
0.49
1.30
2.03
2.31
1.34
1.08
0.70
0.89
0.96
1.62
0.90
0.93
1.16
2.38
0.79
1.50
FRM
3.69
0.39
3.00
13.20
1.49
4.56
3.04
0.26
2.76
0.14
5.41
9.34
4.24
1.81
3.21
.65
.73
.19
.87
.31
.50
0.20
0.51
0.70
0.23
1.55
0.86
2.52
1.69
0.13
0.26
1.12
2.44
1.93
1.13
2.39
1.21
3.10
1.38
3.66
5.78
2.59
1.80
2.05
4.03
2.43
1.72
2.38
0.78
4.02
3.86
1.83
0.89
1.97
2.82
0.58
0.53
0.23
1.88
0.42
1.21
1.99
2.26
1.45
1.15
0.64
0.85
0.86
1.50
0.91
1.26
1.13
2.30
0.77
1.78
IMP
2.75
0.16
1.85
10.60
0.92
3.43
0.10
2.10
0.06
2.19
0.10
4.24
7.83
4.35
1.48
2.69
0.87
1.01
0.36
1.04
0.83
0.69
0.06
0.25
0.39
0.15
0.70
0.40
1.56
0.95
0.06
0.20
0.49
0.99
0.65
0.52
0.95
3.07
1.08
3.53
5.00
2.14
1.40
1.63
3.56
2.32
1.36
1.96
0.51
3.04
3.34
1.52
0.49
1.29
2.08
0.41
0.35
0.12
1.59
2.16
1.23
0.95
0.62
0.89
0.88
1.44
0.84
0.80
0.98
2.30
0.75
1.34
MET
3.13
0.16
2.65
13.48
1.31
4.43
0.05
2.94
0.08
2.66
0.08
5.31
9.27
4.69
1.58
3.39
1.71
1.68
1.09
1.66
1.31
0.97
0.10
0.32
0.38
1.23
0.53
1.94
0.99
0.08
0.25
0.67
1.56
1.22
1.09
3.56
1.34
4.09
5.64
2.53
1.75
2.13
3.96
2.75
1.58
2.11
0.66
4.24
3.80
1.89
0.67
2.44
0.52
0.23
1.89
0.53
1.28
2.14
2.42
1.26
0.99
0.58
0.86
0.89
1.47
0.90
0.80
2.29
0.72
1.41
URG
3.61
0.43
2.96
14.34
1.58
4.86
0.32
3.45
0.37
2.99
0.17
5.98
9.42
5.32
2.09
3.85
.86
.74
.25
.71
.14
.36
0.02
0.37
0.64
0.20
1.18
0.79
2.24
0.81
0.20
0.27
1.08
2.24
2.01
1.18
1.23
3.40
1.30
4.03
5.83
2.61
1.57
1.83
4.26
2.44
1.68
2.35
0.63
4.49
3.64
1.83
0.69
1.66
2.46
0.73
0.65
0.28
1.93
0.45
1.50
2.37
2.59
1.56
1.23
0.82
0.94
1.00
1.75
1.02
1.52
1.40
2.54
0.98
1.70
VAPS
3.50
0.24
1.29
13.31
1.46
4.35
0.24
2.84
0.15
2.55
0.06
4.79
8.39
4.91
1.57
3.48
1.89
1.80
1.49
0.17
0.50
0.70
0.25
1.59
0.86
2.42
0.85
0.12
0.27
1.11
2.14
1.91
1.05
1.87
0.89
2.63
1.24
4.91
2.09
1.17
1.55
3.45
1.98
1.35
1.90
0.57
3.84
3.42
1.46
0.53
1.42
2.45
0.71
0.58
0.27
1.80
0.46
1.29
2.78
1.64
1.36
0.81
1.07
1.92
0.90
1.28
2.85
0.82
1.79
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of OC (Mg/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
10.26
10.04
14.21
4.28
8.13
2.46
7.71
4.79
5.89
1.96
6.97
8.34
8.26
6.67
3
11.22
9.49
12.03
14.63
12.13
7.60
6.87
14.71
5.78
4.73
10.37
7.23
10.78
9.21
5.54
8.50
8.53
10.25
8.25
8.82
8.85
5.94
6.36
3.13
3.55
9.14
5.68
4.60
6.11
6.76
4.04
4.79
6.56
3.62
11.01
6.82
3.10
3.72
4.69
3.95
2.83
5.52
2.67
7.43
2.32
2.45
4.00
5.59
3.11
2.96
3.85
1.70
4.89
3.96
2.54
10.07
3.31
2.76
2.02
3.75
FRM
8.69
4.58
7.72
10.01
2.98
6.89
7.78
4.58
4.65
2.05
5.53
5.69
6.93
5.50
4.19
11.75
7.86
9.48
11.79
10.14
5.60
6.23
12.51
4.34
3.96
8.98
5.60
9.67
7.69
3.99
7.29
7.34
8.34
6.45
7.75
7.59
4.89
5.17
2.31
2.80
9.57
4.05
4.13
3.10
5.32
3.40
4.46
4.60
2.31
8.54
4.41
2.78
2.77
3.66
3.30
3.10
4.42
2.79
7.27
2.49
2.40
3.78
4.66
2.39
2.00
3.75
2.21
3.32
1.81
8.51
3.37
3.11
1.46
3.24
IMP
7.85
2.98
4.16
5.26
1.75
2.97
1.25
3.28
2.56
3.98
1.58
5.86
6.45
5.91
4.58
3.83
7.65
5.80
6.72
9.42
10.21
4.04
5.45
10.11
3.96
3.25
8.10
4.45
7.61
6.58
3.51
5.79
6.09
7.50
5.64
6.12
2.16
5.82
1.93
2.29
7.83
3.82
3.95
2.74
4.54
3.24
3.74
3.71
1.92
7.83
4.71
2.39
2.98
3.71
3.10
2.26
3.04
2.01
5.99
1.79
1.83
3.97
1.78
1.72
2.83
1.29
2.76
2.47
1.62
7.51
2.25
2.31
1.58
2.69
MET
8.81
5.86
9.37
10.42
3.43
7.60
3.41
7.78
3.79
4.29
2.35
7.57
5.46
7.82
7.38
4.64
12.32
10.05
1.64
12.71
13.32
5.51
5.42
12.11
4.85
10.77
5.72
10.35
9.30
4.52
10.41
7.64
9.59
1.41
8.49
5.51
2.64
2.54
2.86
8.00
.98
.95
.24
.44
.75
4.14
4.23
2.02
10.14
4.39
0.39
2.39
3.07
4.62
2.65
7.20
2.80
2.83
3.83
5.33
2.86
4.01
2.44
3.51
4.15
2.27
10.33
3.06
1.77
3.62
URG
8.01
4.18
6.81
11.49
3.08
5.72
2.64
6.39
3.24
3.83
1.66
5.86
6.40
6.32
5.17
3.94
7.78
6.38
7.64
8.33
8.63
4.05
4.21
9.19
3.69
2.16
9.23
5.61
9.72
4.91
3.72
5.64
6.95
7.21
6.56
6.38
3.90
4.72
1.97
2.10
7.57
3.48
3.64
2.93
4.48
3.07
3.36
3.64
1.53
8.29
4.20
2.16
2.20
3.06
3.21
1.89
3.38
1.72
6.22
0.89
1.74
4.22
4.04
2.06
1.95
2.96
1.68
2.87
2.69
1.51
7.95
2.72
2.09
1.47
2.87
VAPS
5.75
2.80
14.99
14.50
3.57
7.65
1.66
2.51
3.62
1.16
5.68
6.70
6.23
4.82
5.45
6.76
4.76
8.32
3.28
7.84
2.94
1.72
5.51
4.04
7.10
4.54
1.81
4.68
4.08
4.21
2.96
3.98
3.92
1.89
4.12
1.99
6.06
2.34
1 .88
2.26
3.84
2.45
2.42
2.46
1.31
5.65
3.30
1.62
1.24
2.07
2.33
2.42
2.94
1.46
5.56
2.02
1.64
6.85
1.86
1.80
3.06
3.13
2.99
1.17
7.34
2.25
1.19
Shaded values are from replicate sampler to replace missing primary OC data.
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of EC (jig/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
6.1
4.77
5.69
1.64
3.05
0.57
3.6
1.45
2.08
0.21
2.29
4.54
3.51
3.31
0.96
5.18
4.23
4.72
5.16
5.43
1.98
2.21
4.09
1.38
1.26
2.98
2 32
3.20
2.41
0.99
1.93
2.73
3.77
2.60
2.65
3.69
1.01
1.59
5.35
2.13
1.55
2.13
1.53
2.25
3.50
1.94
1.91
2.35
1.11
5.28
3.28
1.45
1.46
1.51
1.59
0.77
1.57
0.87
2.96
0.38
0.88
2.08
2.41
0.68
0.90
1.83
0.83
2.29
0.99
0.57
2.66
1.16
1.21
0.62
1.25
FRM
6.51
1.65
5.22
7.70
1.61
3.22
4.44
1.59
2.50
0.56
2.86
5.43
3.08
2.64
1.90
5.01
3.86
4.56
4.82
5.71
2.29
2.37
5.03
1.86
1.53
4.56
3.26
4.28
3.28
1.47
2.70
2.99
3.98
2.37
2.64
3.62
2.10
3.54
0.82
1.64
5.21
2.07
1.73
1.91
3.35
2.45
2.42
2.71
1.16
5.73
3.43
1.63
1.71
1.91
1.57
1.30
2.17
1.41
3.70
0.76
1.16
2.51
2.84
0.90
0.96
1.60
0.78
0.82
0.55
3.06
1.20
1.06
0.65
1.43
IMP
4.55
0.81
1.49
1.85
0.37
0.73
0.35
1.13
0.85
2.45
0.46
2.47
4.24
2.86
2.38
1.68
5.12
3.74
3.92
4.77
4.26
1.49
1.96
3.78
1.41
1.12
3.54
2.66
3.81
2.93
1.41
2.26
2.49
3.33
1.97
2.45
0.58
2.32
0.77
1.43
4.71
2.01
1.82
1.79
2.95
2.10
1.97
2.08
1.11
4.45
3.38
1.70
1.83
1.60
1.42
0.99
1.29
0.96
2.21
0.36
0.88
2.24
0.64
0.76
1.42
0.65
1.86
0.80
0.56
2.40
1.00
1.08
0.76
1.21
MET
5.68
2.18
4.95
7.23
1.56
2.76
0.58
2.99
1.46
2.67
0.51
2.99
4.03
3.38
3.01
1.64
5.48
4.30
0.10
4.50
4.93
2.03
2.11
3.90
1.54
4.32
2.96
4.36
3.33
1.67
2.84
2.76
4.14
0.12
2.85
1.97
1.47
0.79
1.67
4.93
2.13
1.80
1.80
3.13
1.76
2.06
2.17
0.91
6.25
2.77
0.00
1.37
1.34
1.36
1.12
3.12
0.52
1.01
2.20
2.70
0.99
1.60
0.73
1.72
0.75
0.42
2.71
1.13
0.75
1.17
URG
5.90
1.40
4.17
6.73
1.50
2.78
0.87
3.77
1.31
2.44
0.54
3.28
4.97
3.98
3.37
2.03
5.87
3.81
4.26
4.47
5.37
2.05
2.61
4.89
1.92
1.30
3.96
3.36
4.23
2.94
1.53
2.39
2.87
3.61
2.35
2.46
1.91
3.60
0.77
1.68
4.56
2.09
1.80
2.20
3.47
2.07
2.15
2.08
0.98
5.48
3.34
.41
.39
.46
.50
.05
.61
.20
3.36
0.37
1.20
2 22
2.70
0.81
0.93
2.04
0.92
2.29
0.96
0.59
2.94
1.20
1.25
0.83
1.02
VAPS
6.08
1.50
5.48
5.72
1.55
2.95
0.76
1.39
2.63
0.39
2.55
4.53
3.19
3.10
1.92
5.74
4.42
5.19
2.02
4.32
1.66
1.20
3.49
2.76
3.91
2.68
1.13
2.23
2.41
3.22
1.95
2.26
2.33
1.64
2.85
0.64
4.73
1.72
0.93
1.60
2.80
1.50
1.81
2.04
1.22
5.33
3.09
1.53
1.60
1.34
1.03
1.07
1.83
1.09
3.58
0.38
0.94
4.52
0.68
0.81
1.91
1.97
0.66
0.41
2.37
1.07
0.75
Shaded values are from replicate sampler to replace missing primary EC data.
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of S (ng/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND FRM IMP MET URG VAPS
256
93
208
2235
624
706
84
210
104
315
119
636
1473
802
751
268
392
301
357
631
403
84
222
129
148
237
166
268
401
80
237
402
459
399
434
1114
567
1541
980
2363
2953
1454
628
1072
1942
1461
876
1454
434
2154
2467
1270
395
578
941
714
509
259
1884
622
1258
1946
2247
1184
1037
638
890
865
1828
833
689
999
2130
657
1012
247
102
205
1928
606
242
111
311
121
645
1392
646
175
788
256
375
300
366
643
386
71
238
122
162
227
177
285
402
85
237
407
480
435
1131
575
1323
2233
2794
1430
640
1033
2027
1481
900
1474
441
2203
2411
1290
394
590
892
698
506
262
1821
612
1231
2004
1171
692
867
861
1953
889
712
1007
2084
678
1046
237
199
2119
570
580
98
147
89
306
113
605
1359
760
143
718
228
344
328
301
880
362
73
159
124
143
234
170
268
364
72
229
377
425
394
620
1500
935
2124
2756
1326
617
959
1752
612
837
407
1995
2226
397
557
870
646
463
241
1742
570
1148
2010
972
934
563
920
809
1841
855
693
1017
2180
619
931
164
331
2288
664
784
106
292
152
353
128
704
1439
836
286
807
374
484
359
454
784
431
127
296
175
313
149
315
447
92
284
461
448
490
553
639
1665
905
2523
3138
1464
674
1113
2049
1636
961
1558
461
2287
2519
1339
451
987
554
289
1870
632
1364
2086
2366
1188
1026
717
905
870
2102
888
709
2079
715
1122
260
1100
205
2310
814
131
665
1457
818
798
298
358
654
384
221
126
143
238
158
77
392
463
412
440
1219
651
1480
1010
2503
3239
1520
694
1102
2071
1509
927
1492
440
2182
2482
1337
432
663
969
752
547
280
1949
655
1289
2094
2329
1276
1080
755
921
892
954
790
1070
2267
764
1065
273
672
779
106
269
130
344
146
651
862
162
839
266
363
597
151
218
121
138
227
265
424
77
346
436
1372
654
1567
3034
1416
681
1031
2000
1619
867
335
2170
1257
915
778
537
254
1899
624
1332
2313
356
1018
690
1969
732
2304
686
1131
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of Si (ng/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
492
264
449
166
67
86
0
140
166
65
179
174
332
170
113
532
510
469
501
450
472
274
474
179
88
450
60
95
225
72
475
390
475
446
342
70
53
67
0
34
84
60
35
46
89
29
32
40
10
126
64
48
36
71
76
31
72
69
208
41
30
67
100
35
23
47
35
71
36
39
164
26
100
33
91
FRM
420
218
378
217
16
150
146
43
101
121
236
119
280
72
461
500
467
475
403
417
212
444
133
75
334
56
94
172
50
395
312
375
320
57
45
52
49
97
41
31
25
59
40
24
19
26
133
25
22
32
61
72
51
30
71
195
23
34
38
24
72
36
77
19
42
227
42
101
41
106
IMP
621
609
206
24
41
30
116
166
37
150
171
284
135
341
81
478
503
560
472
632
406
257
353
162
115
464
127
103
213
94
452
411
430
487
38
83
21
48
71
35
45
45
64
71
41
14
153
64
49
45
81
53
43
67
205
38
53
104
27
33
24
28
65
14
24
161
41
116
42
67
MET
1715
2409
509
130
208
53
694
915
77
792
672
642
613
1627
134
2732
2530
1752
2918
2573
1236
1409
1749
671
2347
104
578
530
371
1636
1341
1466
1573
1134
99
93
0
77
192
83
64
0
175
62
75
149
38
318
66
67
235
208
66
98
226
0
100
73
228
30
48
157
58
153
0
57
286
34
16
105
URG
430
195
339
116
53
110
126
244
106
83
341
377
295
301
314
90
43
315
42
47
267
294
300
206
78
40
61
15
39
85
44
41
48
90
4
34
32
0
131
50
31
26
46
86
32
44
76
185
29
29
66
115
9
20
83
38
57
42
143
53
104
54
77
VAPS
805
341
407
177
1025
888
191
988
794
916
2270
585
648
634
491
3958
585
196
113
480
157
266
70
390
468
85
76
51
112
47
16
57
70
34
45
3
143
13
97
88
65
69
258
315
109
120
290
35
78
51
258
106
65
116
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
10
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of Ca (ng/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
500
217
473
261
56
106
27
141
293
76
86
177
187
164
77
238
233
233
213
184
178
118
194
63
39
207
38
52
87
44
143
164
206
167
142
35
32
67
20
24
73
26
25
31
43
16
33
45
13
127
27
19
25
32
33
13
14
15
158
29
17
26
61
16
6
36
11
25
15
13
77
12
20
19
15
FRM
397
168
360
343
34
140
246
52
64
112
125
80
243
61
209
213
219
188
159
151
89
167
40
26
149
29
39
65
33
124
130
158
116
32
31
53
24
59
24
27
25
48
17
28
66
13
107
23
14
20
28
23
22
14
17
146
27
16
17
14
44
13
29
17
14
88
14
17
21
18
IMP
587
567
282
57
83
30
101
284
64
87
166
159
138
285
79
219
215
256
185
262
167
122
138
58
50
214
68
49
87
63
144
173
190
177
36
48
25
23
67
24
25
27
38
36
31
13
124
25
26
30
32
20
16
17
172
21
15
65
11
7
21
10
27
19
14
75
12
17
20
12
MET
1606
2407
751
89
198
68
703
1697
66
386
728
321
711
1784
89
1231
1171
815
1262
1087
559
640
697
312
1115
51
238
224
221
506
610
656
637
432
81
51
14
41
136
66
51
23
104
42
60
83
16
316
27
41
92
63
18
36
168
25
30
23
68
19
8
59
23
58
16
15
119
17
23
15
URG
409
169
324
155
72
64
112
122
98
49
170
148
119
110
127
34
24
140
28
30
114
123
103
76
33
40
46
17
18
65
22
26
23
34
13
24
37
10
102
23
16
20
29
22
15
12
19
140
21
16
19
43
10
5
33
11
22
14
70
12
14
17
12
VAPS
715
207
426
125
921
1565
330
465
746
753
1836
344
286
278
212
1747
237
72
39
225
51
96
53
165
196
48
41
55
69
24
19
25
42
16
26
4
122
17
27
20
18
18
182
23
16
52
121
6
40
18
100
19
25
18
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
11
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of Fe (ng/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
520.3
149.2
482.1
292.2
73.5
136.8
31.5
204.9
124.7
80.3
68.2
241.4
282.1
193
110.7
445.8
366.3
453.6
360.1
340.9
208.8
212.5
342.7
128.1
69.7
324.9
94.9
171.4
218
77.9
286.8
244.1
372.6
234.4
228.5
163.2
102.6
158
33.5
74.6
286.3
74.6
64.1
83.7
206.7
84.2
114.8
140.4
19.1
317.8
90.2
35.9
39.4
57.2
71.6
38.9
36.4
50.4
133.3
15.5
41.6
49.9
94.9
26.6
15.8
75.8
43.1
101.4
16.9
27.5
62.4
28.5
36
39.6
41.6
FRM
435.7
140.5
385.9
239.4
54.6
220.1
113.8
72.2
58
166.2
209.9
125.7
276.6
92.5
404
354.1
420.6
337.6
330.3
186
173.7
304.8
94.5
57.1
270.3
73.3
141.5
174
59.8
257.3
200.8
298.4
190.2
158.1
105.6
101.8
69.4
256.3
67.1
69.3
89.3
221.5
68.8
92.3
138
18.1
303.3
76.4
24
34
56.2
53.2
36.7
36.6
65.5
124.6
12
32.7
45.5
21.5
84
54.6
103.1
10.9
31.9
82.2
40.3
34.6
40.8
45.5
IMP
573.3
486.3
324
66.8
105.4
39.6
150.7
120.8
75.1
68.9
203.4
233.5
190.3
291.4
101.9
396.8
333.4
473.2
310.1
485.6
185.2
205
227.7
109.6
77
323.4
125.4
160.8
214.1
122.8
283.9
249.2
331
251.5
99.4
144.1
40.8
67.7
287
70.3
62.6
84.6
194.1
74.1
96.8
16.6
310.3
75.1
44.9
54.5
66.5
39.2
55
53.6
131.8
7.2
36.3
90
17.3
20.2
70.3
46.2
94
14.2
29.7
66.3
31
36.6
39.2
32.2
MET
724.5
1336
401
110.2
184.9
50.3
504.2
426.7
67.5
310.8
498.9
362.8
413.2
928.7
132.2
1217.5
1098.8
871.3
1186.3
1164.2
480.8
597.7
777.9
292.2
971.4
100.3
338.8
316.3
241.9
643.6
559.6
672.4
578.3
420
131.1
169.6
17.7
70
352.7
102.4
83
74.1
278.9
92.3
153.3
189.6
24.2
509.1
85.7
60.7
83.5
102.2
45.5
84
147.2
7.7
48.8
58.6
123.9
31.4
12.3
121.6
58.7
123.1
14.7
35.8
122.4
37.4
49.1
46.6
URG
466.9
134.9
363
228.1
112
51.5
171.7
213.6
159.1
85.6
362.6
279.1
259.6
138.4
258.3
74.7
44.1
259.1
65.5
67.5
180.1
248
171.5
143.4
161
106.4
133.5
25.9
67.5
272.3
67.8
58.3
77.9
217.6
58.4
103.9
132.4
15.9
282.4
74.6
30.4
35
59.1
53.6
33.8
34.4
51.2
120.4
8.6
32.7
40.4
77.2
21.2
14.1
78.2
44
83.9
23.8
68.4
31
35.5
41.9
31.3
VAPS
688.6
253.5
368.4
100
787.9
470.7
195.2
406.3
596
618.6
1137
390.7
498.5
415.6
402
1618.5
398.7
134.8
76.7
350.2
161.4
249.8
78
222.7
344.1
212.9
113.9
159.5
286.4
70.6
26
73.6
209.7
77.6
105.1
9.9
311.7
31.5
57.2
48.9
44.9
60.9
148.6
7.7
42
86
95.2
25.6
101.5
16.5
85.3
44
56.7
52.9
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
12
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of Zn (ng/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
243.9
7.4
251.7
159.5
8.1
13.8
3.9
24.4
136.9
75
4.5
17.4
53.9
19.5
12.1
29.9
27.9
30.5
24.4
27.8
28.4
11
24.4
17.1
3.7
18
8.3
12.6
17.3
4.4
17.1
17.2
20.1
16.7
12.8
37
27.8
80.2
7.6
25
95.3
19.1
15.6
23.6
43.9
32.6
25.9
20
4.9
64.4
37.8
9.8
8.6
29.7
14.9
5.4
9.5
15.7
46.9
6.5
8.6
23.7
21.7
12.8
4.4
14.4
8
11.8
3.3
10.2
13.1
13.5
13.5
7.5
9.8
FRM
244 2
9.2
254.8
85.4
8.2
25.5
139.2
78.8
3.9
17.4
47.1
15.9
156.1
12
30.6
30.3
47.5
25.8
25.1
25
11.4
24.8
15.7
1.5
19.2
11.8
10.1
15.7
3.8
16.5
15
22.9
14.9
40
25.1
65.1
21.2
83.3
19.8
16.8
22.4
48.7
30.4
22.9
35.4
5.2
67.8
35.6
10.6
6.7
30.8
15.4
3.8
11.9
18.7
43.4
5.8
7.6
17.8
54.9
16.7
6.4
11.8
5.3
11.1
14.6
14
13.2
7.9
9.9
IMP
240
246.3
163
9.1
10.7
4.6
18.9
115.4
71.6
4.3
16.3
44.4
16.5
116.9
11
27.6
25.4
28.4
20.1
39.8
25.6
9.9
18.5
17
4.1
19
10.3
10.8
14.3
4.6
16.5
16.1
16.9
14.7
28.7
79.6
7.5
19.6
90.6
16
16.9
20.3
39.2
12.4
21.7
4.9
63.3
35.4
8.7
30.9
15.4
5
10.6
16.1
36.5
3.7
7.7
16.8
1.1
0.9
36.5
9.6
12.7
5.6
10.8
14.4
12.7
13.7
6.9
8.6
MET
17.3
265.5
163.2
11.9
14.8
2.8
41.7
140
73.2
5.8
61.2
55.7
22
180.1
12.7
51.3
51.5
42.4
45.5
53.4
30.2
18.7
37.9
19.9
38.1
11.1
18
18.4
6.4
26.2
24.3
26.9
27
15.4
31.4
75.5
6.1
21.7
98.2
18.4
14.5
25.8
51.3
33.7
23.9
24.4
4.9
64.8
31.3
9.6
10.6
12.9
11.8
16
45.5
6.3
7.9
20.1
3.1
12.2
3.6
14.1
9.5
11.8
3.3
9.7
13.6
11.6
6.5
10.5
URG
257.1
11.9
270
174
16
2 3
16.4
54.2
19.5
13.2
29
25.6
25.1
25.1
24.2
15.9
1.9
20.2
8.5
3.3
15.4
18.9
13.4
11.1
43.2
33
76.9
8
21.5
101.3
19.5
17.7
24
48.2
33.9
23.3
21.9
5
63.1
42
11.4
8.5
33.1
14.6
5.1
11.2
19.2
44
6.5
6.2
21.4
23.7
13.1
6.8
14
9.5
13
11.3
16.8
14.4
13.3
8.7
14.1
VAPS
250.3
12.7
21.5
6.9
32.9
133.5
79.8
9.8
21.8
24.6
62.5
17.8
32.2
28.4
28
32.6
24.6
16.9
4.2
19.8
10.4
20.3
2
10.6
19
50.9
34
86.8
98.9
17.4
15.5
19.9
45.9
36.2
25.5
3
61.7
9.1
14.8
4.5
11.8
18
47.6
4
5.3
24.9
5
5.4
13
5.3
15.9
16.2
10.2
13
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
13
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of K (ng/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
139.3
66.9
135.8
161.9
53.9
112.5
16
104.8
52.1
63.8
30.8
99.7
132.3
90.9
58.4
158.5
155.7
194
225.2
232.9
192.1
87.2
237.1
70.6
30.3
138.7
59.9
116.5
91.1
32 3
144.7
133
163.3
193.2
131.7
91.6
33.6
61.5
26.4
38.4
119.1
46.8
57.6
37.9
66.3
45
49.5
56.9
28.9
116.4
68
42.1
33.8
38.3
54.3
42
85.2
28.7
183
43.8
35.9
75.6
91.4
49.5
34.7
68.5
32.7
57.4
69.1
35.7
134.6
43.9
69.2
30
87.6
FRM
130.9
63.6
121
122.5
40.1
124.7
48.4
62.5
23.5
93.4
112
77.1
80.7
49.9
150.9
150.2
194.5
225.6
237.2
181.1
81.9
244.8
62.4
29.5
124.2
58.8
123.6
83.5
31.5
134.6
127.2
159.7
188.2
95.3
31.4
53.9
35.7
111.9
44.2
58.6
36.1
71
48.5
45.5
57.4
31.1
115.1
63.9
40.2
28.5
37.5
51.9
39.9
80.5
32.5
176.9
38.7
35.2
77.1
44.6
74.9
31.9
58.1
72.3
36.5
146.4
49.1
69.6
31.7
95.2
IMP
156.2
143.8
162
53.7
87.9
18.9
72.7
43.7
60.8
30.2
95.1
115
82.9
79.7
56
139.2
141.3
211.2
194.5
327.4
173.5
86.6
172.8
68.8
34.3
138.2
65.3
117.3
87.3
38.3
137.3
136.6
153.2
191.3
34.2
62.7
28
35.8
111.4
43.4
58.9
36.8
58.6
70.5
46.7
28.4
105
62.5
34.1
35
52.6
38.2
79.3
31.4
179.9
45.4
33.6
81.4
40.3
35.8
48.1
34
54.6
66.8
36.4
137.5
46.8
66.5
30.8
78.6
MET
330.1
473.9
231.5
65.7
146.8
25.7
200.4
167.1
60.7
130.8
186.8
175.7
170.3
320.9
71.2
491.9
492.8
379.3
628.3
561.3
325.1
268.2
450.5
149.8
436.3
60.4
196.2
148.3
81.9
337.6
292.9
326.4
376.7
230
36.6
57.7
23.3
46.6
137.7
60.1
65.4
36
75.1
55.6
59.1
64.2
35.6
147.1
68.8
51.8
62.9
73.7
91.2
45.4
189.8
43.6
46.8
79.1
99.3
52.3
31.5
85.9
45
75.8
77.5
39.9
167.2
62.9
32.6
93.9
URG
127.5
65.2
125.9
164
122.3
24.6
94.6
114.4
85.5
47.2
177.8
213.5
226.8
165.9
229.1
58.6
24.7
124.5
55.3
31.3
122.1
143.6
177.4
117.1
100.4
36.9
68.7
23 3
35.3
125.6
49.6
65.9
38.3
65.7
45.9
46.8
58.3
29
115.3
71.2
43.3
34.6
38.5
54.5
43.6
87.7
34.5
190.6
36.8
38.4
77.2
92.5
48.3
34.1
80.8
32
56.2
39.6
149.3
49.6
68.1
35.2
90.8
VAPS
193.3
116.7
181.1
32.9
246.9
147.7
86.2
149.5
200.2
218.4
374.4
184.6
177.7
167.6
238
610.1
264.6
69.7
32.9
145.9
117.6
95.8
35.2
130.2
166.1
115.9
43.3
66.5
128.4
44.8
62.1
36.3
68.9
52.6
49
19
122.8
39.2
53.7
46.2
90.4
33.3
204.6
47.8
36.9
94.8
66.3
37.6
77
72.9
155.9
72.7
34.4
95
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
14
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of Mn (ng/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
11.1
1.7
9.0
6.4
1.0
2.6
1.3
3.9
2.0
0.6
1.1
4.5
7.9
4.5
3.7
7.4
6.0
7.5
7.0
5.1
5.5
3.8
4.8
3.4
1.3
6.6
1.8
1.8
4.9
0.0
3.1
4.2
5.2
3.8
3.0
2.6
1.6
7.4
0.0
3.3
7.6
1.1
1.3
2.1
5.0
1.9
3.1
3.9
1.2
6.1
4.7
0.8
1.6
3.2
3.2
0.0
0.0
2.3
6.4
1.6
1.0
0.0
1.9
0.0
0.0
1.6
1.1
2.0
2.1
2.8
2.0
1.1
4.7
1.8
1.4
FRM
9.8
1.5
10.6
8.1
1.2
4.5
2.7
1.6
1.7
3.3
6.0
2.3
7.0
2.6
8.3
5.7
6.6
6.5
4.7
4.1
2.8
3.7
2.3
0.0
4.7
0.0
2.6
3.6
0.9
4.5
5.3
5.6
3.8
2.3
1.7
4.9
2.0
6.0
2.4
0.0
2.6
3.9
1.9
4.1
5.2
2.1
7.1
2.6
1.3
0.6
2.2
2.6
1.0
0.7
2.3
7.1
1.2
1.0
1.6
0.0
2.1
0.6
1.7
1.4
0.8
3.2
2.4
2.2
1.8
1.2
IMP
12.0
8.7
6.1
1.1
2 2
1.3
2.5
2.0
1.7
1.9
3.6
5.9
3.7
6.7
3.1
7.5
7.2
6.5
4.5
7.8
5.5
3.8
2.7
2.5
1.6
6.9
2.1
2 3
3.4
2.5
3.4
4.0
5.7
3.8
1.8
3.4
0.4
0.9
7.8
2.6
1.7
2.5
5.1
1.1
3.4
1.1
6.6
2.4
1.2
2.7
2.6
0.0
0.8
3 2
5.4
0.0
0.0
3.1
0.0
0.0
1.8
1.4
2.1
1.9
1.9
2.6
0.7
2.7
0.9
1.1
MET
17.8
31.5
10.9
2.6
3.9
3.9
7.9
4.8
0.0
9.0
7.1
8.8
10.8
24.7
4.4
24.1
24.4
18.1
25.1
23.8
13.2
13.4
14.4
4.8
18.2
2.9
7.5
5.1
6.6
10.1
14.7
16.8
11.3
8.4
3.6
1.5
2 3
4.8
7.0
2.4
1.8
3.8
4.3
2 2
6.5
9.5
0.0
8.0
5.0
3.3
4.6
0.0
1.7
2.1
6.2
0.0
2.4
0.0
5.3
2 3
0.0
2 3
2.9
3.9
2.7
2.0
7.4
2.0
2.1
0.9
URG
10.2
3.0
7.6
6.4
3.3
1.1
2.3
8.2
3.1
2.9
7.3
4.2
3.3
5.0
2.9
1.8
1.4
7.0
0.7
1.7
4.2
4.0
4.6
2.4
1.7
2.4
4.6
1.6
2.1
8.8
2.5
0.0
2.4
4.6
3.4
3.9
3.9
1.0
5.7
3.7
1.4
0.0
2.2
2.7
0.0
0.9
2.7
7.7
0.0
0.8
1.5
3.7
0.0
0.0
1.9
2.8
3.1
2.8
3.6
1.1
2.1
2.7
1.8
VAPS
13.1
3.6
4.7
1.6
11.7
9.8
1.9
9.7
8.7
13.1
21.8
7.8
9.4
7.9
5.6
32.3
5.3
3.0
0.8
6.9
4.0
3.4
1.4
5.4
5.6
1.1
1.1
3.6
8.1
0.7
0.0
2.9
5.3
1.7
2.7
0.0
8.1
0.0
2.4
0.0
0.0
2.1
6.4
0.0
0.0
3.7
4.0
0.0
3.6
1.6
5.5
4.4
1.1
0.0
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
15
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of Cu (ng/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
16.3
2.4
10.5
10.8
1.7
5.2
0
4.5
2.9
6.4
1.7
3.5
10.1
4.7
2.8
17.8
18.6
11.6
13.9
22.9
7.6
3.9
15.2
1.6
1.9
9.1
3
5
12.6
1.5
9.9
5.6
11.3
5.3
11.2
6.9
2.8
6.1
2 3
6.1
12.8
3.4
4.4
1.2
10.8
2.7
5.2
3.5
0
10.8
4.4
1.9
1.2
2.6
2.8
1.9
0
2 3
64
0
2.7
2.9
2.6
5.8
0.8
3.6
9.8
6.3
0.7
1.5
1.4
5.8
3 2
9.4
6.2
FRM
11.4
2.3
8.9
30.7
1.1
5.6
2.5
4
0
4.6
7
3
10.1
2.1
17
15.8
37.6
11.7
19
6.7
2.9
14.1
3.1
2.2
7.4
2.9
6.1
9.3
1.6
9.7
4.5
12.5
4.7
3.8
2.9
6.3
5.8
11.2
1.7
2.7
2.7
12.7
3.8
4.7
6
1.1
13.3
6.5
0.9
2.9
3.1
6
2.3
3.1
2.8
64.3
0
1.5
2.3
7.9
3.9
9.5
7.9
0
0
0.7
5
2.7
8.9
7.1
IMP
15.3
7.6
9.2
1.2
3
1.4
5.8
1.6
3
0.6
4.1
7.5
5.2
8.5
2.1
16.2
16.1
12.7
10.5
29.4
6.4
3.3
13.2
2.5
1
10.4
2.9
5.3
11.8
1.2
8.9
5.4
8.3
5.2
3.5
4.8
1.3
6.6
12.2
2.9
1.8
3.6
10.4
2.7
4
1.5
12.8
3
1.5
3
4.9
2.2
3.3
2.6
57.3
0
0.5
3.4
5.8
1.5
4.2
8.5
7.2
1.5
0
2.8
6.2
3.9
8.1
5.9
MET
4.1
11.4
8.1
0
3.1
0
9.5
4.8
7.1
0
6.8
12.2
3.7
15.1
0
30.6
26.8
17.5
22.8
34.6
6.5
3.9
19
9.9
15.9
0
7.5
14.6
2 2
14.2
6
16.4
10.1
13.7
4.3
5.8
0
5.7
15.6
0
0
2
11.9
0
6.2
4
3.6
18.1
0
2.6
0
2.1
0
3.6
69.9
0
0
3.7
0
9
0
0
11.9
9
3.1
0
3.3
1.9
13.3
4
URG
15.1
2 3
7.8
9.8
6.6
1.5
3.4
10.2
4.8
1.3
9
9
19.4
6.3
13.4
2.8
0.8
7.5
0.7
1.6
2.8
8.8
4.8
10.6
4.9
4
5.4
2.6
7.6
12.3
2 3
.1
.6
1 .2
.4
.6
.6
0
12.2
5.9
2 3
2
3.5
4.6
2
1.2
1.7
65
1.7
1.4
1.7
2.9
5.9
0
4.6
8
5.2
0
1.3
4.6
4.1
9.9
4.7
VAPS
22
4.2
14
2.3
15
5.1
6.4
0.9
7.5
9.5
9.9
2.8
22.2
16.4
22.6
15.2
16.5
5.3
2.2
10.9
4.7
15.6
1.2
5.4
10.7
9
2.8
6.5
11.3
0
4.8
3.2
11.1
2.7
5.2
0
11.4
1.7
4.3
3.3
2.9
1.7
64.8
0.9
0
5.2
6.1
1.6
4.5
0
0
4.3
10.6
4.3
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
16
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of Pb (ng/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
49.7
3.9
12.6
35.3
3.3
11.1
0.0
8.3
12.9
25.2
0.0
10.1
15.5
6.0
2.9
11.0
10.8
4.7
7.6
7.9
4.3
.7
2 .0
.0
.3
5.4
1.6
4.3
12.0
2.9
4.4
5.8
9.2
6.0
6.9
17.7
3.5
11.6
37.7
9.0
3 2
4.9
31.1
9.0
8.4
4.2
4.3
14.2
9.3
5.7
2 3
7.2
8.4
9.1
1.4
4.1
15.9
1.9
2.8
.9
.1
.7
.8
.3
.6
.4
.3
.7
.5
.7
.4
.3
1.5
FRM
49.3
3.9
11.2
18.5
6.2
11.2
12.3
26.4
2.0
7.7
14.6
3.1
49.6
1.8
10.3
10.3
10.5
8.0
4.8
21.9
2.1
0.0
4.8
2.9
3 2
6.3
0.0
4.2
8.6
8.8
5.6
8.0
15.8
7.9
35.9
11.0
5.7
6.9
38.7
10.9
6.3
3.7
3.5
20.2
7.2
6.5
2.6
7.9
8.3
1.1
3.9
15.8
0.0
1.3
4.4
4.4
4.0
5.2
1.8
2.5
3.5
4.1
3.8
7.5
2.9
2.9
IMP
47.7
9.8
37.2
4.0
8.1
2.8
5.5
8.9
25.1
0.0
8.1
15.8
6.3
34.8
4.2
6.2
8.0
14.1
3.3
2.8
17.4
1.4
0.0
6.0
4.1
5.0
6.6
0.0
3.2
7.1
7.9
6.1
17.6
17.0
0.0
7.5
32.1
8.2
3.3
3.5
29.2
2.6
7.9
5.4
16.0
7.9
2.0
7.4
4.4
2.4
2.5
14.5
3.0
2.0
5.1
2.4
1.1
11.7
2.7
3.0
0.0
2.5
5.1
3.9
4.8
3.3
2.6
MET
0.0
16.8
41.7
5.5
13.9
0.0
11.6
17.7
21.7
0.0
5.6
16.1
11.4
47.7
0.0
14.0
21.7
14.4
7.0
0.0
28.2
4.6
5.4
7.1
5.6
12.9
3.2
4.1
8.1
9.0
9.9
12.0
12.8
9.6
8.0
36.9
14.6
0.0
4.8
31.2
11.4
7.8
5.5
8.8
21.7
3.7
12.6
4.4
10.3
10.0
0.0
17.7
3.2
3.2
8.6
8.0
0.0
0.0
4.4
0.0
0.0
3.4
0.0
3.4
7.7
5.7
5.5
URG
52.0
8.3
10.7
35.2
13.5
1.8
7.1
16.3
10.8
1.9
4.9
12.4
12.7
3.8
23.5
0.0
1.8
6.0
2.1
4.6
7.0
8.1
9.1
6.6
11.9
18.9
3.4
11.4
40.0
10.9
4.0
7.0
32.9
10.9
13.4
5.6
5.7
17.8
7.4
6.2
2.5
7.2
7.8
13.3
0.0
0.0
17.4
0.0
3.1
4.3
6.4
5.1
1.7
6.6
1.7
2.7
4.5
5.2
5.4
5.3
4.0
3.3
VAPS
50.5
7.8
13.0
2.4
9.8
10.0
25.5
3.4
7.4
7.4
13.6
5.4
58.1
12.0
25.1
1.9
3.0
4.5
4.6
8.2
0.0
5.2
4.6
16.3
37.5
9.7
3.5
3.9
34.1
10.9
7.2
3.4
20.1
5.4
10.0
1.5
3 2
14.9
0.0
0.0
4.5
0.0
5.6
6.2
2 3
3.8
4.8
4.8
0.0
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
17
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of As (ng/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
1.90
1.00
0.70
0.00
0.00
0.70
1.30
0.00
0.00
0.00
0.90
0.00
1.70
0.00
0.00
1.40
3.20
1.20
2.50
4.90
1.30
0.00
3.50
0.00
0.00
1.70
0.00
0.80
0.00
0.00
0.00
1.70
0.00
0.00
1.90
0.00
0.00
0.00
1.80
1.70
0.00
0.00
1.10
1.30
1.10
0.00
0.00
4.40
1.10
1.80
0.00
0.00
1.60
1.90
0.00
0.00
3.20
0.00
0.00
0.00
1.10
2.00
0.00
1.00
0.00
0.00
0.70
0.00
0.00
0.00
0.70
0.00
1.10
FRM
2.30
1.30
1.40
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
1.40
1.10
0.00
0.00
2.00
3.40
0.00
0.00
3.50
0.00
0.70
0.00
0.00
1.20
3.30
2.10
0.00
0.00
0.00
1.70
2.40
0.00
2.00
0.00
0.00
0.00
0.00
0.00
0.90
2.70
0.00
1.00
1.70
3.60
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.70
0.00
1.00
0.80
0.00
2.90
0.00
0.00
0.80
0.00
0.00
1.00
1.20
IMP
0.00
1.70
0.00
0.00
1.10
0.00
0.00
0.50
0.00
0.00
0.00
0.90
1.10
0.00
0.00
1.30
0.50
4.00
1.70
0.00
2.40
0.00
0.00
0.00
0.00
0.90
3.00
0.00
1.30
0.70
0.00
0.00
0.00
1.80
1.10
0.90
2.60
1.60
0.50
0.00
1.80
1.00
1.40
0.00
2.60
1.30
0.90
1.10
1.20
0.00
0.00
0.60
0.00
0.00
1.20
0.70
1.00
0.00
0.00
0.90
0.70
0.60
0.00
0.00
1.10
0.00
0.00
MET
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
3.00
2.30
2.00
0.00
0.00
1.10
0.00
0.00
0.00
3.30
0.00
0.00
4.10
0.00
2.10
0.00
0.00
0.00
0.00
2.90
0.00
0.00
0.00
0.00
3.70
0.00
0.00
0.00
0.00
4.20
0.00
3.10
0.00
3.20
0.00
0.00
2.10
2.20
0.00
0.00
0.00
1.00
0.00
0.00
0.00
0.00
0.00
0.00
2.30
0.00
2.50
0.00
0.00
0.00
0.00
2.50
0.00
0.00
0.00
URG
2.20
0.80
1.30
3.00
0.00
0.00
1.20
0.00
0.00
0.00
1.70
0.80
1.00
2.00
2.60
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
1.20
1.50
0.00
0.00
0.00
1.50
1.30
0.00
0.00
3.60
0.00
0.00
0.00
0.00
1.70
3.40
1.20
0.00
0.00
0.00
0.00
1.30
0.00
0.00
0.70
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.90
0.00
0.00
0.00
1.40
1.70
0.00
VAPS
0.00
0.00
1.10
0.00
0.00
1.10
0.00
0.00
1.10
0.00
0.00
0.00
13.00
1.80
2.30
0.00
0.00
1.90
0.00
2.10
0.00
0.00
1.00
0.00
0.80
2.10
0.00
1.50
2.20
1.60
1.50
0.00
1.00
1.10
0.00
0.00
0.00
3.10
0.00
0.00
2.00
0.80
0.00
0.00
0.00
2.90
2.20
0.00
2.30
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
18
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study
Average of PM-10 (jig/m3)
Site
1
2
3
4
Date
1/6/1999
1/9/1999
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/6/1999
1/8/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/15/1999
1/17/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Sampler
AND
FRM
IMP
MET
URG
VAPS
82.0
7.3
19.0
33.7
59.7
6.7
51.3
14.9
27.0
10.5
75.8
62.6
59.0
44.4
65.9
60
66.63
55.84
23.56
47.87
20.98
10.57
45.35
41.47
29.37
9.12
17.26
30.32
19.41
29.0
38.4
3.8
55.3
21.1
15.4
18.5
39.3
20.3
18.7
7.9
5.0
52.5
27.3
15.0
34.9
23.0
67.1
8.0
34.3
13.5
11.6
26.9
12.9
10.5
15.6
7.7
20.0
1.8
32.2
17.6
7.3
19.2
04 Appendices D, D1.xls/AII Data bySpecies.Date.Sampler 11/7/20000
19
-------
Appendix Dl
Final Data Archive, By Species, Site, Date,
and Sampler for Replicate No. 2 Measurements at Rubidoux
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study, Measurements at Rubidoux.
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
AND FRM IMP MET URG
32.52
28.44
87.47
12.69
29.7
2.41
21.29
5.07
19.5
2.26
34.59
60.06
34.08
18.06
23.23
34.26
7.42
27.68
84.58
12.59
29.36
2.96
21.35
5.04
19.4
2.58
34.01
57.99
29.98
17.07
22.65
33.99
7.94
28.77
89.44
13.28
32.34
2.7
13.28
4.77
19.48
2.67
32.01
57.34
30.12
18.1
21.67
71.26
21.93
98.49
34.69
3.14
31.01
15.48
19.94
8.25
41.86
64.38
40.68
41.57
25.14
38.29
8.57
31.84
111.18
16.77
38.3
3.09
5.12
2.58
41.3
71.99
35.97
22.34
35.39
Average of SO4-T (jig/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
0.58
0.51
6.08
1.61
1.74
0.23
0.52
0.25
0.7
0.32
1.65
2.03
0.48
1.93
0.61
0.29
0.52
5.92
1.67
1.77
0.24
0.53
0.12
0.68
0.34
1.6
3.64
2.04
0.43
1.94
1.14
0.47
6.09
1.7
1.87
0.29
0.7
0.43
0.82
0.42
1.81
3.79
2.2
0.84
1.99
^VVCIilJJC Ul k5W+V^ V^tJi/lIU^
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
kXllll|Jlt;l
AND FRM IMP MET URG
0.64
0.28
0.57
5.69
1.58
0.48
0.42
1 .88
4.08
1.81
0.53
2 2
0.67
0.42
0.64
5.76
1.57
1.71
0.29
0.59
0.42
0.75
0.46
1.68
3.69
1.9
0.52
1.91
04 Appendices D, D1 .xls/Replicate final Data base 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study, Measurements at Rubidoux.
Average of NO3T (jig/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
10.4
8.86
38.63
3.3
11.57
0.24
8.25
0.24
7.68
0.07
15.64
13.56
5.78
8.93
12.31
0.94
39.51
3.37
12.34
0.24
9.15
0.37
7.82
0.11
16.14
26.65
13.8
6.34
8.97
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
AND FRM IMP MET URG
11.35
10.51
44.07
3.84
13.31
0.77
10.62
0.45
8.68
0.13
17.27
30.21
16.8
6.91
10.56
11.74
0.99
9.63
36.17
2.95
4.13
0.11
17.72
28.49
13.88
7.27
10.51
10.82
0.96
9.16
37.43
3.63
12.1
0.69
8.44
0.19
7.69
0.08
15.27
25.91
14.87
6.18
10.02
12.52
1.19
40
3.64
12.25
0.75
9.57
0.52
8.21
0.14
16.6
27.01
16.61
8.6
10.31
12.58
1.22
10.81
36.78
3.21
11.11
1.33
10.9
0.62
16.27
0.23
16.37
26.82
17.17
7
9.47
Average of NH4 (jig/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
3.07
2.54
13.24
1.29
4
0.1
2.55
0.11
2.46
0.11
4.87
4.51
1.68
2.99
3.38
0.33
2.81
13.05
1.33
1.2
0.16
6
10.26
4.68
2 22
3.91
2.77
0.18
2.23
11.55
0.93
3.56
0.1
2.08
0.02
2.12
0.09
4.23
7.86
4.28
1.54
2.75
3.68
0.2
13.95
1.42
4.6
0.06
2.97
0.08
2.81
0.08
5.53
9.65
4.82
1.71
3.45
3.75
0.46
3.25
12.87
1.44
4.24
0.29
3.16
0.29
3
0.15
5.45
9.41
4.85
2.14
3.6
04 Appendices D, D1 .xls/Replicate final Data base 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study, Measurements at Rubidoux.
Average of OC (jig/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
10.26
10.04
14.21
4.28
8.13
2.46
7.71
4.79
5.89
1.96
6.97
8.34
8.26
6.67
3
8.71
4.2
7.41
12.86
4.02
6.41
4.5
2.11
6.35
7.28
7.11
5.25
4.13
#DIV/0!
4.22
7.09
9.79
3.05
6.88
2.09
6.3
3.3
3.8
1.42
5.31
6.89
6.15
6.19
3.84
11.32
4.03
11.68
3.12
6.39
2.68
8.72
3 23
4.12
2 33
6.66
7.51
8.58
8.29
5.63
7.08
3.43
7.71
10.92
2.77
6.18
2.26
7.67
2.99
4.31
2.06
5.3
8.33
6.79
5.49
4.13
Average of EC (jig/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
6.1
4.77
5.69
1.64
3.05
0.57
3.6
1.45
2.08
0.21
2.29
4.54
3.51
3.31
0.96
5.93
1.43
4.58
6.79
1.75
4
1.36
0.73
3.29
5.69
3.69
3.6
1.89
4.5
1.35
4.4
6.27
1.35
3.09
0.86
3.37
1.5
2.48
0.4
2.35
4.74
3.13
2.99
1.48
6.68
1.47
7.16
1.45
2.72
0.87
4.02
1.35
2.62
0.49
2.9
4.44
3.28
3.23
1.88
6.38
1.53
4.77
6.65
1.31
2.91
0.8
4.34
1.44
2.71
0.48
2.79
4.18
3.43
3.21
1.83
Average of S (ng/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
237.6
203.6
2233.1
584.5
703.2
94.1
194.6
104.9
301.4
102.7
616.8
1453.2
776.5
170.2
790.4
258.9
99.1
195.1
2271
562.6
599.3
94.3
213.6
111.2
306.7
113
634.2
1412.8
628.2
172.5
790.9
251.5
103.8
210.9
6409.4
546.6
631.1
98.6
135.6
306.6
103.5
585.4
1363.8
748.5
169.4
723
418.6
178.4
2417.3
754
109
284.7
164.1
353.6
156.7
698.4
1528.2
936.3
317.3
791.4
273.2
108.1
224.1
2395.7
682
798.6
99.6
104.5
127.5
639.4
1437
785.4
160.1
792.7
04 Appendices D, D1 .xls/Replicate final Data base 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study, Measurements at Rubidoux.
Average of Si (ng/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
481.6
477.4
145.3
45.4
63.7
22.1
132.7
162.4
38.3
122.7
148
305.3
145.7
386.9
89.1
494.7
244.7
345.1
115.2
95.8
105.5
60.5
326.3
132.1
24.8
101.6
123.4
255.1
93.7
289.2
75.7
672.5
316.3
645.2
600.8
34.6
58.6
37.5
107.2
39.3
144.6
133
319
169.7
463.7
102.3
3329.6
1743.9
445.1
137.1
58.7
613.2
905.3
104.2
473.7
445.7
681.5
717.3
2513.9
266.2
501.6
214.9
422.9
152.3
54.5
56.1
17.3
132.4
168.3
146.2
286.1
194.5
319.3
57.1
Average of K (ng/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
129.2
133.3
156.2
45.6
109.4
16.2
99.5
48.5
63.1
25.2
95.2
121.2
84.1
91.7
58.4
139.6
70.9
117.7
163.5
53.8
104.8
18.1
130.2
45.3
64.6
24.1
86.9
109
72.4
77.6
58.2
161.5
85.6
155.7
472
45.8
102.2
18.6
66.4
62.2
29
93.5
121.6
81.6
106.2
56.9
674.3
331.5
231.4
132.6
27.1
215.4
168.8
74
78.3
156
186.4
200
456.8
89.6
151.5
68
136.8
164.3
45.5
126.9
21.7
43.9
32
96
121.7
92.3
86.4
51.9
Average of Ca (ng/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
462.1
434.6
203.4
46.4
85.5
22.3
119.1
255.8
62.7
72.7
135.1
164.8
137.5
330.5
68.5
446.4
187
330.7
154.2
77.5
137.3
45.2
260.9
224.1
51.8
58.2
106.5
124.5
80
233.8
62.9
630.2
293.8
565.1
757.2
51.3
93.3
28.3
90.6
65.6
79.6
140.5
163.6
148.3
393
73.5
3453.8
1589.2
736.3
136.9
58.2
686.6
1837.1
136.4
275.3
479.5
343.7
803.4
2439.1
152.1
493.5
186.3
352.2
170.5
38.5
75.7
24.2
212.3
88.9
121
138.9
131.9
260.4
54
04 Appendices D, D1 .xls/Replicate final Data base 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study, Measurements at Rubidoux.
Average of Mn (ng/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
10
8.9
8.3
0
2 2
0
2.7
2.4
0.7
2.6
2.7
6.5
4.4
8.1
2.8
11.8
3.9
7.2
5.9
1
2.1
1.2
5.4
2.4
1.5
1.2
2.7
5.8
4.7
6.4
2.1
13.9
3.8
11.4
21.8
1.1
2.8
1
1.7
2.1
2.5
3.1
5.5
3.8
8.7
3.2
39.1
14.7
10.2
3.4
0
9.9
5.9
4.5
5.1
6.7
8.6
8.1
32.1
0
12.4
2.6
9.3
6.9
0.6
1.6
1.2
2.6
2
2 3
6.5
3.5
7.9
3
Average of Fe (ng/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
457.5
430.7
268
59.7
116.7
40
191.2
116.4
78.1
57.6
199.5
243.8
180.6
330.3
100.2
480.1
150
360.6
217.3
89.8
153.8
60.4
302.6
99.5
70.8
49.6
152.5
216.3
125.6
259.1
93.3
580.3
200.4
498.7
886.8
64.2
116
40
134.3
76.9
64.4
196.6
249.6
180.5
365
103.7
1799.4
734
423.7
156.6
48
515.4
427.5
95.1
203.8
379
393.5
446.8
1234.1
172.3
532.1
144
413.3
249.5
53.9
111.2
59.4
104.9
77.9
177.7
225.8
180
282.3
92.6
Average of Cu (ng/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
16.5
10
9.2
1.4
6.7
1.6
6.6
3.1
6.4
0.8
4.6
7.5
3.5
11.9
1
17.2
2.7
8.5
7.2
11.2
16.4
13.6
28.1
3.8
4.2
0
1.7
10.1
4.7
11.1
0
14.2
2
8.5
29.2
1.7
4.8
0.5
2.4
4.1
0
4
8.5
3.5
9.7
1.6
21.8
3.9
10.4
5
0
8.6
8.9
4.1
3.3
2.8
10.6
7.1
15.1
0
17.5
0
8.1
15.3
1.7
6.8
2.1
0
4.4
14
5.2
9
1.6
04 Appendices D, D1 .xls/Replicate final Data base 11/7/20000
-------
Appendix D. Final Data Archive, By Species, Site, Date, Sampler for the 4 City Study, Measurements at Rubidoux.
Average of Zn (ng/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
234.9
264
151.3
6.8
14.3
5.5
21.9
133.3
77.7
3.5
16.1
53.2
20
162.3
12.8
246.1
7.7
240
164.2
9.5
16.2
5.1
24.8
127.7
76.5
3.7
15.8
52.7
10.4
148
11.5
225.9
9.8
229.7
461
7.4
12.8
5.7
15
69.4
3
15.6
50
14.7
149.1
13.1
283.7
19.6
168.6
16.2
3.4
32.2
147.1
78.4
6.9
27.4
51.8
21.7
176
12
277.2
12.5
294.1
177.2
8.8
14.4
6.1
134.3
4.2
17.4
53.2
17.3
149
10.6
Average of Pb (ng/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
45.1
16.1
33.1
9 9
7.9
0
6.8
12.6
24.2
0
7.7
13.7
6.6
45
4.4
50
5.3
10.2
32.3
4.6
9.3
1.5
8.4
13.8
23.4
0
10.6
15.6
5.3
46.9
4.2
50.1
4.5
11.7
99.8
5
11.5
0
4.9
23.1
0.1
7.4
14.9
5.4
43.6
3.4
67.3
5.7
36.3
4.9
0
12.3
8.8
27.3
6.7
10.5
16.4
5.1
39.5
0
51.7
6.3
13.9
39.6
5.7
12.2
0
15.5
3.2
12.1
17
7.1
44.8
2.1
Average of As (ng/m3)
Site
1
Date
01/06/99
01/09/99
01/15/99
01/18/99
01/21/99
01/24/99
01/27/99
01/30/99
02/02/99
02/05/99
02/1 1/99
02/14/99
02/17/99
02/20/99
02/23/99
02/26/99
Sampler
AND FRM IMP MET URG
2.3
0
2.4
1.2
1.2
0
0.8
0
0.9
0
0.8
1.8
0
1.3
0
0
0
0
4
0
0
0
0
0
1.5
0
0
1.1
0
0
0
0
1.2
0.9
4.2
0
0
0
0
1.3
0
1
1.8
0
1.3
0
0
0
0
2.8
0
0
0
0
0
0
0
1.1
2
2.6
-)
1.4
0
0
0
0.9
0.8
0
0
0
1.3
0
0
1.4
04 Appendices D, D1 .xls/Replicate final Data base 11/7/20000
-------
Appendix E
Time Series Plots for All Species Measured in the 4-City Study
-------
PM2.5 Mass at Rubidoux, Jan. - Feb. 1999
120
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Sample Number
05 Appendix El TS-Mass & SO4.xls/Mass @ RUB1 (11/8/2000)
-------
PM2.5 Mass at Phoenix, Jan. - Feb. 1999
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
05 Appendix El TS-Mass & SO4.xls/Mass@Phoenix (11/8/2000)
-------
PM2.5 Mass at Philadelphia, Jan. - Feb. 1999
50
45
40
35
t 30
)
in
25
20
15
10
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
05 Appendix El TS-Mass & SO4.xls/Mass@Phil (11/8/2000)
-------
PM2.5 Mass at RTP, Jan. - Feb. 1999
1234567
8 9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
05 Appendix El TS-Mass & SO4.xls/Mass@RTP (11/8/2000)
-------
Sulfate at Rubidoux, Jan. - Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Sample Number
05 Appendix El TS-Mass & SO4.xls/SO4@RUBl (11/8/2000)
-------
Sulfate at Phoenix, Jan. - Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
05 Appendix El TS-Mass & SO4.xls/Sulfate@Phoenix (11/8/2000)
-------
Sulfate at Philadelphia, Jan. - Feb. 1999
CO
£
0)
CO
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
05 Appendix El TS-Mass & SO4.xls/Sulfate@Phil (11/8/2000)
-------
Sulfate at RTP, Jan. - Feb. 1999
CO
£
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
05 Appendix El TS-Mass & SO4.xls/Sulfate at RTP (11/8/2000)
-------
Nitrate at Rubidoux, Jan.-Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Sample Number
05 Appendix E2 TS-NO3 & NH4.xls/Nitrate@RUB (11/8/2000)
-------
Nitrate at Phoenix, Jan - Feb 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
05 Appendix E2 TS-NO3 & NH4.xls/Nitrate@Phoenix (11/8/2000)
-------
Nitrate at Philadelphia, Jan.-Feb. 1999
AND
FRM
--A-- IMP
MET
URG
VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
05 Appendix E2 TS-NO3 & NH4.xls/Nitrate@Phil (11/8/2000)
-------
3.5
2.5
1.5
0.5
0
Nitrate at Research Triangle Park, Jan.-Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
05 Appendix E2 TS-NO3 & NH4.xls/Nitrate@RTP (11/8/2000)
-------
Ammonium at Rubidoux, Jan.-Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
05 Appendix E2 TS-NO3 & NH4.xls/NH4@RUB (11/8/2000)
-------
Ammonium at Phoenix, Jan.-Feb. 1999
2.5
1.5
0.5
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
05 Appendix E2 TS-NO3 & NH4.xls/NH4@Phoenix (11/8/2000)
-------
Ammonium at Philadelphia, Jan.-Feb. 1999
o
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
05 Appendix E2 TS-NO3 & NH4.xls/NH4@Phil (11/8/2000)
-------
Ammonium at Research Triangle Park, Jan.-Feb. 1999
2.5
"TO
1.5
o
I
0.5
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sample Number
05 Appendix E2 TS-NO3 & NH4.xls/NH4@RTP (11/8/2000)
-------
OC at Rubidoux, Jan.-Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Sampling Day
-------
OC at Phoenix, Jan.-Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Sampling Day
-------
OC at Philadelphia, Jan.-Feb. 1999
12
10
8
CO
£
"B)
3
O
o
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Sampling Day
-------
OC at RTP, Jan.-Feb. 1999
12
10
8
CO
£
"B)
3
O
o
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Sampling Day
-------
EC at Rubidoux, Jan.-Feb. 1999
8
CO
£ 5
0)
o
LU
1
0
8
10 11 12 13 14 15
Sampling Day
-------
EC at Phoenix, Jan.-Feb. 1999
CO
£
o
LU
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Sampling Day
-------
7
CO
£ 4
"B)
LU
0
EC at Philadelphia, Jan.-Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Sampling Day
-------
EC at RTP, Jan.-Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Sampling Day
-------
2500
2000
1500
1000
500
0
Sulfur at Rubidoux, Jan. - Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Sample Number
-------
Si at Rubidoux, Jan. - Feb. 1999
3000
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Sample Number
-------
Ca at Rubidoux, Jan. - Feb. 1999
rc
O
3000
2500
2000
1500
1000
500
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Sample Number
-------
Fe at Rubidoux, Jan. - Feb. 1999
O)
C
o
1600
1400
1200
1000
800
600
400
200
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Sample Number
-------
Zn at Rubidoux, Jan. - Feb. 1999
300
250
,,200
£
o
c
N
150
100
50
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Sample Number
-------
K at Rubidoux, Jan. - Feb. 1999
500
450
£350
6 7 8 9 10 11 12 13 14
Sample Number
-------
Mn at Rubidoux, Jan. - Feb. 1999
6 7 8 9 10 11 12 13 14
Sample Number
-------
Cu at Rubidoux, Jan. - Feb. 1999
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Sample Number
-------
Pb at Rubidoux, Jan. - Feb. 1999
60
50
30
rc
0)
20
10
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Sample Number
-------
As at Rubidoux, Jan. - Feb. 1999
3.50
I 1.50
0.50
0.00
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Sample Number
-------
Sulfur at Phoenx, Jan. - Feb. 1999
1000
900
AND
FRM
IMP
MET
URG
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
Si at Phoenx, Jan. - Feb. 1999
4500
4000
-I-AND
-I-FRM
-A-IMP
-B-MET
-9-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
-------
Ca at Phoenx, Jan. - Feb. 1999
2000
-I-AND
-I-FRM
-A-IMP
-B-MET
-9-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
-------
Fe at Phoenx, Jan. - Feb. 1999
1800
1600
-I-AND
-I-FRM
-A-IMP
-B-MET
-9-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
-------
Zn at Phoenx, Jan. - Feb. 1999
60
50
40
30
20
10
0
-I-AND
-I-FRM
-A-IMP
-B-MET
-9-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
-------
K at Phoenx, Jan. - Feb. 1999
700
600
500
400
300
200
100
0
-I-AND
-I-FRM
-A-IMP
-B-MET
-9-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
-------
Mn at Phoenx, Jan. - Feb. 1999
35
30
25
20
15
10
0
-I-AND
-I-FRM
-A-IMP
-B-MET
-9-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
-------
Cu at Phoenx, Jan. - Feb. 1999
-I-AND
-I-FRM
-A-IMP
-B-MET
-9-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
-------
Pb at Phoenx, Jan. - Feb. 1999
70
60
50
40
30
20
10
0
x
-I-AND
-I-FRM
-A-IMP
-B-MET
-9-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
-------
As at Phoenx, Jan. - Feb. 1999
14.00
12.00
_ 10.00
CO
£
"B)
£ 8.00
o
'E
8> 6.00
<
4.00
2.00
0.00
x
AND
FRM
IMP
MET
URG
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
Sulfur at Philadelphia Jan. - Feb. 1999
3500
3000
AND
FRM
IMP
MET
-9-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
Si at Philadelphia Jan. - Feb. 1999
350
AND
FRM
IMP
MET
URG
VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
Ca at Philadelphia Jan. - Feb. 1999
350
AND
FRM
IMP
MET
URG
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
Fe at Philadelphia Jan. - Feb. 1999
600
500
-400
£
"B)
C
o
300
200
100
0
AND
FRM
IMP
MET
URG
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
Zn at Philadelphia Jan. - Feb. 1999
120
100
& 80
o
c
N
60
40
20
0
AND
FRM
IMP
MET
URG
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
K at Philadelphia Jan. - Feb. 1999
160
AND
FRM
IMP
MET
URG
6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
Mn at Philadelphia Jan. - Feb. 1999
CO
AND
FRM
IMP
MET
URG
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
Cu at Philadelphia Jan. - Feb. 1999
AND
FRM
IMP
MET
URG
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
Pb at Philadelphia Jan. - Feb. 1999
AND
FRM
IMP
MET
URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
As at Philadelphia Jan. - Feb. 1999
5.00
4.50
0.00
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
AND
FRM
IMP
MET
URG
-------
Sulfur at RTP Jan. - Feb. 1999
2500
AND
FRM
IMP
MET
URG
3 4
7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
SiatRTP Jan.-Feb. 1999
350
300
250
200
150
100
50
0
AND
FRM
IMP
MET
URG
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
-------
CaatRTP Jan.-Feb. 1999
200
-I-AND
-I-FRM
-A-IMP
-B-MET
-9-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
-------
FeatRTP Jan.-Feb. 1999
160
140
-I-AND
-I-FRM
-A-IMP
-B-MET
-9-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
-------
ZnatRTP Jan.-Feb. 1999
60
50
40
30
20
10
0
-I-AND
-I-FRM
-A-IMP
-B-MET
-9-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
-------
KatRTP Jan.-Feb. 1999
250
200
150
100
50
0
-I-AND
-I-FRM
-A-IMP
-B-MET
-9-URG
-X-VAPS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
-------
Mn at RTP Jan. - Feb. 1999
AND
FRM
IMP
MET
URG
8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
CuatRTP Jan.-Feb. 1999
AND
FRM
IMP
MET
URG
7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
PbatRTP Jan.-Feb. 1999
AND
FRM
IMP
MET
URG
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
As at RTP Jan. - Feb. 1999
3.50
co 2-50
0.50
0.00
AND
FRM
IMP
MET
URG
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number
-------
Appendix F
Regression Analysis Plots for All Species Measured in the 4-City Study
-------
FRM vs Speciation Samplers for Mass at Rubidoux
CO
£
rc
'o
0)
Q.
120
100
80
*> 60
Q.
E
^ 40
20
0
-20
I AND
A IMP
D MET
O URG
X PTEMP
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (PTEMP)
FRM (ug/m3)
-------
FRM vs Speciation Samplers for Mass at Phoenix
100
90
80
70
60
50
40
30
20
10
0
a--'
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
•Linear (IMP)
Linear (MET)
— Linear (URG)
Linear (VAPS)
0
10
15
20
25
30
-------
FRM vs Speciation Samplers for Mass at Philadelphia
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
0
10
15
20
25
30
35
40
-------
30
25
20
15
10
0
FRM vs Speciation Samplers for Mass at Research
Triangle Park
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
0
10
15
20
25
-------
FRM vs Speciation Sampler for Sulfate at Rubidoux
o
I AND
A IMP
D MET
O URG
X PTEMP
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (PTEMP)
0
-------
FRM vs Speciation Samples for Sulfate at Phoenix
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
0
0.5
1.5
-------
FRM vs Speciation Samplers for Sulfate in Philadelphia
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
0
8
-------
CO
E
5 4
"5.
E
ro
c
o
1
o
0)
Q.
0
0
FRM vs Speciation Samplers for Sulfate at Research
Triangle Park
i
A
O
X
AND
IMP
MET
URG
VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
3 4
FRM (ug/m3)
-------
FRM vs Speciation Samplers for Particluate Nitrate at
Rubidoux
CO
£
Q.
E
o
Q.
0)
(O
o
I AND
A IMP
D MET
O URG
X PTEMP
Linear (AND)
Linear (IMP)
- Linear (MET)
— Linear (URG)
Linear (PTEMP)
10 15 20 25
FRM (ug/m3)
30
35
40
-------
12
10
8
0
FRM vs Speciation Samples for Particulate Nitrate at
Phoenix
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
-Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
0
8
-------
FRM vs Speciation Samplers for Particulate Nitrate
at Philadelphia
i
A
O
X
AND
IMP
MET
URG
VAPS
Linear (/
•Linear (I
Linear (P
Linear (I
•Linear 0
n
-------
FRM vs Speciation Samplers for Particulate Nitrate at
RTF
0.5
I AND
A IMP
D MET
O URG
X VAPS
Linear (>
Linear (I
Linear ([
Linear (I
Linear
n
-------
FRM vs Speciation Samplers for Ammonium
at Rubidoux
I AND
A IMP
D MET
O URG
X PTEMP
Linear (AN
Linear (ME
Linear (UF
Linear (IMI
Linear (PT
-------
2.5
1.5
0.5
FRM vs Speciation Samplers for Particulate
Ammonium at Phoenix
I AND
A IMP
D MET
O URG
X VAPS
Linear (>
Linear (I
Linear ([
Linear (I
Linear 0
-------
6
FRM vs Speciation Samplers for Particulate
Ammonium at Philadelphia
I AND
A IMP
D MET
O URG
X VAPS
Linear (>
Linear (I
Linear (I
Linear (I
Linear
-------
^ 2.5
CO
£
(0
2
Q.
E
CO
c
o
.Is
'o
CD
Q.
(O
1.5
0.5
FRM vs Speciation Samplers for Particulate
Ammonium at RTP
n
I AND
A IMP
D MET
O URG
X VAPS
Linear (>
Linear (I
Linear (r
Linear (I
Linear 0
-------
FRM vs Speciation Samplers for OC at Rubidoux
01234567
FRM (ug/m3)
8
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
10 11
-------
FRM vs Speciation Samplers for OC at Phoenix
0 1
6789
FRM (ug/m3)
I AND
O URG
X VAPS
D MET
A IMP
Linear (AND)
Linear (URG)
Linear (VAPS)
Linear (MET)
Linear (IMP)
10 11 12 13 14
-------
FRM vs Speciation Samplers for OC at Philadelphia
o
I AND
A IMP
O URG
D MET
X VAPS
Linear (AND)
Linear (IMP)
Linear (URG)
- Linear (MET)
Linear (VAPS)
456
FRM (ug/m3)
8
10
-------
FRM vs Speciation Samplers for OC at RTF
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
0
8
FRM (ug/m3)
-------
FRM vs Speciation Samplers for EC at Rubidoux
8
(/> 5
"5.
I 4
§ 3
I 2
(/)
1
0
0
3456
FRM (ug/m3)
8
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
-------
FRM vs Speciation Samplers for EC at Rubidoux
8
(/> 5
"5.
I 4
§ 3
I 2
(/)
1
0
0
3456
FRM (ug/m3)
8
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
-------
FRM vs Speciation Samplers for EC at Phoenix
CO
£
"B)
3. 4
12
_o>
Q.
I 3
(/)
c
o
o
0)
Q.
(O
0
0
234
FRM (ug/m3)
I AND
A IMP
O URG
X VAPS
D MET
Linear (AND)
Linear (IMP)
Linear (URG)
Linear (VAPS)
Linear (MET)
-------
FRM vs Speciation Samplers for EC at Philadelphia
7
CO
_o> 4
Q.
E
CG
o 2
0)
Q.
(/)
0
0
3 4
FRM (ug/m3)
I AND
A IMP
O URG
X VAPS
D MET
Linear (AND)
Linear (IMP)
Linear (URG)
Linear (VAPS)
Linear (MET)
7
-------
FRM vs Speciation Samplers for EC at RTF
o
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
•Linear (IMP)
- Linear (MET)
— Linear (URG)
Linear (VAPS)
FRM (ug/m3)
-------
FRM vs Speciation Samplers for Sulfur at Rubidoux
2500
~ 2000
o> 1500
Q.
E
1000
o
0)
500
0
0
500
1000 1500
FRM (ng/m3)
2000
I AND
A IMP
D MET
O URG
Linear (AND)
-Linear (IMP)
- Linear (MET)
Linear (URG)
2500
-------
FRM vs Speciation Samplers for Si at Rubidoux
3000
2500
2000
1500
1000
500
0
-500
0
D
50
100
150
200
250
300
350
4(
I AND
A IMP
D MET
O URG
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
0
-------
FRM vs Speciation Samplers for Ca at Rubidoux
3000
2500
2000
1500
1000
500
0
I AND
A IMP
D MET
O URG
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
o
50
100 150 200 250 300 350 400
-------
FRM vs Speciation Samplers for Fe at Rubidoux
1600
1400
1200
1000
800
600
400
200
0
I AND
A IMP
D MET
O URG
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
0
100
200
300
400
500
-------
FRM vs Speciation Samplers for Zn at Rubidoux
350
300
250
200
150
100
50
0
I AND
A IMP
D MET
O URG
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
0
50
100
150
200
250
300
-------
FRM vs Speciation Samplers for K at Rubidoux
350
100
0
D
0
20
-B-
40
60
80
100
120
I AND
A IMP
D MET
O URG
Linear (AND)
Linear (IMP)
Linear (URG)
140
-------
35
30
25
20
15
10
0
FRM vs Speciation Samplers for Mn at Rubidoux
D
n
I AND
A IMP
D MET
O URG
Linear (AND)
-Linear (IMP)
- Linear (MET)
Linear (URG)
0
8
10
12
-------
FRM vs Speciation Samplers for Cu at Rubidoux
FRM = 30.7 ng/m3 on 1/18/99
as outlier on
I AND
A IMP
D MET
O URG
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
-------
FRM vs Speciation Samplers for Pb at Rubidoux
80
70
60
50
40
30
20
10
0
I AND
A IMP
D MET
O URG
Linear (AND)
-Linear (IMP)
Linear (MET)
Linear (URG)
0
10
20
30
40
50
60
-------
FRM vs Speciation Samplers for As at Rubidoux
3.50
LOO EJ
2.
[]
-S 2.00 EJ
_0>
Q.
E
rc
c
o
o
0)
Q.
(/)
.00
0.50
0.00
I AND
A IMP
D MET
O URG
Linear (AND)
-Linear (IMP)
- Linear (MET)
Linear (URG)
0.00 0.20 0.40 0.60 0.80
FRM (ng/m3)
1.00
1.20
1.40
1.60
-------
FRM vs Speciation Samplers for Sulfur at Phoenix
CO
£
"B)
Q.
E
rc
.5
'o
0)
Q.
(O
1000
900
I
A
O
X
AND
IMP
MET
URG
VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
0
100
200
300
400
500
600
700
FRM (ng/m3)
-------
3500
3000
2500
2000
1500
1000
500
0
FRM vs Speciation Samplers for Si at Phoenix
VAPS = (212, 3958 ng/m3) on
1/22/99 removed as an outlier in
this plot.
D D
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
0
100
200
300
400
500
-------
1400
1200
1000
800
600
400
200
0
FRM vs Speciation Samplers for Ca at Phoenix
VAPS = (89, 1747 ng/m3) on
1/22/99 removed as an outlier
from this plot.
0
50
100
o^o ^^^
150
n
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
•Linear (IMP)
Linear (MET)
— Linear (URG)
Linear (VAPS)
200
250
-------
1400
1200
1000
800
600
400
200
0
FRM vs Speciation Samplers for Fe at Phoenix
VAPS = (174, 1689 ng/m3) on
1/22/99 removed as an outlier
from this plot.
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
•Linear (IMP)
Linear (MET)
— Linear (URG)
Linear (VAPS)
0
100
200
300
400
500
-------
FRM vs Speciation Samplers for Zn at Phoenix
60
50
40
30
20
10
0
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
0
10
20
30
40
50
-------
700
600
500
400
300
200
100
0
FRM vs Speciation Samplers for K at Phoenix
VAPS = (82, 610ng/m3)on
1/22/99 removed as an outlier
from this plot.
D
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
•Linear (IMP)
- Linear (MET)
— Linear (URG)
Linear (VAPS)
0
50
100
150
200
250
300
-------
FRM vs Speciation Samplers for Mn at Phoenix
30
25
20
15
10
0
VAPS = (2.8, 32.3 ng/m3) on
1/22/99 removed as an outlier
from this plot.
-B-
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
0
-------
FRM vs Speciation Samplers for Mn at Phoenix
30
25
20
15
10
0
VAPS = (2.8, 32.3 ng/m3) on
1/22/99 removed as an outlier
from this plot.
-B-
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
0
-------
FRM vs Speciation Samplers for Cu at Phoenix
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
-Linear (IMP)
Linear (MET)
— Linear (URG)
Linear (VAPS)
0
10
15
20
25
30
35
40
-------
FRM vs Speciation Samplers for Pb at Phoenix
70
60
50
40
30
20
10
0
x
D
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
-Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
0
10
15
20
25
-------
FRM vs Speciation Samplers for As at Phoenix
CO
£
12
_fe
Q.
E
CG
0)
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
x
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
-Linear (IMP)
Linear (MET)
— Linear (URG)
Linear (VAPS)
0.00 0.50
1.00
2.50 3.00 3.50 4.00
FRM (ng/m3)
-------
FRM vs Speciation Samplers for Sulfur at Philadelphia
3500
_ 3000
CO
£
g 2500
^
-§_ 2000
|
W 1500
c
o
I 1000
0)
W 500
0
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
0
500
1000
1500
2000
2500
3000
FRM (ng/m3)
-------
FRM vs Speciation Samplers for Si at Philadelphia
350
300
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
0
20
40
60
80
100
120
140
-------
FRM vs Speciation Samplers for Ca at Philadelphia
350
300
250
200
150
100
50
0
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
0
20
40
60
80
100
120
-------
FRM vs Speciation Samplers for Fe at Philadelphia
600
500
400
300
200
100
0
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
0
50
100
150
200
250
300
350
-------
FRM vs Speciation Samplers for Zn at Philadelphia
450
400
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
-Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
100
-------
FRM vs Speciation Samplers for K at Philadelphia
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
0
20
40
60
80
100
120
140
-------
FRM vs Speciation Samplers for Mn at Philadelphia
12
10
8
0
-2
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
-------
20
15
10
0
FRM vs Speciation Samplers for Cu at Philadelphia
-5
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
-Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
-------
FRM vs Speciation Samplers for Pb at Philadelphia
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
0
10
20
30
40
50
-------
FRM vs Speciation Samplers for As at Philadelphia
Speciation Samplers (ng/m3)
u.uu
4 en
.ou
[
A nn -i
't.UU *
^ ^n ^
o.ou \
^ nn ^
o.uu
9 ^n L
z.ou
^
9 nn '
Z.UU
/
1 ^n i
\ .ou s
1 nn *
I .UU ,
4
n ^n v
U.OU z
n nn E
I
]
O
.-.-••"D
^ 0 .""""""
X
J L- ^ -----^
L _--r=-"n A A
^'. ..-•• A
k
I C3JH PH r\
I AND
A IMP
D MET
0 URG
X VAPS
Linear (AND)
Linear (IMP)
I inaar fMP~T\
Lineal (JVIt I )
Linear (URG)
— Linear (VAPS)
0.00 0.50
1.00
1.50 2.00
2.50
3.00 3.50
4.00
FRM (ng/m3)
-------
2500
CO
£
"B)
Q.
E
2000
1500
1000
.5
'o
0)
Q.
500
0
FRM vs Speciation Samplers for Sulfur at RTF
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
0
500
1000
1500
2000
2500
FRM (ng/m3)
-------
FRM vs Speciation Samplers for Si at RTF
350
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
•Linear (IMP)
Linear (MET)
— Linear (URG)
Linear (VAPS)
0
50
100
150
200
250
-------
200
180
FRM vs Speciation Samplers for Ca at RTF
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
Linear (MET)
Linear (URG)
Linear (VAPS)
0
20
40
60
80
100
120
140
160
-------
180
FRM vs Speciation Samplers for Fe at RTF
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
0
20
40
60
80
100
120
140
-------
FRM vs Speciation Samplers for Zn at RTF
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
-Linear (IMP)
Linear (MET)
— Linear (URG)
Linear (VAPS)
0
10
20
30
40
50
60
-------
250
200
150
100
50
0
FRM vs Speciation Samplers for K at RTF
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
0
50
100
150
200
-------
FRM vs Speciation Samplers for Mn at RTF
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
0
8
-------
FRM vs Speciation Samplers for Cu at RTF
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
-Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
-------
FRM vs Speciation Samplers for Pb at RTF
20
15
10
0
-5
I AND
A IMP
D MET
O URG
X VAPS
Linear (AND)
-Linear (IMP)
- Linear (MET)
Linear (URG)
Linear (VAPS)
-------
FRM vs Speciation Samplers for As at RTF
^— »
CO
£
B)
^
12
0
Q.
E
rc
(/)
c
'^
<5
o
0)
A
Q.
(/)
o.uu
^
o nn
O.UU n
9 ^n
z.ou
s
/
9 nn -
Z.UU H
1 ^n
I .OU
1 nn
I .UU
.AAi- /ii\fln\
-Linear (IMP)
1 in^or /MP~T\
Linear ^ivit i )
Linear (URG)
1 ina^r A/APQN
Linear ^vAr o^
FRM (ng/m3)
-------
APPENDIX G-l
Absolute Differences Between the FRM,
(Reference Sampler), and the Speciation Samplers
by Site and by Sampling Period
(Manganese, Copper, Lead and Arsenic Excluded Because of Too Few Samples)
07 Appendices G, Gl.xls/Appendix Gl - Abs Dif Final (11/9/2000)
-------
DIFFERENCES FROM FRM FOR PM-2.5 (micrograms/m3)
SITE
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
DATE
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
ERMMass
ERM-AND
FRM-IMP
ERM-MET
FRM-URG
FRM-VAPS
28.57
74.28
14.88
29.36
2.96
29.46
5.58
21.02
2.21
35.62
59.01
29.13
17.56
23.6
22.04
25.33
24.38
12.71
9.04
19.83
7.58
4.12
17.29
11.46
23.33
13.79
3.92
12.42
12.75
20.29
13.42
1.78
-10.85
1.59
-0.73
0.29
7.45
0.15
3.01
-0.03
1.42
0
-2.08
0.2
1.05
0.16
-0.22
-0.19
-0.22
-0.1
-0.74
-0.81
-0.28
0.24
0.48
1.07
0.24
-0.23
0.12
0.16
0.08
-0.09
2.31
-13.25
1.64
2.39
0.2
13.06
-0.09
1.33
-0.43
1.42
0.35
-3.69
1.53
1.04
6.38
4.12
1.78
2.05
0.18
5.17
0.68
0
1.77
1.39
5.17
1.8
-0.02
-0.82
0.53
2.87
1.07
-37.58
-18.76
1.55
-4.07
-0.08
-3.95
-8.65
1.7
-7.25
-5.97
-6.71
-9.72
-14.42
-1.63
-15.96
-21.46
-19.29
-13.99
-11.53
-18.16
-2.22
-16.95
-0.63
-5.33
-7.29
-2.43
-12.79
-13.25
-15.91
-15.49
-4.37
-30.82
-1.89
-9.32
-0.13
0.46
-0.33
-17.84
-21.53
-10.62
-4.78
-11.45
3.08
0.79
0.88
1.13
0.29
-0.54
0.16
1.24
-1.21
0.54
1.07
-0.53
-0.09
-17.65
-28.18
-1.95
-20.26
-7
-5.21
-6.87
-39.18
-25.08
-18.85
-39.84
-1.87
-0.08
-1.84
-2.64
0.05
-0.19
-8.93
0.16
0.15
0.57
-1.17
25.76
0.43
-0.91
0.77
0.73
-7.45
19.35
37.58
15.03
12.36
13.77
25.84
16.52
11.86
15.56
5.16
32.17
24.85
10.45
5
10.78
14.36
4.38
23.75
6.54
8.25
15.25
0.03
-0.93
0.05
0.58
-0.08
0.05
-0.39
0.02
0.04
0.11
-0.36
-0.08
-0.52
-0.25
-0.06
-0.42
0.14
0.67
0.3
-0.27
-0.46
0.57
-0.98
0.31
0.7
0.2
0.14
-0.04
0.45
-0.2
1.84
-0.25
-0.02
-0.72
-0.06
-0.05
-0.64
0.77
0.83
-0.22
-1.09
-3.15
-0.42
1.45
0.48
0.75
0.13
-0.92
0.01
-0.38
-6.38
0.46
-1.86
-1.5
-1.55
-0.57
-0.66
0.55
-1.32
-0.64
-1.84
-3.33
-0.39
0.06
-0.27
-0.52
0.01
0.14
0.26
-0.01
0.55
0.44
-0.89
-0.86
-0.89
-0.66
0.17
-0.88
0.37
-0.1
-0.88
-9.83
-1.52
-1.18
-2.4
-8.77
-2.05
-3.05
1.65
-9.35
-0.54
-2.77
-2.63
-0.19
-1.69
-1.69
-1.15
8.71
-2.01
1.53
0.41
-0.38
-1.8
10.42
6.58
11.5
12.83
6.5
20.54
9.75
13.87
5.21
12
0.98
0.51
0.46
0.48
0.8
0.86
0.21
-0.06
0.21
0.1
1.29
0.39
0.33
0.27
0.28
0.18
-0.1
-0.34
0.19
0.62
-0.53
-0.17
-1.4
-0.73
-0.21
-3.18
-1.48
-0.42
-0.28
0.54
-0.59
-0.47
-0.71
-0.76
-1.22
-1.31
-0.33
-1.1
-2.12
-3.45
-3.91
-2.22
-0.21
-3.13
07 Appendices G, Gl.xls/Appendix Gl - Abs Dif Final (11/9/2000)
-------
DIFFERENCES FROM FRM FOR SULFATE BY 1C (micrograms/m3)
SITE
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
DATE
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
FRM MEAN
FRM-AND
FRM-IMP
FRM-MET
FRM-URG
FRM-VAPS
0.57
5.98
1.62
1.89
0.08
-0.08
-0.02
0.14
0.12
0.56
-0.02
0.2
-0.4
-0.04
-0.04
0.11
0.04
-0.66
-0.11
-0.07
-0.34
-0.73
-0.3
-0.14
0.59
0.34
1.16
0.39
1.79
3.69
1.61
0.43
1.98
0.75
1.02
1.78
1.16
0.24
0.63
0.33
0.47
0.7
0.44
0.71
1.19
0.28
0.7
1.13
1.57
1.15
1.2
3.96
2.62
6.66
8.42
4.42
2.03
3.03
5.68
4.01
2.68
4.14
1.54
6.54
7.33
3.74
1.46
2.01
2.74
0.71
5.24
1.7
3.51
5.38
6.2
3.32
2.84
2.03
2.57
2.25
5.51
2.63
2.24
2.75
5.86
1.97
2.88
0.06
0.08
0.42
0.05
0.1
-0.09
-0.62
-0.02
-0.08
0.05
0.08
0.12
0.11
0.05
0.08
0.04
0.08
0.14
0.08
0.11
0.15
0.08
0.1
0.15
0.42
0.12
0.08
-0.08
1.32
0
0.21
0.44
0.37
0.11
0.3
-0.09
0.26
-0.01
0.31
0.28
0.25
0.1
0.37
0.33
0.12
0.06
-0.06
-0.1
0.1
-0.16
0.09
0.18
0.09
0.21
0.1
-0.07
-0.1
0.16
0.23
-0.05
-0.31
0.1
0.1
0.06
0.07
0.47
0.06
0.19
0.12
-0.39
0.01
0.1
0.05
0.11
0.14
0.1
0.06
0.09
0.04
0.09
0.12
0.07
0.1
0.1
0.07
0.09
0.13
0.44
0.12
0.07
-0.26
-0.21
-0.06
0.34
0.4
0.33
0.08
0.28
-0.12
0.21
-0.02
0.32
0.47
0.45
0.16
0.39
0.36
0.28
0.03
-0.06
0.01
0.06
0.11
0.23
0.02
-0.12
0.19
0.19
0.33
0.08
-0.2
0.35
0.1
-0.07
-0.06
0.41
-0.04
0.06
0.05
-0.5
-0.38
0.06
-0.24
-0.17
-0.11
-0.11
-0.13
-0.15
-0.11
-0.11
-0.02
-0.03
-0.08
-0.04
-0.17
-0.1
0.1
-0.11
-0.32
0.09
-0.12
0.07
0.43
0.27
0.01
0.33
-0.17
0.18
-0.26
0.26
0.07
0.31
0.15
0.31
0.22
-0.01
0.1
-0.07
0.01
-0.3
-0.01
0.07
0.06
0.2
0.07
-0.16
-0.11
0.1
0.77
-0.24
0.06
0.05
-0.06
-0.05
0.42
0
-0.06
-0.06
-0.33
-0.03
-0.06
0.01
0.04
-0.06
0.06
0
0.03
-0.02
0.04
0.04
0.02
0
-0.06
-0.05
-0.03
0.05
0.33
0
-0.12
-0.13
0.34
-0.15
0.06
0.24
0.28
0.28
-0.19
0.08
0.11
0.13
0.19
-0.02
0.52
-0.11
0.02
0.04
0.2
0.05
0.27
-0.06
-0.17
-0.25
-0.19
0.11
0.19
0.16
0.07
-0.18
-0.02
-0.06
-0.18
-0.11
-0.23
-0.29
0.32
-0.07
-0.07
0.36
-0.04
-0.02
-0.05
-0.65
-0.28
-0.14
0.06
0.02
0.05
0.03
0.05
0.04
0.04
0.03
0.09
0.02
0
0.1
0.48
0.08
0.01
0.33
0.29
0.96
0.94
0.65
0.46
0.77
0.63
0.42
0.61
0.34
0.7
0.86
0.43
0.25
0.27
0.15
-0.02
0.38
-0.12
0.18
-0.35
-0.1
0.01
0.14
-0.01
0.14
0.51
0.15
-0.34
0.24
-0.15
07 Appendices G, Gl.xls/Appendix Gl - Abs Dif Final (11/9/2000)
-------
DIFFERENCES FROM FRM FOR PARTICLE NITRATE (micrograms/m3)
SITE
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
DATE
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
FRM MEAN
FRM-AND
FRM-IMP
FRM-MET
FRM-URG
FRM-VAPS
9.64
38.03
3.06
12.13
1.36
3.38
-0.16
1.21
1.88
3.45
-0.45
0.51
-0.55
-2.39
-0.76
-0.34
0.11
-3.75
-3.46
-0.46
-4.31
-1.1
-1.18
9.69
0.44
9.52
0.13
16.71
25.83
12.22
6.2
9.7
3.46
5.33
2.55
3.71
0.45
1.37
1.99
0.44
4.42
2.23
7.43
4.24
0.2
0.29
2.36
6.81
5.47
2.42
5.46
1.63
4.62
8.64
3.39
3.46
2.71
6.53
3.14
2.18
2.62
0.92
6.14
3.86
1.59
1.07
3.84
6.09
0.13
1
0.25
0.23
1.01
1.07
1.47
0.67
0.25
0.17
0.29
0.11
0.34
1.75
0.73
1.14
0.4
2.25
1.71
0.05
2.57
0.03
2.47
1.69
-1.82
1.06
0.56
-0.72
0.27
-0.95
0.1
0.05
0.08
0.06
0.03
0.34
0.09
0.41
0.06
0.04
0.01
0
0.37
0.08
-0.09
0.15
-2.36
0.3
0.69
0.49
0.66
0.07
0.63
0.11
0.21
0.02
0.27
-0.18
0.36
0.23
0.54
0.99
0.96
-0.01
-0.1
-0.19
-0.04
-0.32
0
-0.07
0.01
-0.19
0.01
-0.28
-0.24
-0.18
0.15
-0.03
-0.53
-0.13
-0.05
1.32
0.03
1.64
0.05
1.32
-0.02
-2.61
0.17
0.02
0.81
0.91
-0.19
0.81
0.04
0.12
0.08
0.07
0.9
0.13
1.15
0.73
0.06
0.04
0.53
2.38
1.96
0.43
-0.45
-0.28
-0.16
0.34
0.27
0.4
-0.08
0.23
-0.19
0.11
-0.06
0.24
0.63
0.12
0.17
0.53
0.81
0.77
0.01
-0.08
-0.02
-0.15
-0.02
-0.15
0.03
-0.3
-0.22
-0.21
0.47
0.01
-0.45
-0.18
-0.15
-0.02
-0.05
1.2
-0.04
-0.23
-1.46
-4.05
-0.32
-0.8
-1.17
-0.42
-1.42
-0.1
-0.22
0.44
-0.14
-0.01
0.01
0.17
0.01
-0.08
-0.21
-0.03
-0.35
-0.4
-0.23
-0.08
-0.05
0.07
0.21
0.32
-0.06
0.19
-0.22
0.03
-0.11
0.19
-0.77
1.52
0.08
0.46
0.49
-0.11
0.28
-0.24
-0.11
-0.56
-0.29
-0.22
-0.09
-0.26
-0.02
-0.42
-0.28
-0.3
1.47
-0.78
-0.24
-0.51
-0.02
-1.15
-0.88
-5.69
-0.83
-2.54
-2.01
-1.91
-1.56
-0.49
-1.1
-0.23
-1.53
-0.84
-0.04
-0.23
-1.42
-2.49
-3.26
-1.92
-0.27
-0.31
-0.41
-0.2
-0.28
0.34
0.13
-0.44
-0.09
-0.43
-0.45
0.18
-2.14
0.11
-0.25
0.34
0.49
0.07
-0.04
-0.58
-0.06
-0.16
-1.06
-0.55
-0.83
-0.28
-0.45
-0.04
-0.32
-0.39
-1.38
-0.28
-1.35
-0.36
-0.83
-0.18
0.94
-0.03
-0.72
-1.21
-8.03
-0.61
-1.43
-1.74
-0.08
-0.38
-0.58
-0.18
-1.54
-2.03
-1.68
-0.35
0
-1.49
-1.96
-1.87
-0.85
-1.6
0.57
0.39
1.06
-0.59
-0.33
-0.08
0.19
-0.16
0.19
-1.79
-0.14
0.45
0.31
-0.31
-0.07
-0.34
-0.17
-0.17
-0.69
-1.21
-0.35
-0.55
-0.3
-1.07
-2.11
-0.46
-0.43
07 Appendices G, Gl.xls/Appendix Gl - Abs Dif Final (11/9/2000)
-------
DIFFERENCES FROM FRM FOR AMMONIUM (micrograms/m3)
SITE
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
DATE
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
FRM MEAN
FRM-AND
FRM-IMP
FRM-MET
FRM-URG
FRM-VAPS
3
13.2
1.49
4.56
0.55
0
0.36
0.56
1.15
2.6
0.57
1.13
0.35
-0.28
0.18
0.13
0.04
-1.14
-0.09
-0.3
1.71
-0.11
0.03
0.21
3.04
0.26
2.76
0.14
5.41
9.34
4.24
1.81
3.21
1.19
1.87
1.31
1.5
0.2
0.51
0.7
0.23
1.55
0.86
2.52
1.69
0.13
0.26
1.12
2.44
1.93
1.13
3.1
1.38
3.66
5.78
2.59
1.8
2.05
4.03
2.43
1.72
2.38
0.78
4.02
3.86
1.83
0.89
1.97
2.82
0.23
1.88
0.42
1.21
1.99
2.26
1.45
1.15
0.64
0.85
0.86
1.5
0.91
1.26
1.13
2.3
0.77
1.78
0.59
0.15
0.32
0.02
0.51
0.34
0.06
0.12
0.33
0.21
0.5
0.29
0.71
0.14
0.27
0.37
0.07
0.66
0.39
1.04
0.95
0.08
0.05
0.55
1.02
1.16
0.61
-0.12
0.2
-0.1
0.25
0.24
0.24
0.09
0.21
-0.12
0.22
0.29
0.19
0.05
0.18
0.11
0.34
0.5
0.3
-0.01
0.06
-0.07
-0.09
-0.04
-0.05
0.11
0.07
-0.06
-0.04
-0.1
-0.12
0.01
0.33
-0.03
-0.08
-0.02
0.28
0.94
0.2
0.57
0.04
1.17
1.51
-0.11
0.33
0.52
0.83
0.83
0.48
0.81
0.14
0.26
0.31
0.08
0.85
0.46
0.96
0.74
0.07
0.06
0.63
1.45
1.28
0.61
0.03
0.3
0.13
0.78
0.45
0.4
0.42
0.47
0.11
0.36
0.42
0.27
0.98
0.52
0.31
0.4
0.68
0.74
0.11
0.29
0.1
0.22
0.2
0.02
-0.04
-0.02
0.06
0.07
0.46
0.15
0
0.02
0.44
0.1
0.18
0.1
0.06
0.1
0.07
-0.45
0.23
-0.18
0.1
0.21
0
0.53
0.1
0.19
0.32
0.32
0.33
0.58
0.7
0.05
0.01
0.45
0.88
0.71
-0.46
0.04
-0.43
0.14
0.06
0.05
-0.08
0.07
-0.32
0.14
0.27
0.12
-0.22
0.06
-0.06
0.22
0.38
0
-0.01
-0.11
-0.07
-0.15
-0.16
0.19
0.16
0.06
-0.01
-0.03
0.03
0.01
0.46
0.01
0.05
0.37
-0.41
-0.11
-0.23
-0.03
-0.57
-0.08
-1.08
-0.28
-0.64
-0.06
0.16
0.17
0.14
0.18
0.14
0.06
0.03
0.37
0.07
0.28
0.88
-0.07
-0.01
0.04
0.2
-0.08
-0.05
-0.3
0.08
-0.37
-0.05
-0.02
0.23
0.22
-0.23
-0.01
0.04
0.03
0.15
-0.47
0.22
0
0.2
0.31
0.36
-0.05
-0.05
-0.03
-0.29
-0.38
-0.33
-0.11
-0.08
-0.18
-0.09
-0.14
-0.25
-0.11
-0.26
-0.27
-0.24
-0.21
0.08
0.2
0.11
0.21
0.08
0.62
0.95
-0.67
0.24
-0.27
-0.18
0.03
0.01
0
-0.02
-0.04
0
0.1
0.84
0.01
-0.01
0.01
0.3
0.02
0.08
0.47
0.14
0.87
0.5
0.63
0.5
0.58
0.45
0.37
0.48
0.21
0.18
0.44
0.37
0.36
0.55
0.37
-0.04
0.08
-0.04
-0.08
-0.52
-0.19
-0.21
-0.17
-0.21
-0.42
0.01
-0.02
-0.55
-0.05
-0.01
07 Appendices G, Gl.xls/Appendix Gl - Abs Dif Final (11/9/2000)
-------
DIFFERNCES FROM FRM FOR ORGANIC CARBON (micrograms/m3)
SITE
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
DATE
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
FRM MEAN
FRM-AND
FRM-IMP
FRM-MET
FRM-URG
FRM-VAPS
7.72
10.01
2.98
6.89
-2.32
-4.2
-1.3
-1.24
3.56
4.75
1.23
3.92
-1.65
-0.41
-0.45
-0.71
0.91
-1.48
-0.1
1.17
-7 27
-4.49
-0.59
-0.76
7.78
4.58
4.65
2.05
5.53
5.69
6.93
5.5
4.19
9.48
11.79
10.14
5.6
6.23
12.51
4.34
3.96
8.98
5.6
9.67
7.69
3.99
7.29
7.34
8.34
6.45
7.75
5.17
2.31
2.8
9.57
4.05
4.13
3.1
5.32
3.4
4.46
4.6
2.31
8.54
4.41
2.78
2 77
3.66
3.3
2 79
7.27
2.49
2.4
3.78
4.66
2.39
2
3.75
2.21
0.07
-0.21
-1.24
0.09
-1.44
-2.65
-1.33
-1.17
1.19
-2.55
-2.84
-1.99
_2
-0.64
-2.2
-1.44
-0.77
-1.39
-1.63
-1.11
-1.52
-1.55
-1.21
-1.19
-1.91
-1.8
-1.07
-1.19
-0.82
-0.75
0.43
-1.63
-0.47
-3.01
-1.44
-0.64
-0.33
-1.96
-1.31
-2.47
-2.41
-0.32
-0.95
-1.03
-0.65
0.12
-0.16
0.17
-0.05
-0.22
-0.93
-0.72
-0.96
-0.1
0.51
4.5
2.02
0.67
0.47
-0.33
-0.76
1.02
0.92
0.36
2.76
2.37
-0.07
1.56
0.78
2.4
0.38
0.71
0.88
1.15
2.06
1.11
0.48
1.5
1.25
0.84
0.81
1.63
-0.65
0.38
0.51
1.74
0.23
0.18
0.36
0.78
0.16
0.72
0.89
0.39
0.71
-0.3
0.39
-0.21
-0.05
0.2
0.78
1.28
0.7
0.57
0.69
0.61
0.28
0.92
0.92
0
0.79
0.36
-0.3
-2.04
0.23
-0.89
-1.88
-0.45
7.84
-0.92
-3.18
0.09
0.81
0.4
-0.51
-1.79
-0.12
-0.68
-1.61
-0.53
-3.12
-0.3
-1.25
5.04
-0.74
2.53
-0.23
-0.06
1.57
0.07
0.18
-0.14
0.88
-0.35
0.32
0.37
0.29
-1.6
0.02
2.39
0.38
0.23
0.14
0.07
-0.31
-0.43
-0.05
-0.67
-0.86
-0.26
-0.23
1.39
1.34
0.82
0.39
-0.33
-0.71
0.61
0.33
0.25
1.84
3.46
1.51
1.55
2.02
3.32
0.65
1.8
-0.25
-0.01
-0.05
2.78
0.27
1.65
0.39
1.13
-0.11
1.37
0.45
0.34
0.7
2
0.57
0.49
0.17
0.84
0.33
1.1
0.96
0.78
0.25
0.21
0.62
0.57
0.6
0.09
1.07
1.05
1.6
0.66
-0.44
0.62
0.33
0.05
0.79
0.53
2.07
1.03
0.89
-0.15
-1.01
0.7
0.68
-1.26
1.82
2.95
4.67
1.4
2.24
3.47
1.56
2.57
3.15
2.18
2.61
3.26
4.13
3.49
3.77
1.05
0.32
3.51
1.71
2.25
0.84
1.48
0.95
2.04
2.14
1
2.89
1.11
1.16
1.53
1.59
0.97
1.33
1.71
0.47
0.76
-2.19
0.53
0.2
0.69
3.32
1.81
8.51
3.37
3.11
1.46
3.24
-0.64
-0.73
-1.56
0.06
0.35
-0.56
-0.51
0.85
0.19
1
1.12
0.8
-0.12
0.55
-0.83
-0.46
-1.82
0.05
-0.31
-0.38
0.63
0.3
0.56
0.65
1.02
-0.01
0.37
0.33
0.64
1.17
0.86
0.27
07 Appendices G, Gl.xls/Appendix Gl - Abs Dif Final (11/9/2000)
-------
DIFFERENCES FROM FRM FOR ELEMENTAL CARBON (micrograms/m3)
SITE
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
DATE
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
FRM MEAN
FRM-AND
FRM-IMP
FRM-MET
FRM-URG
FRM-VAPS
5.22
7.7
1.61
3.22
0.45
2.01
-0.03
0.17
3.73
5.85
1.24
2.49
0.27
0.47
0.05
0.46
1.05
0.97
0.11
0.44
-0.26
1.98
0.06
0.27
4.44
1.59
2.5
0.56
2.86
5.43
3.08
2.64
1.9
4.56
4.82
5.71
2.29
2.37
5.03
1.86
1.53
4.56
3.26
4.28
3.28
1.47
2.7
2 99
3.98
2.37
2.64
3.54
0.82
1.64
5.21
2.07
1.73
1.91
3.35
2.45
2.42
2.71
1.16
5.73
3.43
1.63
1.71
1.91
1.57
1.41
3.7
0.76
1.16
2.51
2.84
0.9
0.96
1.6
0.78
0.84
0.14
0.42
0.35
0.57
0.89
-0.43
-0.67
0.94
-0.16
-0.34
0.28
0.31
0.16
0.94
0.48
0.27
1.58
0.94
1.08
0.87
0.48
0.77
0.26
0.21
-0.23
-0.01
1.95
-4.53
-0.49
3.66
-0.06
0.2
-0.34
-0.15
0.51
0.51
0.36
0.05
0.45
0.15
0.18
0.25
0.4
-0.02
0.54
0.74
0.38
0.28
0.43
0.43
0.22
0.06
-0.23
-0.05
3.31
0.74
0.05
0.1
0.39
1.19
0.22
0.26
0.22
0.64
0.05
1.45
0.8
0.41
1.25
0.45
0.41
1.02
0.6
0.47
0.35
0.06
0.44
0.5
0.65
0.4
0.19
1.22
0.05
0.21
0.5
0.06
-0.09
0.12
0.4
0.35
0.45
0.63
0.05
1.28
0.05
-0.07
-0.12
0.31
0.15
0.45
1.49
0.4
0.28
0.6
0.26
0.2
0.18
0.13
1.45
0.13
-0.17
0.05
-0.13
1.4
-0.3
-0.37
0.26
4.46
0.32
0.78
0.26
0.26
1.13
0.32
0.24
0.3
-0.08
-0.05
-0.2
-0.14
0.23
-0.16
2.25
-0.21
2.07
0.03
-0.03
0.28
-0.06
-0.07
0.11
0.22
0.69
0.36
0.54
0.25
-0.52
0.66
1.63
0.34
0.23
0.29
0.58
0.24
0.15
0.31
0.14
-0.03
0
0.05
0.67
0.28
0.06
0.02
-0.42
0.46
-0.9
-0.73
-0.13
0.3
0.35
0.34
0.24
-0.24
0.14
-0.06
0.23
0.6
-0.1
0.05
0.34
-0.06
0.31
0.12
0.37
0.02
0.18
-0.06
0.05
-0.04
0.65
-0.02
-0.07
-0.29
-0.12
0.38
0.27
0.63
0.18
0.25
0.09
0.22
0.32
0.45
0.07
0.21
0.34
0.39
-0.04
0.29
0.14
0.09
0.03
-0.44
-0.14
0.2
-0.13
0.17
0.31
0.9
-0.11
-0.46
-0.02
0.52
0.35
0.71
0.2
0.33
1.07
0.5
0.37
0.6
0.34
0.47
0.58
0.76
0.42
0.38
0.69
0.18
0.48
0.35
0.8
0.31
0.55
0.95
0.61
0.67
-0.06
0.4
0.34
0.1
0.11
0.57
0.54
0.32
0.12
0.38
0.22
-1.68
0.22
0.15
-0.31
0.82
0.55
3.06
1.2
1.06
0.65
1.43
-0.17
-0.02
0.4
0.04
-0.15
0.03
0.18
0.02
-0.01
0.66
0.2
-0.02
-0.11
0.22
0.07
0.13
0.35
-0.07
-0.1
0.26
-0.14
-0.04
0.12
0
-0.19
-0.18
0.41
0.16
0.14
0.69
-0.01
-0.1
07 Appendices G, Gl.xls/Appendix Gl - Abs Dif Final (11/9/2000)
-------
DIFFERENCES FROM FRM FOR SULFUR BY XRF (micrograms/m3)
SITE
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
DATE
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
FRM MEAN
FRM-AND
FRM-IMP
FRM-MET
FRM-URG
FRM-VAPS
204.7
1928.3
605.9
-3.1
-306.4
-17.7
5.8
-190.8
35.9
-125.8
-359.6
-58.1
-0.3
-381.8
-66.4
242.3
111
310.6
120.6
644.8
1391.7
645.7
174.7
787.5
299.6
365.5
642.8
385.5
71.2
237.9
122.3
161.9
226.6
176.7
285.2
402.3
85.3
237.1
407
479.5
434.9
32
6.7
-4.2
1.6
8.7
-81.4
-156.3
36.3
-1.6
8.9
11.4
-17.5
-13.1
15.9
-6.5
13.7
-10.1
11
17.2
1.6
5
0
4.8
20.5
35.7
95.6
22.4
4.3
7.8
40.2
32.5
-113.8
32.2
69.3
-28.5
64.2
-237.6
23.8
_2
78.6
-1.7
18.8
-7.2
7.2
17.3
38.4
13.1
7.7
29.7
54.1
40.9
-49.2
-40.9
-42.7
-7.3
-59.5
-46.8
-190.4
-111.3
-19.1
-59.4
-88.8
-141.2
-45.7
-55.7
-58.2
-52.3
-86.2
27.9
-29.8
-44.8
-6.8
-46.8
-53.9
32
-55.2
-10.8
-19.8
-65.3
-171.8
-10.6
1.8
7.7
-10.7
1.5
17.2
-4.1
18.9
-11.6
19
8
15.5
16.7
23
-26.8
-18.8
-33.6
-25.3
-5.7
-216.5
12.4
-51.6
63.7
-79.9
19.5
1.8
24.1
-0.4
20.5
-21.8
8.6
61.4
43.7
1322.8
-218.2
-177.6
-342.2
-157.5
-244.6
2232.9
2794
1429.6
640.2
1033.4
2026.7
1480.9
900.1
1474.4
440.7
2202.8
2411.2
1289.9
394.2
589.7
891.5
261.7
1820.8
611.7
1231
2004.1
-130.2
-158.9
-24.1
12.3
-39
84.5
19.5
24.1
20.3
7.1
49.1
-55.9
20.2
-1.1
11.8
-49.5
2.5
-63.1
-9.9
-26.6
57.9
109.3
38.4
103.4
23.1
74.8
274.5
868.6
63
34.2
208.3
185.1
-2 9
33.2
21.3
20.9
78.6
41.8
83.1
-290.4
-344.3
-34
-33.5
-80
-21.9
-155.2
-60.7
-83.5
-19.8
-84.3
-108.2
-49.3
-56.6
-95.4
-27
-48.8
-20.3
-132.8
-82
-270.1
-445.4
-90.7
-53.4
-68.2
-44.2
-27.9
-27.3
-17.5
0.9
21.1
-71.2
-47.2
-38
-73.6
-77.8
-18.1
-128.4
-42.8
-57.9
-90.3
-239.9
14
-40.6
2.2
26.9
-137.9
33.3
106
33
32.9
-23.1
8.1
-78.4
-12.7
-101.4
1171.4
-12.2
199.4
-16.7
-104.3
815.5
692.4
867.2
861.3
1953.4
888.5
712.1
1007.4
2083.7
677.9
1046.2
54.5
-22.4
-3.9
125.4
55.7
23.5
8.7
-45.8
21.3
34.4
129
-53.1
52.4
112.6
33.7
19.5
-9.7
-96.3
58.6
114.8
-24.7
-37.6
-8.2
-148.5
0.5
3
4.7
-36.9
-75.3
-62.6
-53.7
-31
-65.7
-78.3
-62.6
-183.7
-86.3
-19
2.6
-16
-19.9
-220.6
-7.7
-84.3
07 Appendices G, Gl.xls/Appendix Gl - Abs Dif Final (11/9/2000)
-------
DIFFERENCES FROM FRM FOR SILICON (nanograms/m3)
SITE
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
DATE
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
FRM MEAN
FRM-AND
FRM IMP
FRM MET
FRM URG
FRM VAPS
377.8
216.5
16.2
-70.7
50.1
-50.7
-231.6
10.8
-8
-2031.5
-292.9
-113.3
39.1
100.9
-325
150.1
145.7
43
101.1
121.4
235.9
118.7
280.1
71.7
467.1
475.4
403.3
416.5
211.5
444.2
133.1
75.1
333.6
55.7
94
172
49.9
395.3
312.2
374.7
320.4
10.3
-20.5
-22.2
-78.2
-52.1
-95.6
-51.2
-41.3
-1.5
-25.3
-47.1
-55.2
-62.2
-30
-46.1
-13.3
-116.4
-4
-1.4
-53.4
-22.1
-79.7
-77.9
-100.3
-125.1
33.7
-20.5
5.9
-48.7
-49.8
-47.8
-16
-60.8
-9
-92.6
3.3
-228.5
10.8
-45.1
91.5
-28.6
-39.7
-130.4
-71.4
-8.5
-41
-43.8
-56.4
-98.7
-54.8
-166.5
-544
-769.4
-33.5
-691.1
-550.1
-406.5
-494
-1346.9
-62.1
-1284.4
-2442.4
-2169.5
-819.8
-1197.6
-1304.3
-537.7
-2013.8
-48.1
-484.3
-358.2
-321.3
-1241
-1028.6
-1090.8
-1252.4
-9.1
-4.1
-8.3
12.8
-11.4
126.5
98.8
108.3
115.6
130.4
42.8
32.3
18.2
13.6
2.7
45.1
80.3
20.4
-875.1
-742.2
-147.9
-887.2
-672.7
-797.2
-1990
-513.3
-88.1
-3746.9
-140.4
-62.6
-37.7
-146.7
-63.3
-93.6
-19.8
-77.4
-92.8
51.8
-15.2
-30.8
-40.9
-8.9
0.4
48.5
96.5
40.5
31.2
25
59.1
40.3
23.7
19.3
25.9
132.8
24.7
22.1
31.8
61.3
71.6
70.5
194.9
22.6
34.4
38.4
14.4
12.6
-19.6
-3.9
-20.7
-30.3
11.4
-8
-20.6
16.4
6.5
-39.4
-25.4
-3.9
-10
-3.9
1.9
-13
-18.7
4
-28.2
1
25.2
5.1
-13.8
-19.5
-4.7
-30.9
-17.5
12
-19.9
-39
-17.6
16.5
-9.8
4
-10.3
-15.7
-18.5
-28.8
-95.8
-42.3
-32.4
25
-115.5
-21.7
-51.6
-129 2
-12.2
-185.2
-41.1
-44.5
-203.3
-136.7
-27.3
-31.5
22.6
-65.6
-34.1
9.9
11.9
-3.3
-9.6
-22.6
-31.2
36.7
-10.3
-12.7
25.9
2.1
-25.1
-8.8
6.1
15.7
-14.4
-5.1
9.9
-6.1
5.3
-28
-15
-6.1
15.6
-32.2
-11.2
6
-20.8
22.5
-10
9.4
-25.1
1.8
-62.6
-291.9
-74.1
24
-11.2
-2.8
-6.2
14.8
-265.8
72.3
35.7
76.8
18.8
42.1
226.9
41.9
100.8
40.9
106.2
25.6
0.4
5.8
-17.4
3.2
62.6
15.6
1
8.1
15.7
48.1
7.3
11.8
5.1
18.3
65.7
1
-14.7
-1.2
39.4
-84.6
-22.4
-75.8
18.8
-14.5
-58.9
66.8
24.9
1.3
-11
-2.3
20
-0.2
84.4
-10.9
-3.6
-13.2
29.6
-6
-32.3
-31
-5.5
-23.7
-9.8
07 Appendices G, Gl.xls/Appendix Gl - Abs Dif Final (11/9/2000)
-------
DIFFERENCES FROM FRM FOR POTASSIUM (nanograms/m3)
SITE
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
DATE
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
FRM MEAN
FRM-AND
FRM IMP
FRM MET
FRM URG
FRM VAPS
121
122.5
40.1
-14.8
-39.4
-13.8
-22.8
-39.8
-13.6
-352.9
-109
-25.6
-4.9
-41.5
-76.6
124.7
48.4
62.5
23.5
93.4
112
77.1
80.7
49.9
194.5
225.6
237.2
181.1
81.9
244.8
62.4
29.5
124.2
58.8
123.6
83.5
31.5
134.6
127.2
159.7
188.2
19.9
-3.7
-1.3
-7.3
-6.3
-20.3
-13.8
-8.5
0.5
0.4
4.3
-11
-5.3
7.7
-8.2
-0.8
-14.5
-1.1
7.1
-7.6
-0.8
-10.1
-5.8
-3.6
-5
52
4.7
1.7
-6.7
-1.7
-3
-5.8
1
-6.1
-16.7
31.1
-90.2
7.6
-4.7
72
-6.4
-4.8
-14
-6.5
6.3
-3.8
-6.8
-2 7
-9.4
6.5
-3.1
-75.7
-118.7
1.8
-107.3
-93.4
-63.7
-93.2
-240.2
-21.3
-184.8
-402.7
-324.1
-144
-186.3
-205.7
-87.4
-312.1
-1.6
-72.6
-64.8
-50.4
-203
-165.7
-166.7
-188.5
-1.1
-1.2
-2.4
-8.4
2.7
16.7
12.1
10.4
15.2
15.7
3.8
4.8
-0.3
3.5
0.2
5.1
16.1
10.8
-122.2
-99.3
-23.7
-126
-106.8
-141.3
-293.7
-134.7
-0.9
-528.2
-19.8
-7.3
-3.4
-21.7
6
-12.3
-3.7
-3
-6.4
53.9
-7.6
-8.8
-3.8
-14.8
-12.6
35.7
111.9
44.2
58.6
36.1
71
48.5
45.5
57.4
31.1
115.1
63.9
40.2
28.5
37.5
51.9
32.5
176.9
38.7
35.2
77.1
-2 7
-7.2
-2.6
1
-1.8
4.7
3.5
-4
0.5
2.2
-1.3
-4.1
-1.9
-5.3
-0.8
-2.4
3.8
-6.1
-5.1
-0.7
1.5
-0.1
0.5
0.8
-0.3
-0.7
12.4
-22
-1.2
2 7
10.1
1.4
-5.6
2.5
-0.7
1.1
-3
-6.7
1.6
-10.9
-25.8
-15.9
-6.8
0.1
-4.1
-7.1
-13.6
-6.8
-4.5
-32
-4.9
-11.6
-34.4
-21.8
-12.9
-12.9
-4.9
-11.6
-2
0.4
-13.7
-5.4
-7.3
-2.2
5.3
2.6
-1.3
-0.9
2.1
-0.2
-7.3
-3.1
-6.1
-1
-2.6
_2
-13.7
1.9
-3.2
-0.1
-16.5
-0.6
-3.5
-0.2
2.1
-4.1
-3.5
12.1
-7.7
1
-1.8
-0.8
-27.7
-9.1
-1.7
44.6
-4.9
4.3
-7.7
-3.7
-21.7
74.9
31.9
58.1
72.3
36.5
146.4
49.1
69.6
31.7
95.2
6.4
-0.8
0.7
3.2
0.8
11.8
5.2
0.4
1.7
7.6
26.8
-2.1
3.5
5.5
0.1
8.9
2.3
3.1
0.9
16.6
-11
-13.1
-17.7
-5.2
-3.4
-20.8
6.7
-0.9
1.3
-5.9
-0.1
1.9
-3.1
-2.9
-0.5
1.5
-3.5
4.4
-2.1
-0.6
-9.5
-3.1
-2 7
0.2
07 Appendices G, Gl.xls/Appendix Gl - Abs Dif Final (11/9/2000)
-------
DIFFERENCES FROM FRM FOR CALCIUM (nanograms/m3)
SITE
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
DATE
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
FRM MEAN
FRM-AND
FRM IMP
FRM MET
FRM URG
FRM VAPS
360.3
343.1
33.6
-113
82.3
-22.8
-206.2
60.8
-22.9
-2046.9
-408.1
-55.2
36.3
187.8
-173.1
140.4
246.1
52.3
63.7
112.4
124.5
80.1
243.1
60.6
218.6
187.9
159.4
150.8
88.8
167.1
40
25.6
148.9
29.1
38.9
65.4
32.7
123.9
129.6
157.8
116.3
-0.6
-47.1
-23.3
-21.8
-64.7
-62.5
-84.3
-16.1
-14.7
-24.6
-24.8
-26.9
-29.2
-26.9
-22.6
-12.9
-58
-8.5
-13.3
-21.9
-11.3
-19.4
-34.4
-48.2
-50.4
39
-38
-12.1
-23.4
-53.9
-34.2
-58.1
-42.1
-18.5
-37.2
3.3
-102.6
-16.6
-32.7
28.7
-17.7
-24.4
-64.8
-39.2
-10.1
-21.8
-29.9
-20.5
-43.2
-32.3
-60.2
-562.5
-1450.4
-14.1
-321.9
-615.1
-196.3
-631.2
-1540.6
-28.5
-596.6
-1073.7
-927.4
-407.8
-551.6
-530.2
-272
-965.6
-22
-198.7
-159
-187.9
-382.2
-480.6
-498
-520.3
-0.5
0.3
2.3
-18
11.8
48.2
39.5
40.9
41
40.4
6.5
2
9.1
1.4
3
15.7
35.2
13.5
-780.8
-1319
-277.6
-401.5
-633.7
-672.8
-1592.8
-283.1
-53.1
-1657.8
-69.4
-31.8
-13.2
-76
-12.1
-30.9
-20.4
-34.9
-38.3
53
-14.4
5.5
2.5
6.9
-1.6
24.1
58.9
23.8
26.7
24.8
47.7
16.7
27.7
65.5
12.6
106.9
22.5
13.7
20.3
28.1
23.4
16.8
146.1
27 2
16.4
16.8
-0.1
-13.8
-1.9
1.9
-6.6
4.7
0.4
-5.5
20.4
-0.3
-19.6
-4.9
-4.9
-5.1
-3.4
-9.6
1.7
-12.1
-2.1
-0.4
-8.8
1.3
-7.6
-0.4
1.4
-2.5
9.6
-19.1
-3.4
-0.4
-17.3
-2.4
-6
-1.8
-8.8
0.1
-25.5
6
1.6
-16.4
-77.5
-41.8
-23.9
1.6
-56.7
-25
-32.5
-17.8
-3.2
-209
-4.7
-27
-72
-39.4
-18.7
-21.6
1.8
-13.1
-5.8
6.1
-6.5
1.4
0.7
2.1
14
3.4
3.8
28.1
2.3
5.3
-0.5
-2.2
0.5
-1.1
1.2
-2.1
6.3
6.2
0.6
-1.8
-10.3
-0.6
8.2
0.3
5.6
0.4
2.2
8.6
-15.2
-3
-3.6
-1.6
-36.2
3.9
0.5
14
-2.2
3.2
-5.2
3.6
-106.7
44.3
13.1
28.5
17.1
13.6
88
13.8
17
20.8
18.3
8.4
2.2
3.7
2
0.8
11.4
2.3
-2.8
1.6
3.5
23.7
3.4
2
-1.6
-0.2
13.3
1.8
0.1
1.2
5.9
-14.7
-9.4
-29 9
0.7
-1.8
-31.2
-0.2
-1.8
3.8
11.1
1.8
6.8
-0.6
18.4
2.2
3.3
4.3
6.2
4
-0.6
-12.1
-1.6
-3.9
0.3
07 Appendices G, Gl.xls/Appendix Gl - Abs Dif Final (11/9/2000)
-------
DIFFERENCES FROM FRM FOR IRON (nanograms/m3)
SITE
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
DATE
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
FRM MEAN
FRM-AND
FRM IMP
FRM MET
FRM URG
FRM VAPS
385.9
239.4
54.6
-96.2
-52.8
-18.9
-100.4
-84.8
-12.2
-950.1
-161.6
-55.6
22.9
11.3
-198.9
220.1
113.8
72.2
58
166.2
209.9
125.7
276.6
92.5
420.6
337.6
330.3
186
173.7
304.8
94.5
57.1
270.3
73.3
141.5
174
59.8
257.3
200.8
298.4
190.2
15.2
-10.9
-8.1
-10.2
-75.2
-72.2
-67.3
-18.2
-33
-22.5
-10.6
-22.8
-38.8
-37.9
-33.6
-12.6
-54.6
-21.6
-29.9
-44
-18.1
-29.5
-43.3
-74.2
-44.2
69.4
-7
-2 9
-10.9
-37.2
-23.6
-64.6
-14.8
-9.4
-52.6
27.5
-155.3
0.8
-31.3
77.1
-15.1
-19.9
-53.1
-52.1
-19.3
-40.1
-63
-26.6
-48.4
-32.6
-61.3
-284.1
-312.9
4.7
-252.8
-332.7
-152.9
-287.5
-652.1
-39.7
-450.7
-848.7
-833.9
-294.8
-424
-473.1
-197.7
-701.1
-27
-197.3
-142.3
-182.1
-386.3
-358.8
-374
-388.1
6.5
-5.5
-3.7
-33.4
6.9
58
58.5
70.7
47.6
46.5
19.8
13
11.2
7.8
-7.7
20.7
50.4
18.7
-567.8
-356.9
-123
-348.3
-429.8
-492.9
-860.4
-298.2
-71.7
-1444.8
-93.9
-40.3
-19.6
-79.9
-19.9
-75.8
-18.2
-21.9
-45.7
101.8
-56.2
-42.3
-67.8
-31.7
-57.7
69.4
256.3
67.1
69.3
89.3
221.5
68.8
92.3
138
18.1
303.3
76.4
24
34
56.2
53.2
65.5
124.6
12
32.7
45.5
-5.2
-30
-7.5
5.2
5.6
14.8
-15.4
-22.5
-2.4
-1
-14.5
-13.8
-11.9
-5.4
-1
-18.4
15.1
-8.7
-3.5
-8.9
-4.4
1.7
-30.7
-3.2
6.7
4.7
27.4
-5.3
-4.5
1.5
-7
1.3
-10.9
1.7
-13.3
11.9
-7.2
4.8
-3.6
-0.6
-96.4
-35.3
-13.7
15.2
-57.4
-23.5
-61
-51.6
-6.1
-205.8
-9.3
-36.7
-49.5
-49
-18.5
-22.6
4.3
-16.1
-13.1
1.9
-16
-0.7
11
11.4
3.9
10.4
-11.6
5.6
2.2
20.9
1.8
-6.4
-1
-2.9
-0.4
14.3
4.2
3.4
0
5.1
-30.1
-3.5
43.3
15.7
11.8
-8.8
-12.8
8.2
-8.4
-7.5
-4
4.6
-24
4.3
-9.3
21.5
-5.1
4.2
-9.9
0.3
-73.7
84
54.6
103.1
10.9
31.9
82.2
40.3
34.6
40.8
45.5
8.2
11.5
1.7
-6
4.4
19.8
11.8
-1.4
1.2
3.9
13.7
8.4
9.1
-3.3
2.2
15.9
9.3
_2
1.6
13.3
-37.6
-4.1
-20
-3.8
-3.9
-40.2
-2.8
-8.3
-1.1
5.8
10.6
19.2
8.1
13.8
9.3
-0.9
-1.1
14.2
-17.5
-5.6
-3.1
-9.4
-15.9
-7.4
07 Appendices G, Gl.xls/Appendix Gl - Abs Dif Final (11/9/2000)
-------
DIFFERENCES FROM FRM FOR ZINC (nanograms/m3)
SITE
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
DATE
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
FRM MEAN
FRM-AND
FRM IMP
FRM MET
FRM URG
FRM VAPS
254.8
85.4
8.2
3.1
-74.1
0.1
8.5
-77.5
-0.9
-10.7
-77.8
-3.7
-15.2
-88.6
-4.5
25.5
139.2
78.8
3.9
17.4
47.1
15.9
156.1
12
47.5
25.8
25.1
25
11.4
24.8
15.7
1.5
19.2
11.8
10.1
15.7
3.8
16.5
15
22.9
14.9
1.1
2.3
3.8
-0.6
0
-6.8
-3.6
-0.1
17
1.4
-2 7
-3.4
0.4
0.4
-1.4
-2.2
1.2
3.5
-2.5
-1.6
-0.6
-0.6
-2.2
2.8
-1.8
6.6
23.8
7.2
-0.4
1.1
2 7
-0.6
39.2
1
19.1
5.7
-14.7
-0.6
1.5
6.3
-1.3
-2.6
0.2
1.5
-0.7
1.4
-0.8
0
-1.1
6
0.2
-16.2
-0.8
5.6
-1.9
-43.8
-8.6
-6.1
-24
-0.7
5.1
-19.7
-28.3
-5.2
-7.3
-13.1
-4.2
-18.9
0.7
-7.9
-2.7
-2.6
-9.7
-9.3
-4
-12.1
1.6
1
-7.1
-3.6
-1.2
18.5
0.2
0
-0.1
0.6
-0.2
-0.4
-1
3.3
0.5
-0.4
4
1.5
-7.4
5.7
-1
-5.9
-4.4
-8.7
93.6
-5.8
-2 9
-21.2
0.2
-1.2
-2 7
-0.6
-0.3
-4.6
1.8
4.4
3.9
65.1
-15.1
-14.5
-10.4
-11.8
-21.7
21.2
83.3
19.8
16.8
22.4
48.7
30.4
22.9
35.4
5.2
67.8
35.6
10.6
6.7
30.8
15.4
18.7
43.4
5.8
7.6
17.8
-3.8
-12
0.7
1.2
-1.2
4.8
-2.2
-3
15.4
0.3
3.4
-2.2
0.8
-1.9
1.1
0.5
3
-3.5
-0.7
-1
-5.9
1.6
-7.3
3.8
-0.1
2.1
9.5
18
1.2
0.3
4.5
0.2
_2
-0.1
0
2.6
6.9
2.1
-0.1
-0.5
-14.9
1.4
2.3
-3.4
-2.6
-3.3
-1
11
0.3
3
4.3
1
-3.9
2.5
2.7
-2.1
-0.5
-0.3
-2.3
-0.3
-18
0.3
-0.9
-1.6
0.5
-3.5
-0.4
13.5
0.2
4.7
-6.4
-0.8
-1.8
-2.3
0.8
-0.5
-0.6
-0.7
1.4
-3.6
-15.6
2.4
1.3
2.5
2.8
-5.8
-2.6
2.2
6.1
1.5
0.6
0.7
-4.2
1.8
2.3
54.9
42.1
53.8
42.7
41.8
49.9
16.7
6.4
11.8
5.3
11.1
14.6
14
13.2
7.9
9.9
2.3
-1.6
0
2
0.9
1.5
0.5
-0.3
0.4
0.1
-19.8
-3.2
-0.9
-0.3
0.3
0.2
1.3
-0.5
1
1.3
2.6
-3.1
0
2
1.4
1
1.6
1.4
-0.6
2.7
-3.1
-1.2
-0.2
-2.2
-0.4
-0.1
-0.8
-4.2
3.7
0
-1.3
-3
-2.3
-3.1
07 Appendices G, Gl.xls/Appendix Gl - Abs Dif Final (11/9/2000)
-------
APPENDIX G-2
Percent Differences Between the FRM,
(Reference Sampler), and the Speciation
Samplers by Site and by Sampling Period
(Manganese, Copper, Lead and Arsenic Excluded Because of Too Few Samples)
07 Appendices G, Gl.xls/Appendix G2%Dif Final (11/7/2000)
-------
Percent Differences From FRM For PM-2.5
% Difference = [(FRM Cone- Sampler Conc)/FRM]s100
SITE DATE FRM (Ug/m3) AND %
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Difference
IMP %
Difference
MET%
Difference
URG %
Difference
VAPS %
Difference
28.6
74.3
14.9
29.4
3.0
29.5
5.6
21.0
2.2
35.6
59.0
29.1
17.6
23.6
22.0
25.3
24.4
12.7
9.0
19.8
7.6
4.1
17.3
11.5
23.3
13.8
3.9
12.4
12.8
20.3
13.4
6.2
-14.6
10.7
-2.5
9.8
25.3
2 7
14.3
-1.4
4.0
0.0
-7.1
1.1
4.5
0.7
-0.9
-0.8
-1.7
-1.1
-3.7
-10.7
-6.8
1.4
4.2
4.6
1.7
-5.9
1.0
1.3
0.4
-0.7
8.1
-17.8
11.0
8.1
6.8
44.3
-1.6
6.3
-19.5
4.0
0.6
-12.7
8.7
4.4
29.0
16.3
7.3
16.1
2.0
26.1
9.0
0.0
10.2
12.1
22.2
13.1
-0.5
-6.6
4.2
14.1
8.0
-131.5
-25.3
10.4
-13.9
-2.7
-13.4
-155.0
8.1
-328.1
-16.8
-11.4
-33.4
-82.1
-6.9
-72.4
-84.7
-79.1
-110.1
-127.5
-91.6
-29.3
-98.0
-5.5
-22.9
-52.9
-62.0
-103.0
-103.9
-78.4
-115.4
-15.3
-41.5
-12.7
-31.7
-4.4
8.2
-14.9
-50.1
-36.5
-36.5
-27.2
-48.5
14.0
3.1
3.6
8.9
1.5
-7.1
3.9
7.2
-10.6
13.8
8.4
-2.6
-0.7
-118.6
-96.0
-65.9
-68.8
-125.5
-24.8
-310.9
-110.0
-86.1
-107.4
-168.8
-7.7
-0.9
-9.3
-34.8
1.2
-1.1
-38.3
1.2
3.8
4.5
-5.8
25.8
1.7
-3.5
3.0
2.8
-28.9
19.4
37.6
15.0
12.4
13.8
25.8
16.5
11.9
15.6
5.2
32.2
24.9
10.5
5.0
10.8
14.4
4.4
23.8
6.5
8.3
15.3
0.2
-2.5
0.3
4.7
-0.6
0.2
-2.4
0.2
0.3
2.1
-1.1
-0.3
-5.0
-5.0
-0.6
-2 9
3.2
2.8
4.6
-3.3
-3.0
3.0
-2.6
2.1
5.7
1.5
0.5
-0.2
3.8
-3.9
5.7
-1.0
-0.2
-14.4
-0.6
-0.4
-14.6
3.2
12.7
-2 7
-5.6
-8.4
-2.8
11.7
3.5
2.9
0.8
-7.8
0.1
-7.4
-19.8
1.9
-17.8
-30.0
-10.8
-13.0
-2.8
8.4
-16.0
-4.2
-9.5
-8.9
-2.6
0.5
-2.0
-2.0
0.1
1.2
1.7
-0.2
1.7
1.8
-8.5
-17.2
-8.3
-4.6
3.9
-3.7
5.7
-1.2
-5.8
-26.2
-10.1
-9.6
-17.4
-33.9
-12.4
-25.7
32.0
-29.1
-2.2
-26.5
-18.3
-4.3
-7.1
-25.8
-13.9
8.7
-23.1
17.6
4.7
-4.4
-20.7
10.4
6.6
11.5
12.8
6.5
20.5
9.8
13.9
5.2
12.0
9.4
7.8
4.0
3.7
12.3
4.2
2.2
-0.4
4.0
0.8
12.4
5.9
2 9
2.1
4.3
0.9
-1.0
-2.5
3.7
5.2
-5.1
-2.6
-12.2
-5.7
-3.2
-15.5
-10.7
-8.1
-2.3
5.2
-9.0
-4.1
-10.9
-3.7
-12.5
-9.4
-6.3
-9.2
-20.4
-26.9
-19.0
-16.0
-4.0
-26.1
07 Appendices G, Gl.xls/Appendix G2%Dif Final (11/7/2000)
-------
Percent Differences From FRM For Sulfate by 1C
% Difference = [FRM Cone- Sampler Conc)/FRM]*100
SITE DATE FRM (Ug/m3)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
AND %
Difference
IMP Percent
Difference
MET%
Difference
URG %
Difference
VAPS %
Difference
0.6
6.0
1.6
1.9
14.0
-1.3
-1.2
7.4
21.1
9.4
-1.2
10.6
-70.2
-0.7
-2.5
5.8
7.0
-11.0
-6.8
-3.7
-59.7
-12.2
-18.5
-7.4
0.6
0.3
1.2
0.4
1.8
3.7
1.6
0.4
2.0
0.8
1.0
1.8
1.2
0.2
0.6
0.3
0.5
0.7
0.4
0.7
1.2
0.3
0.7
1.1
1.6
1.2
1.2
4.0
2.6
6.7
8.4
4.4
2.0
3.0
5.7
4.0
2.7
4.1
1.5
6.5
7.3
3.7
1.5
2.0
2.7
0.7
5.2
1.7
3.5
5.4
6.2
3.3
2.8
2.0
2.6
2.3
5.5
2.6
2.2
2.8
5.9
2.0
2.9
10.2
23.5
36.2
12.8
5.6
-2.4
-38.5
-4.7
-4.0
6.7
7.8
6.7
9.5
20.8
12.7
12.1
17.0
20.0
18.2
15.5
12.6
28.6
14.3
13.3
26.8
10.4
6.7
-2.0
50.4
0.0
2.5
10.0
18.2
3.6
5.3
-2.2
9.7
-0.2
20.1
4.3
3.4
2 7
25.3
16.4
4.4
8.5
-1.2
-5.9
2 9
-3.0
1.5
5.4
3.2
10.3
3.9
-3.1
-1.8
6.1
10.3
-1.8
-5.3
5.1
3.5
10.2
20.6
40.5
15.4
10.6
3.3
-24.2
2.3
5.1
6.7
10.8
7.9
8.6
25.0
14.3
12.1
19.2
17.1
15.9
14.1
8.4
25.0
12.9
11.5
28.0
10.4
5.8
-6.6
-8.0
-0.9
4.0
9.1
16.3
2.6
4.9
-3.0
7.8
-0.5
20.8
7.2
6.1
4.3
26.7
17.9
10.2
4.2
-1.2
0.2
1.8
3.9
11.3
0.8
-5.3
3.5
7.2
14.7
2 9
-3.4
17.8
3.5
-11.9
-17.7
35.3
-10.3
3.4
1.4
-31.1
-88.4
3.0
-32.0
-16.7
-6.2
-9.5
-54.2
-23.8
-33.3
-15.7
-4.6
-4.2
-6.7
-14.3
-24.3
-8.9
6.4
-9.6
-8.1
3.4
-1.8
0.8
9.7
13.3
0.3
5.8
-4.2
6.7
-6.3
16.9
1.1
4.2
4.0
21.2
8.0
-1.4
1.9
-4.1
0.3
-5.6
-0.2
2.1
2.1
9.9
2.7
-7.1
-2.0
3.8
34.4
-4.1
3.1
1.7
-10.2
-14.7
36.2
0.0
-3.4
-1.6
-20.5
-7.0
-3.0
1.3
3.9
-3.4
5.2
0.0
4.8
-6.1
8.5
5.7
4.6
0.0
-5.0
-17.9
-4.3
4.4
21.0
0.0
-10.0
-3.3
13.0
-2.3
0.7
5.4
13.8
9.2
-3.4
2.0
4.1
3.1
12.3
-0.3
7.1
-2 9
1.4
2.0
7.3
7.0
5.2
-3.5
-4.8
-4.7
-3.1
3.3
6.7
7.9
2 7
-8.0
-0.4
-2.3
-8.0
-4.0
-3.9
-14.7
11.1
-11.9
-20.6
31.0
-10.3
-1.1
-1.4
-40.4
-65.1
-7.1
3.4
8.3
7.9
9.1
10.6
5.7
9.1
4.2
7.6
7.1
0.0
8.9
30.6
7.0
0.8
8.3
11.1
11.4
21.3
32.0
15.2
13.6
15.7
15.7
14.7
22.1
10.7
11.7
11.5
17.1
13.4
5.5
-2.8
7.3
-7.1
5.1
-5.7
-3.0
0.4
6.9
-0.4
2.5
19.4
6.7
-5.8
12.2
-5.2
07 Appendices G, Gl.xls/Appendix G2%Dif Final (11/7/2000)
-------
Percent Differences From FRM for Particle Nitrate
% Difference = [FRM Cone- Sampler Conc)/FRM]*100
SITE DATE FRM (Ug/m3) AND % Difference
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
IMP %
Difference
MET%
Difference
URG %
Difference
VAPS %
Difference
9.6
38.0
3.1
12.1
14.1
8.9
-5.2
10.0
19.5
9.1
-14.7
4.2
-5.7
-6.3
-24.8
-2.8
1.1
-9.9
-113.1
-3.8
-11.3
-36.0
-9.7
9.7
0.4
9.5
0.1
16.7
25.8
12.2
6.2
9.7
3.5
5.3
2.6
3.7
0.5
1.4
2.0
0.4
4.4
2.2
7.4
4.2
0.2
0.3
2.4
6.8
5.5
2.4
5.5
1.6
4.6
8.6
3.4
3.5
2.7
6.5
3.1
2.2
2.6
0.9
6.1
3.9
1.6
1.1
3.8
6.1
0.1
1.0
0.3
0.2
1.0
1.1
1.5
0.7
0.3
0.2
0.3
0.1
0.3
1.8
0.7
1.1
0.4
2.3
17.7
11.4
27.0
23.1
14.8
6.5
-14.9
17.1
5.8
-20.8
5.1
-37.3
2 7
11.1
5.8
3.0
6.8
7.7
4.0
5.5
1.4
20.0
3.5
0.0
5.4
1.5
-3.7
2.8
-144.8
6.5
8.0
14.5
19.1
2.6
9.7
3.5
9.6
0.8
29.4
-2 9
9.3
14.5
50.5
25.8
15.8
-7.7
-10.0
-76.0
-17.4
-31.7
0.0
-4.8
1.5
-76.0
5.9
-96.6
-218.2
-52.9
8.6
-4.1
-46.5
-32.5
-2.2
13.6
6.8
17.2
38.5
7.9
-0.1
-21.4
2 7
0.2
23.4
17.1
-7.5
21.8
8.9
8.8
4.0
15.9
20.4
5.8
15.5
17.2
30.0
13.8
22.5
35.0
35.8
17.8
-8.2
-17.2
-3.5
3.9
8.0
11.6
-3.0
3.5
-6.1
5.1
-2.3
26.1
10.3
3.1
10.7
49.5
21.1
12.6
7.7
-8.0
-1.9
-10.2
-3.0
-60.0
17.7
-103.5
-200.0
-61.8
26.9
1.4
-39.5
-45.0
-6.7
-0.2
-11.4
12.6
-30.8
-1.4
-5.7
-33.1
-5.2
-8.3
-33.8
-7.9
-55.7
-22.2
-16.1
22.1
-3.2
-0.5
0.1
4.0
5.0
-27.6
-8.9
-0.4
-6.4
-16.5
-4.2
-4.9
-1.1
0.8
6.2
9.3
-2.2
2.9
-7.0
1.4
-4.2
20.7
-12.5
39.4
5.0
43.0
8.1
-84.6
28.0
-96.0
-47.8
-55.5
-27.1
-15.0
-13.4
-104.0
-11.8
-144.8
-254.6
-88.2
84.0
-68.4
-60.0
-22.7
-15.4
-6.9
-3.4
-46.6
-8.6
-73.4
-37.7
-74.9
-42.1
-35.8
-55.3
-52.3
-34.6
-37.7
-20.0
-79.3
-60.2
-36.6
-59.6
-79.3
-5.0
-19.0
-8.9
-2.3
-8.3
9.8
4.8
-6.7
-2 9
-19.7
-17.2
19.6
-34.9
2 9
-15.7
31.8
12.8
1.2
-30.8
-58.0
-24.0
-69.6
-105.0
-51.4
-56.5
-41.8
-180.0
-23.5
-110.3
-114.7
-78.9
-38.4
-118.4
-90.0
-36.9
-40.9
9.9
-23.1
-4.3
-4.7
-65.7
-9.8
-14.7
-68.2
-17.8
-27.7
-29.2
-40.9
-34.8
-91.0
-22.6
-8.3
0.0
-63.1
-28.8
-77.3
-15.6
-98.2
6.6
11.5
30.6
-21.8
-5.1
-2.6
8.7
-6.1
20.7
-29.2
-8.8
42.1
8.1
-5.1
-53.9
-34.0
-68.0
-73.9
-64.5
-82.3
-52.2
-220.0
-272 7
-61.1
-185.1
-115.0
-19.1
07 Appendices G, Gl.xls/Appendix G2%Dif Final (11/7/2000)
-------
Percent Difference From FRM For Ammonium
b Difference = [(FRM Cone- Sampler Conc)/FRM Conc]*100
SITE DATE FRM AND % Difference
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
IMP %
Difference
MET%
Difference
URG %
Difference
VAPS %
Difference
3.0
13.2
1.5
4.6
18.3
0.0
24.2
12.3
38.3
19.7
38.3
24.8
11.7
-2.1
12.1
2.9
1.3
-8.6
-6.0
-6.6
57.0
-0.8
2.0
4.6
3.0
0.3
2.8
0.1
5.4
9.3
4.2
1.8
3.2
1.2
1.9
1.3
1.5
0.2
0.5
0.7
0.2
1.6
0.9
2.5
1.7
0.1
0.3
1.1
2.4
1.9
1.1
3.1
1.4
3.7
5.8
2.6
1.8
2.1
4.0
2.4
1.7
2.4
0.8
4.0
3.9
1.8
0.9
2.0
2.8
0.2
1.9
0.4
1.2
2.0
2.3
1.5
1.2
0.6
0.9
0.9
1.5
0.9
1.3
1.1
2.3
0.8
1.8
19.4
57.7
11.6
14.3
9.4
3.6
1.4
6.6
10.3
17.7
26.7
22.1
47.3
70.0
52.9
52.9
30.4
42.6
45.4
41.3
56.2
61.5
19.2
49.1
41.8
60.1
54.0
-3.9
14.5
-2 7
4.3
9.3
13.3
4.4
5.2
-4.9
12.8
12.2
24.4
1.2
4.7
6.0
38.2
25.4
10.6
-4.4
3.2
-16.7
-7.4
-2.0
-2.2
7.6
6.1
-9.4
-4.7
-11.6
-8.0
1.1
26.2
-2 7
-3.5
-2.6
15.7
30.9
76.9
20.7
28.6
21.6
16.2
-2.6
18.2
16.2
69.8
44.4
36.6
54.0
70.0
51.0
44.3
34.8
54.8
53.5
38.1
43.8
53.9
23.1
56.3
59.4
66.3
54.0
1.0
21.7
3.6
13.5
17.4
22.2
20.5
11.7
4.5
20.9
17.7
34.6
24.4
13.5
16.9
44.9
34.5
26.2
47.8
15.4
4.4
15.2
17.4
3.1
-4.7
-2.3
4.0
7.7
36.5
13.3
0.0
2.6
24.7
3.3
69.2
3.6
42.9
1.9
0.8
-10.6
12.7
-5.6
8.4
11.2
0.0
35.3
50.0
37.3
45.7
20.7
38.4
23.0
41.4
38.5
3.9
40.2
36.1
36.8
-14.8
2.9
-11.8
2.4
2.3
2.8
-3.9
1.7
-13.2
8.1
11.3
15.4
-5.5
1.6
-3.3
24.7
13.5
0.0
-0.5
-26.2
-5.8
-7.5
-7.1
13.1
13.9
9.4
-1.2
-3.5
2.0
1.1
36.5
0.4
6.5
20.8
-13.5
-42.3
-8.3
-21.4
-10.5
-0.9
-25.5
-15.5
-19.9
-5.0
8.6
13.0
9.3
90.0
27.5
8.6
13.0
23.9
8.1
11.1
52.1
-53.9
-3.9
3.6
8.2
-4.2
-4.4
-9.7
5.8
-10.1
-0.9
-0.8
12.8
10.7
-5.7
-0.4
2.3
1.3
19.2
-11.7
5.7
0.0
22.5
15.7
12.8
-21.7
-2 7
-7.1
-24.0
-19.1
-14.6
-7.6
-7.0
-28.1
-10.6
-16.3
-16.7
-12.1
-20.6
-23.9
-10.4
-27.3
4.5
6.6
42.3
7.6
57.1
11.5
10.2
-15.8
13.3
-8.4
-13.7
15.0
2.0
0.0
-8.7
-2.6
0.0
4.0
49.7
7.7
-3.9
0.9
12.3
1.0
7.1
15.2
10.1
15.1
19.3
35.0
24.4
14.4
18.5
21.5
20.2
26.9
4.5
11.4
20.2
40.5
27.9
13.1
-17.4
4.3
-9.5
-6.6
-23.0
-13.1
-18.3
-26.6
-24.4
-28.0
1.1
-1.6
-23.9
-6.5
-0.6
07 Appendices G, Gl.xls/Appendix G2%Dif Final (11/7/2000)
-------
Percent Differences from FRM For Organic Carbon
% Difference = [(FRM Cone- Sampler Conc)/FRM Conc]*100
SITE DATE FRM AND % Difference
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
IMP %
Difference
MET%
Difference
URG %
Difference
VAPS %
Difference
7.7
10.0
3.0
6.9
-30.1
-42.0
-43.6
-18.0
46.1
47.5
41.3
56.9
-21.4
-4.1
-15.1
-10.3
11.8
-14.8
-3.4
17.0
-94.2
-44.9
-19.8
-11.0
7.8
4.6
4.7
2.1
5.5
5.7
6.9
5.5
4.2
9.5
11.8
10.1
5.6
6.2
12.5
4.3
4.0
9.0
5.6
9.7
7.7
4.0
7.3
7.3
8.3
6.5
7.8
5.2
2.3
2.8
9.6
4.1
4.1
3.1
5.3
3.4
4.5
4.6
2.3
8.5
4.4
2.8
2.8
3.7
3.3
2.8
7.3
2.5
2.4
3.8
4.7
2.4
2.0
3.8
2.2
0.9
-4.6
-26.7
4.4
-26.0
-46.6
-19.2
-21.3
28.4
-26.9
-24.1
-19.6
-35.7
-10.3
-17.6
-33.2
-19.4
-15.5
-29.1
-11.5
-19.8
-38.9
-16.6
-16.2
-22 9
-27.9
-13.8
-23.0
-35.5
-26.8
4.5
-40.3
-11.4
-97.1
-27.1
-18.8
-7.4
-42.6
-56.7
-28.9
-54.7
-11.5
-34.3
-28.1
-19.7
4.3
-2.2
6.8
-2.1
-5.8
-20.0
-30.1
-48.0
-2 7
23.1
57.8
44.1
14.4
22 9
-6.0
-13.4
14.7
16.7
8.6
29.1
20.1
-0.7
27.9
12.5
19.2
8.8
17.9
9.8
20.5
21.3
14.4
12.0
20.6
17.0
10.1
12.6
21.0
-12.6
16.5
18.2
18.2
5.7
4.4
11.6
14.7
4.7
16.1
19.4
16.9
8.3
-6.8
14.0
-7.6
-1.4
6.1
28.0
17.6
28.1
23.8
14.8
25.5
14.0
24.5
41.6
0.0
17.3
7.7
-14.6
-36.9
4.0
-12.8
-34.2
-10.7
82.7
-7.8
-31.4
1.6
13.0
3.2
-11.8
-19.9
-2.1
-7.0
-20.9
-13.3
-42.8
-4.1
-15.0
78.1
-9.6
48.9
-10.0
-2.1
16.4
1.7
4.4
-4.5
16.5
-10.3
7 2
8.0
12.6
-18.7
0.5
86.0
13.7
7.0
5.0
1.0
-12.5
-17.9
-1.3
-14.4
-43.0
-6.9
-10.4
17.9
29.3
17.6
19.0
-6.0
-12.5
8.8
6.0
6.0
19.4
29.4
14.9
27.7
32.4
26.5
15.0
45.5
-2.8
-0.2
-0.5
36.2
6.8
22.6
5.3
13.6
-1.7
17.7
8.7
14.7
25.0
20.9
14.1
11.9
5.5
15.8
9.7
24.7
20.9
33.8
2 9
4.8
22.3
20.6
16.4
2 7
38.4
14.4
64.3
27.5
-11.6
13.3
13.8
2.5
21.1
24.0
45.2
22.2
43.4
-2 7
-17.8
10.1
12.4
-30.1
18.0
47.4
37.3
32.3
56.6
38.6
27.9
26.6
41.0
54.6
35.8
44.4
49.5
54.1
48.7
20.3
13.9
36.7
42.2
54.5
27.1
27.8
27.9
45.7
46.5
43.3
33.8
25.2
41.7
55.2
43.4
29.4
47.7
23.5
18.9
31.7
-47.0
22.2
10.0
18.4
3.3
1.8
8.5
3.4
3.1
1.5
3.2
-19.3
-40.3
-18.3
1.8
11.3
-38.4
-15.7
25.6
10.5
11.8
33.2
25.7
-8.2
17.0
-25.0
-25.4
-21.4
1.6
-21.2
-11.7
19.0
16.6
6.6
19.3
32.8
-0.7
11.4
9.9
35.4
13.8
27.7
18.5
07 Appendices G, Gl.xls/Appendix G2%Dif Final (11/7/2000)
-------
Percent Differences From FRM For Elemental Carbon
% Difference = [(FRMCOnc- Sampler Conc)/FRM Conc]*100
SITE DATE FRM AND %
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Difference
IMP %
Difference
MET%
Difference
URG %
Difference
VAPS %
Difference
5.2
7.7
1.6
3.2
8.6
26.1
-1.9
5.3
71.5
76.0
77.0
77.3
5.2
6.1
3.1
14.3
20.1
12.6
6.8
13.7
-5.0
25.7
3.7
8.4
4.4
1.6
2.5
0.6
2.9
5.4
3.1
2.6
1.9
4.6
4.8
5.7
2.3
2.4
5.0
1.9
1.5
4.6
3.3
4.3
3.3
1.5
2.7
3.0
4.0
2.4
2.6
3.5
0.8
1.6
5.2
2.1
1.7
1.9
3.4
2.5
2.4
2.7
1.2
5.7
3.4
1.6
1.7
1.9
1.6
1.4
3.7
0.8
1.2
2.5
2.8
0.9
1.0
1.6
0.8
18.9
8.8
16.8
62.5
19.9
16.4
-14.0
-25.4
49.5
-3.5
-7.1
4.9
13.5
6.8
18.7
25.8
17.7
34.7
28.8
25.2
26.5
32.7
28.5
8.7
5.3
-9.7
-0.4
55.1
-552.4
-29.9
70.3
-2.9
11.6
-17.8
-4.5
20.8
21.1
13.3
4.3
7.9
4.4
11.0
14.6
20.9
-1.3
38.3
20.0
50.0
24.1
17.1
15.1
24.4
6.3
-14.4
-6.4
74.6
46.5
2.0
17.9
13.6
21.9
7.1
9.9
11.6
14.0
1.0
25.4
34.9
17.3
24.9
24.2
26.8
22.4
18.4
11.0
10.7
4.1
16.3
16.7
16.3
16.9
7.2
34.5
6.1
12.8
9.6
2.9
-5.2
6.3
11.9
14.3
18.6
23.3
4.3
22.3
1.5
-4.3
-7.0
16.2
9.6
31.9
40.3
52.6
24.1
21.1
28.9
20.8
11.3
16.7
32.7
8.2
-6.8
8.9
-4.6
25.8
-9.7
-14.0
13.7
97.8
6.6
13.7
11.4
11.0
22.5
17.2
5.3
9.2
-1.9
-1.5
-13.6
-5.2
7.7
-4.0
94.9
-8.0
58.5
3.7
-1.8
5.4
-2.9
-4.1
5.8
6.6
28.2
14.9
19.9
21.6
-9.1
19.2
100.0
19.9
14.7
20.6
15.7
31.6
12.9
12.4
4.9
-3.1
0.0
6.4
15.1
17.6
2.4
3.6
-14.7
8.5
-29.2
-27.7
-6.8
6.6
7.3
6.0
10.5
-10.1
2.8
-3.2
15.0
13.2
-3.1
1.2
10.4
-4.1
11.5
4.0
9.3
0.8
6.8
-1.7
6.1
-2.4
12.5
-1.0
-4.1
-15.2
-3.6
15.5
11.2
23.3
15.5
4.4
2.6
13.5
18.7
23.6
4.5
14.9
9.2
51.3
-3.5
11.6
4.9
10.0
3.1
-27.5
-18.0
12.6
-5.2
30.4
10.8
16.6
-3.6
-17.4
-1.1
9.1
14.8
14.1
10.8
21.6
23.5
15.3
8.6
18.3
23.1
17.4
19.4
19.1
17.7
14.4
19.5
22.0
9.2
16.9
46.2
16.2
16.4
38.8
25.2
24.7
-5.2
7.0
9.9
6.1
6.4
29.8
34.4
22.7
3.2
50.0
19.0
-59.2
24.4
15.6
-19.4
0.8
0.6
3.1
1.2
1.1
0.7
1.4
-20.7
-3.6
13.1
3.3
-14.2
4.6
12.6
2.4
-1.8
21.6
16.7
-1.9
-16.9
15.4
8.5
23.6
11.4
-6.6
-15.4
18.2
-17.1
-7.3
3.9
0.0
-17.9
-27.7
28.7
19.5
25.5
22.6
-0.9
-15.4
07 Appendices G, Gl.xls/Appendix G2%Dif Final (11/7/2000)
-------
Differences From FRM For Sulfur By XRF
% Difference =[(FRM Cone - Sampler Conc)/FRM Conc]*100
SITE DATE FRM (ug/m3)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
AND %
Difference
IMP %
Difference
MET %
Difference
URG %
Difference
VAPS %
Difference
204.7
1928.3
605.9
-1.5
-15.9
-2.9
2.8
-9.9
5.9
-61.5
-18.7
-9.6
-0.2
-19.8
-11.0
242.3
111.0
310.6
120.6
644.8
1391.7
645.7
174.7
787.5
299.6
365.5
642.8
385.5
71.2
237.9
122.3
161.9
226.6
176.7
285.2
402.3
85.3
237.1
407.0
479.5
434.9
13.2
6.0
-1.4
1.3
1.4
-5.9
-24.2
4.6
-0.5
2.4
1.8
-4.5
-18.4
6.7
-5.3
8.5
-4.5
6.2
6.0
0.4
5.9
0.0
1.2
4.3
8.2
39.5
20.2
1.4
6.5
6.2
2.3
-17.6
18.4
8.8
-9.5
17.6
-37.0
6.2
-2.8
33.0
-1.4
11.6
-3.2
4.1
6.1
9.6
15.4
3.3
7.3
11.3
9.4
-20.3
-36.9
-13.8
-6.1
-9.2
-3.4
-29.5
-63.7
-2.4
-19.8
-24.3
-22.0
-11.9
-78.2
-24.5
-42.8
-38.0
15.8
-10.5
-11.1
-8.0
-19.7
-13.2
6.7
-12.7
-9.0
-3.1
-4.7
-26.6
-1.4
0.6
2.1
-1.7
0.4
7.2
-3.4
11.7
-5.1
10.8
9.4
3.8
3.5
5.3
-11.1
-16.9
-10.8
-21.0
-0.9
-33.5
7.1
-6.6
9.9
-112.2
8.2
1.5
14.9
-0.2
7.2
-5.4
10.1
15.1
9.1
1322.8
-16.5
-13.4
-25.9
-11.9
-18.5
2232.9
2794.0
1429.6
640.2
1033.4
2026.7
1480.9
900.1
1474.4
440.7
2202.8
2411.2
1289.9
394.2
589.7
891.5
261.7
1820.8
611.7
1231.0
2004.1
-5.8
-5.7
-1.7
1.9
-3.8
4.2
1.3
2.7
1.4
1.6
2.2
-2.3
1.6
-0.3
2.0
-5.6
1.0
-3.5
-1.6
-2.2
2.9
4.9
1.4
7.2
3.6
7.2
13.5
58.7
7.0
7.8
9.5
7.7
-0.7
5.6
2.4
8.0
4.3
6.8
6.8
-13.0
-12.3
-2.4
-5.2
-7.7
-1.1
-10.5
-6.7
-5.7
-4.5
-3.8
-4.5
-3.8
-14.4
-10.7
-10.3
-2.7
-3.3
-10.8
-4.1
-12.1
-15.9
-6.3
-8.3
-6.6
-2.2
-1.9
-3.0
-1.2
0.2
1.0
-3.0
-3.7
-9.6
-12.5
-8.7
-6.9
-7.1
-7.0
-4.7
-4.5
-8.6
1.0
-6.3
0.2
1.3
-9.3
3.7
24.1
1.5
2.6
-2.6
3.1
-4.3
-2.1
-8.2
1171.4
-1.0
17.0
-1.4
-8.9
69.6
692.4
867.2
861.3
1953.4
888.5
712.1
1007.4
2083.7
677.9
1046.2
7.9
-2.6
-0.5
6.4
6.3
3.3
0.9
-2.2
3.1
3.3
18.6
-6.1
6.1
5.8
3.8
2.7
-1.0
-4.6
8.6
11.0
-3.6
-4.3
-1.0
-7.6
0.1
0.4
0.2
-5.4
-7.2
-9.0
-6.2
-3.6
-7.4
-11.0
-6.2
-8.8
-12.7
-1.8
0.4
-0.8
-2.8
-10.6
-1.1
-8.1
07 Appendices G, Gl.xls/Appendix G2%Dif Final (11/7/2000)
-------
Percent Differences From FRM For Silicon
% Difference = [(FRM Cone- Sampler Conc)/FRM Conc]*100
SITE DATE FRM (Ug/m3) AND % Difference
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
IMP %
Difference
MET%
Difference
URG %
Difference
VAPS %
Difference
377.8
216.5
16.2
-18.7
23.1
-313.0
-61.3
5.0
-49.4
-537.7
-135.3
-699.4
10.4
46.6
-2006.2
150.1
145.7
43.0
101.1
121.4
235.9
118.7
280.1
71.7
467.1
475.4
403.3
416.5
211.5
444.2
133.1
75.1
333.6
55.7
94.0
172.0
49.9
395.3
312.2
374.7
320.4
6.9
-14.1
-51.6
-77.4
-42.9
-40.5
-43.1
-57.6
-0.3
-5.3
-11.7
-13.3
-29.4
-6.8
-34.6
-17.7
-34.9
-7.2
-1.5
-31.1
-44.3
-20.2
-25.0
-26.8
-39.0
22.5
-14.1
13.7
-48.2
-41.0
-20.3
-13.5
-21.7
-12.6
-19.8
0.7
-56.7
2.6
-21.3
20.6
-21.5
-52.9
-39.1
-128.2
-9.0
-23.8
-87.8
-14.3
-31.6
-14.6
-52.0
-362.4
-528.1
-77.9
-683.6
-453.1
-172.3
-416.2
-480.9
-86.6
-275.0
-513.8
-537.9
-196.8
-566.2
-293.6
-404.0
-603.7
-86.4
-515.2
-208.3
-643.9
-313.9
-329.5
-291.1
-390.9
-9.0
-3.4
-3.5
10.8
-15.9
27.1
20.8
26.9
27.8
29.4
32.2
43.0
5.5
24.4
5.4
14.5
21.4
6.4
-583.0
-509.4
-344.0
-877.6
-554.1
-671.6
-710.5
-715.9
-21.8
-1771.6
-31.6
-47.0
-50.2
-44.0
-67.3
-54.4
-39.7
-24.8
-24.8
51.8
-29.3
-59.5
-79.0
-17.2
0.8
48.5
96.5
40.5
31.2
25.0
59.1
40.3
23.7
19.3
25.9
132.8
24.7
22.1
31.8
61.3
71.6
70.5
194.9
22.6
34.4
38.4
29.7
13.1
-48.4
-12.5
-82.8
-51.3
28.3
-33.8
-106.7
63.3
4.9
-159.5
-114.9
-12.3
-16.3
-5.5
2 7
-6.7
-82.7
11.6
-73.4
2.1
26.1
12.6
-44.2
-78.0
-8.0
-76.7
-73.8
46.3
-15.0
-157.9
-55.4
26.9
-13.7
5.7
-5.3
-69.5
-53.8
-59.4
-99.3
-104.4
-103.9
100.0
-195.4
-53.9
-217.7
-669.4
-47.1
-139.5
-166.4
-201.4
-639.3
-190.9
-38.7
-16.2
100.0
-190.7
-88.8
20.4
12.3
-8.2
-30.8
-90.4
-52.8
91.1
-43.5
-65.8
100.0
1.6
-101.6
-39.8
19.2
25.6
-20.1
-7.2
5.1
-27.0
15.4
-72 9
-15.5
-15.1
50.0
-128.8
-19.0
14.9
-87.8
86.9
-7.5
42.5
-35.1
2.6
-32.1
-1291.6
-215.4
24.0
-46.7
-11.7
-25.8
61.7
-1107.5
72.3
35.7
76.8
18.8
42.1
226.9
41.9
100.8
40.9
106.2
35.4
1.1
7.6
-92.6
7.6
27.6
37.2
1.0
19.8
14.8
66.5
20.5
15.4
27.1
43.5
29.0
2.4
-14.6
-2 9
37.1
-117.0
-62.8
-98.7
100.0
-34.4
-26.0
66.3
60.9
1.2
-15.2
-6.4
26.0
-0.5
37.2
-26.0
-3.6
-32.3
27.9
-8.3
-171.8
-13.7
-5.5
-58.0
-9.2
07 Appendices G, Gl.xls/Appendix G2%Dif Final (11/7/2000)
-------
Percent Differences From FRM For Potassium
% Difference = [(FRM Cone-Sampler Conc)/FRM Conc]*100
SITE DATE FRM (ng/m3)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
AND %
Difference
IMP %
Difference
MET %
Difference
URG %
Difference
VAPS %
Difference
121.0
122.5
40.1
-12.2
-32.2
-34.4
-18.8
-32.5
-33.9
-291.7
-89.0
-63.8
-4.1
-33.9
-191.0
124.7
48.4
62.5
23.5
93.4
112.0
77.1
80.7
49.9
194.5
225.6
237.2
181.1
81.9
244.8
62.4
29.5
124.2
58.8
123.6
83.5
31.5
134.6
127.2
159.7
188.2
16.0
-7.6
-2.1
-31.1
-6.8
-18.1
-17.9
-17.0
0.3
0.2
1.8
-6.1
-6.5
3.2
-13.1
-2.7
-11.7
-1.9
5.7
-9.1
-2.5
-7.5
-4.6
-2.3
-2.7
41.7
9.7
2.7
-28.5
-1.8
-2.7
-7.5
1.2
-12.2
-8.6
13.8
-38.0
4.2
-5.7
29.4
-10.3
-16.3
-11.3
-11.1
5.1
-4.6
-21.6
-2.0
-7.4
4.1
-1.7
-60.7
-245.3
2.9
-456.6
-100.0
-56.9
-120.9
-297.7
-42.7
-95.0
-178.5
-136.6
-79.5
-227.5
-84.0
-140.1
-251.3
-2.7
-58.7
-77.6
-160.0
-150.8
-130.3
-104.4
-100.2
-4.7
-1.3
-2.1
-10.9
5.4
8.6
5.4
4.4
8.4
6.4
6.1
16.3
-0.2
6.0
0.6
4.0
10.1
5.7
-98.0
-205.2
-37.9
-536.2
-114.4
-183.3
-363.9
-269.9
-0.4
-644.9
-8.1
-11.7
-11.5
-17.5
4.9
-14.7
-11.8
-2.4
-4.0
53.9
-14.1
-16.3
-7.1
-27.5
-23.4
35.7
111.9
44.2
58.6
36.1
71.0
48.5
45.5
57.4
31.1
115.1
63.9
40.2
28.5
37.5
51.9
32.5
176.9
38.7
35.2
77.1
-7.6
-6.4
-5.9
1.7
-5.0
6.6
7.2
-8.8
0.9
7.1
-1.1
-6.4
-4.7
-18.6
-2.1
-4.6
11.7
-3.5
-13.2
-2.0
2.0
-0.3
0.5
1.8
-0.5
-1.9
17.5
-45.4
-2.6
8.7
8.8
2.2
-19.7
6.7
-1.4
3.4
-1.7
-17.3
4.6
-30.5
-23.1
-36.0
-11.6
0.3
-5.8
-14.6
-29.9
-11.9
-14.5
-27.8
-7.7
-28.9
-120.7
-42.0
-39.7
-7.3
-12.7
-33.0
-2.6
1.1
-12.2
-12.2
-12.5
-6.1
7.5
5.4
-2.9
-1.6
6.8
-0.2
-11.4
-7.7
-21.4
-2.7
-5.0
-6.2
-7.7
4.9
-9.1
-0.1
-14.8
-1.4
-6.0
-0.6
3.0
-8.5
-7.7
38.9
-6.7
2.5
-3.5
-2.5
-15.7
-23.5
-4.8
44.6
-11.0
9.6
-17.3
-8.3
-48.7
74.9
31.9
58.1
72.3
36.5
146.4
49.1
69.6
31.7
95.2
8.5
-2.5
1.2
4.4
2.2
8.1
10.6
0.6
5.4
8.0
35.8
-6.6
6.0
7.6
0.3
6.1
4.7
4.5
2.8
17.4
-14.7
-41.1
-30.5
-7.2
-9.3
-14.2
9.6
-2.8
1.4
-7.9
-0.3
3.3
-8.5
-2.0
-1.0
2.2
-11.0
4.6
-2.8
-0.8
-6.5
-4.5
-8.5
0.2
07 Appendices G, Gl.xls/Appendix G2%Dif Final (11/7/2000)
-------
Percent Differences From FRM For Calcium
% Difference = [(FRM Cone- Sampler Conc)/FRM Conc]*100
SITE DATE FRM (Ilg/m3)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
AND %
Difference
IMP%
Difference
MET%
Difference
URG %
Difference
VAPS %
Difference
360.3
343.1
33.6
-31.4
24.0
-67.9
-57.2
17.7
-68.2
-568.1
-118.9
-164.3
10.1
54.7
-515.2
140.4
246.1
52.3
63.7
112.4
124.5
80.1
243.1
60.6
218.6
187.9
159.4
150.8
88.8
167.1
40.0
25.6
148.9
29.1
38.9
65.4
32.7
123.9
129.6
157.8
116.3
-0.4
-19.1
-44.6
-34.2
-57.6
-50.2
-105.2
-26.6
-6.7
-13.1
-15.6
-17.8
-32.9
-16.1
-56.5
-50.4
-39.0
-29.2
-34.2
-33.5
-34.6
-15.7
-26.5
-30.5
-43.3
27.8
-15.4
-23.1
-36.7
-48.0
-27.5
-72.5
-17.3
-30.5
-17.0
1.8
-64.4
-11.0
-36.8
17.2
-44.3
-95.3
-43.5
-134.7
-26.0
-33.3
-91.4
-16.6
-33.3
-20.5
-51.8
-400.6
-589.4
-27.0
-505.3
-547.2
-157.7
-788.0
-633.7
-47.0
-272.9
-571.4
-581.8
-270.4
-621.2
-317.3
-680.0
-648.5
-75.6
-510.8
-243.1
-574.6
-308.5
-370.8
-315.6
-447.4
-0.8
0.3
1.9
-22.5
19.5
22.1
21.0
25.7
27.2
24.2
16.3
7.8
6.1
4.8
9.2
12.1
22.3
11.6
-556.1
-536.0
-530.8
-630.3
-563.8
-840.0
-655.2
-467.2
-33.3
-1866.9
-41.5
-79.5
-51.6
-51.0
-31.1
-47.3
-62.4
-26.9
-24.3
53.0
-27.2
10.4
4.7
13.0
-3.0
24.1
58.9
23.8
26.7
24.8
47.7
16.7
27.7
65.5
12.6
106.9
22.5
13.7
20.3
28.1
23.4
16.8
146.1
27.2
16.4
16.8
-0.4
-23.4
-8.0
7.1
-26.6
9.9
2.4
-19.9
31.2
-2.4
-18.3
-21.8
-35.8
-25.1
-12.1
-41.0
10.1
-8.3
-7.7
-2.4
-52.4
5.4
-12.9
-1.7
5.2
-10.1
20.1
-114.4
-12.3
-3.2
-16.2
-10.7
-29.6
-6.4
-37.6
0.6
-17.5
22.1
9.8
-68.1
-131.6
-175.6
-89.5
6.5
-118.9
-149.7
-117.3
-27.2
-25.4
-195.5
-20.9
-197.1
-354.7
-168.4
-111.3
-14.8
6.6
-79.9
-34.5
25.3
-11.0
5.9
2.6
8.5
29.4
20.4
13.7
42.9
18.3
5.0
-2.2
-16.1
2.5
-3.9
5.1
-12.5
4.3
22.8
3.7
-10.7
-17.5
-2.5
30.7
1.2
11.7
2.4
7.9
68.3
-14.2
-21.9
-15.4
-9.5
-24.8
14.3
3.1
14.0
-15.7
22.9
-37.1
25.7
-762.1
44.3
13.1
28.5
17.1
13.6
88.0
13.8
17.0
20.8
18.3
19.0
16.8
13.0
11.7
5.9
13.0
16.7
-16.5
7.7
19.1
53.5
26.0
7.0
-9.4
-1.5
15.1
13.0
0.6
5.8
32.2
-33.2
-71.8
-104.9
4.1
-13.2
-35.5
-1.2
-8.7
20.8
25.1
13.7
23.9
-4.4
20.9
15.9
19.4
20.7
33.9
9.0
-3.5
-13.8
-9.4
-18.8
1.6
07 Appendices G, Gl.xls/Appendix G2%Dif Final (11/7/2000)
-------
Percent Differences From FRM For Iron
% Difference = [(FRM Cone- Sampler Conc)/FRM Conc]*100
SITE DATE FRM (Ilg/m3) AND %
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Difference
IMP %
Difference
MET%
Difference
URG %
Difference
VAPS %
Difference
385.9
239.4
54.6
-24.9
-22.1
-34.6
-26.0
-35.4
-22.3
-246.2
-67.5
-101.8
5.9
4.7
-364.3
220.1
113.8
72.2
58.0
166.2
209.9
125.7
276.6
92.5
420.6
337.6
330.3
186.0
173.7
304.8
94.5
57.1
270.3
73.3
141.5
174.0
59.8
257.3
200.8
298.4
190.2
6.9
-9.6
-11.2
-17.6
-45.3
-34.4
-53.5
-19.7
-7.9
-6.7
-3.2
-12.3
-22.3
-12.4
-35.6
-22.1
-20.2
-29.5
-21.1
-25.3
-30.3
-11.5
-21.6
-24.9
-23.2
31.5
-6.2
-4.0
-18.8
-22.4
-11.2
-51.4
-5.4
-10.2
-12.5
8.2
-47.0
0.4
-18.0
25.3
-16.0
-34.9
-19.6
-71.1
-13.6
-23.1
-105.4
-10.3
-24.1
-10.9
-32.2
-129.1
-275.0
6.5
-435.9
-200.2
-72.8
-228.7
-235.8
-42.9
-107.2
-251.4
-252.5
-158.5
-244.1
-155.2
-209.2
-259.4
-36.8
-139.4
-81.8
-304.5
-150.1
-178.7
-125.3
-204.1
11.2
-3.3
-1.8
-26.6
7.5
13.8
17.3
21.4
25.6
15.3
21.0
22.8
4.1
10.6
-12.9
10.3
16.9
9.8
-258.0
-313.6
-170.4
-600.5
-258.6
-392.1
-311.1
-322.4
-21.7
-831.8
-30.8
-42.7
-34.3
-29.6
-14.1
-43.6
-30.4
-10.9
-15.3
101.8
-55.2
-41.6
-66.6
-31.1
-56.7
69.4
256.3
67.1
69.3
89.3
221.5
68.8
92.3
138.0
18.1
303.3
76.4
24.0
34.0
56.2
53.2
65.5
124.6
12.0
32.7
45.5
-7.5
-11.7
-11.2
7.5
6.3
6.7
-22.4
-24.4
-1.7
-5.5
-4.8
-18.1
-49.6
-15.9
-1.8
-34.6
23.1
-7.0
-29.2
-27.2
-9.7
2.5
-12.0
-4.8
9.7
5.3
12.4
-7.7
-4.9
8.3
-2.3
1.7
-32.1
3.0
-25.0
18.2
-5.8
40.0
-11.0
-0.9
-37.6
-52.6
-19.8
17.0
-25.9
-34.2
-66.1
-37.4
-33.7
-67.9
-12.2
-152.9
-145.6
-92.1
-28.2
-18.1
35.8
-49.2
-28.8
2.7
-6.2
-1.0
15.9
12.8
1.8
15.1
-12.6
4.1
12.2
6.9
2.4
-26.7
-2.9
-5.2
-0.8
21.8
3.4
28.3
0.0
11.2
-11.7
-5.2
62.5
17.6
5.3
-12.8
-13.9
45.3
-2.8
-31.3
-7.5
7.0
-19.3
35.8
-28.4
21.5
-23.7
19.5
-46.1
1.4
-342.8
84.0
54.6
103.1
10.9
31.9
82.2
40.3
34.6
40.8
45.5
9.8
21.1
1.7
-55.1
13.8
24.1
29.3
-4.1
2.9
8.6
16.3
15.4
8.8
-30.3
6.9
19.3
23.1
-5.8
3.9
29.2
-44.8
-7.5
-19.4
-34.9
-12.2
-48.9
-8.1
-20.3
-2.4
6.9
19.4
18.6
25.4
16.8
23.1
-2.6
-2.7
31.2
-20.8
-51.4
-3.8
-27.2
-39.0
-16.3
07 Appendices G, Gl.xls/Appendix G2%Dif Final (11/7/2000)
-------
Percent Differences From FRM For Zinc
% Difference = [(FRM Cone- Sampler Conc)/FRM Conc]*100
SITE DATE FRM (Ilg/m3) AND %
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1/15/1999
1/18/1999
1/21/1999
1/24/1999
1/27/1999
1/30/1999
2/2/1999
2/5/1999
2/11/1999
2/14/1999
2/17/1999
2/20/1999
2/23/1999
2/26/1999
1/14/1999
1/16/1999
1/18/1999
1/20/1999
1/22/1999
1/24/1999
1/26/1999
1/28/1999
2/3/1999
2/5/1999
2/7/1999
2/9/1999
2/11/1999
2/13/1999
2/15/1999
2/17/1999
2/19/1999
2/21/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/4/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
2/26/1999
1/19/1999
1/21/1999
1/23/1999
1/25/1999
1/27/1999
1/29/1999
1/31/1999
2/2/1999
2/6/1999
2/8/1999
2/10/1999
2/12/1999
2/14/1999
2/16/1999
2/18/1999
2/20/1999
2/22/1999
2/24/1999
Difference
IMP %
Difference
MET%
Difference
URG %
Difference
VAPS %
Difference
254.8
85.4
8.2
1.2
-86.8
1.2
3.3
-90.8
-11.0
-4.2
-91.1
-45.1
-6.0
-103.8
-54.9
25.5
139.2
78.8
3.9
17.4
47.1
15.9
156.1
12.0
47.5
25.8
25.1
25.0
11.4
24.8
15.7
1.5
19.2
11.8
10.1
15.7
3.8
16.5
15.0
22.9
14.9
4.3
1.7
4.8
-15.4
0.0
-14.4
-22.6
-0.8
35.8
5.4
-10.8
-13.6
3.5
1.6
-8.9
-146.7
6.3
29.7
-24.8
-10.2
-15.8
-3.6
-14.7
12.2
-12.1
25.9
17.1
9.1
-10.3
6.3
5.7
-3.8
25.1
8.3
40.2
22.1
-58.6
-2.4
13.2
25.4
-8.3
-173.3
1.0
12.7
-6.9
8.9
-21.1
0.0
-7.3
26.2
1.3
-63.5
-0.6
7.1
-48.7
-251.7
-18.3
-38.4
-15.4
-5.8
10.7
-76.4
-112.8
-20.8
-64.0
-52.8
-26.8
-98.4
5.9
-78.2
-17.2
-68.4
-58.8
-62.0
-17.5
-81.2
41.0
5.8
-15.1
-22.6
-10.0
39.0
0.8
0.0
-0.4
2.4
-1.3
-26.7
-5.2
28.0
13.2
-2.7
17.5
10.1
-29.0
4.1
-1.3
-151.3
-25.3
-54.7
60.0
-48.3
-11.6
-186.0
0.8
-7.6
-180.0
-3.1
-3.0
-29.3
47.4
29.3
17.0
65.1
-23.2
-22.3
-16.0
-18.1
-33.3
21.2
83.3
19.8
16.8
22.4
48.7
30.4
22.9
35.4
5.2
67.8
35.6
10.6
6.7
30.8
15.4
18.7
43.4
5.8
7.6
17.8
-17.9
-14.4
3.5
7.1
-5.4
9.9
-7.2
-13.1
43.5
5.8
5.0
-6.2
7.6
-28.4
3.6
3.3
16.0
-8.1
-12.1
-13.2
-33.2
7.6
-8.8
19.2
-0.6
9.4
19.5
59.2
5.2
5.8
6.6
0.6
-29.9
-0.3
0.0
13.9
15.9
36.2
-1.3
-2.4
-17.9
7.1
13.7
-15.2
-5.3
-10.9
-4.4
31.1
5.8
4.4
12.1
9.4
-58.2
16.2
14.4
-4.8
-8.6
-4.0
-12.9
-1.4
-21.6
1.5
-5.4
-7.1
1.0
-11.5
-1.8
38.1
3.9
6.9
-18.0
-7.6
-26.9
-7.5
5.2
-2.7
-1.4
-12.1
18.4
-20.2
-18.7
12.1
7.7
11.2
5.8
-19.1
-11.4
42.3
9.0
14.2
3.9
3.7
-9.7
31.0
30.3
54.9
76.7
98.0
77.8
76.1
90.9
16.7
6.4
11.8
5.3
11.1
14.6
14.0
13.2
7.9
9.9
13.8
-25.0
0.0
37.7
8.1
10.3
3.6
-2.3
5.1
1.0
-118.6
-50.0
-7.6
-5.7
2.7
1.4
9.3
-3.8
12.7
13.1
15.6
-48.4
0.0
37.7
12.6
6.9
12.1
17.7
-6.1
16.2
-48.4
-10.2
-1.8
-15.1
-2.9
-0.8
-10.1
-42.4
22.2
0.0
-8.9
-22.7
-29.1
-31.3
07 Appendices G, Gl.xls/Appendix G2%Dif Final (11/7/2000)
-------
Appendix H
Field Evaluation of a Spiral and Cyclonic
PM2.5 Size Selective Separators
for
The Met One Ambient Chemical Speciation Sampler-SASS
Thomas M. Merrifield
MetOne Instruments Inc.
1600 Washington Ave.
Grants Pass, OR 97526
-------
Abstract
The collection efficiency of several commercial candidate ambient chemical speciation
samplers were field tested during the winter of 1998 and spring of 1999 during the 4-City
study. The separation characteristics of the Met One sampler using the PM2.5 Spiral seperator
of Aerosol Dynamics demonstrated a shallow and higher mass measurement compared to the
FRM 2.5 sampler. Re-testing of the Met One sampler at two of the four cities in April/May
1999 using a new PM2 5 Sharp Cut Cyclone separator (SCC) compared favorably with the
WINS and FRM sampler.
Field Evaluation
The selected field sites for re-testing of the Met One-SASS were Phoenix, AZ
(characterized as a western high PM mass, coarse particles and high organic carbon) and
Research Triangle Park (RTF), NC (characterized as an eastern low PM mass site with mainly
fine particles). Two SASS samplers were collocated with FRM samplers at both sites in
March-April 1999. The Met One-SASS sampler is a five channel device so sampling with
repetitive channels would minimize the number of repeat days of sampling. Channels 1, 2 and 3
were configured with a Sharp Cut Cyclone inlet Model 2.141 (indicating internal cyclone
diameter in cm) followed by 47mm PTFE FRM filter holders for mass concentration
measurement. Channel 4 and 5 were configured with the old Spiral Inlet to demonstrate the
over-sampling problem still existed during the re-test.
Channel 4 was a Spiral inlet which was lightly greased and cleaned after every 24 hour sample
event to reduce particle bounce or re-entertainment. Channel 5 was a Spiral inlet dry which
represents the SASS configuration used in the 4-City study.
Results and Discussion
PTFE filter samples from the two SASS and FRM's were gravimetrically measured and
results compared.
Table 1 presents the re-test field data and calculated precision from Phoenix, AZ from
March 28 to April 1, 1999. Table 2 presents the re-test field data and calculated precision
from RTP,NC from April 4 to 11, 1999.
-------
Table 1 SASS Re-test Phoenix.
SCC-SASS/PQ200FRM Tests- Phoenix, AZ 28 March to 1 April 1999
Test
No.
1
2
3
4
5
FRM
PQ200
Cone.
|jg/m3
9.9
11.4
6.7
7.3
2.5
sec
SASSA1
Cone.
I_ig/m3
11.2
13.0
8.2
8.9
2.4
sec
SASSA2
Cone.
|jg/m3
13.4
13.5
9.9
9.2
1.9
sec
SASSA3
Cone.
I_ig/m3
11.9
14.2
7.4
8.7
1.5
G. Spiral
SASSA4
Cone.
I_ig/m3
11.6
13.0
7.4
8.0
1.5
Dry Spiral
SASSA5
Cone.
|jg/m3
18.3
24.9
15.4
22.6
2.8
SASSB1 SASSB2 SASSB3 SASSB4 SASSB5
11.4
14.4
8.3
8.4
11.9
14.3
7.8
8.9
11.1
14.4
8.1
9.2
10.9
13.3
7.1
8.5
18.7
24.3
15.4
20.5
1.9
2.2
2.2
2.3
Precision Comparisons-|lg/m3
Test No. Six (6) SCC Six (6) SCC+FRM Two (2) Grease Spiral+FRM Two (2) Dry Spiral+FRM
0.85
0.58
0.85
0.30
0.32
1.06
1.11
0.98
0.66
0.36
0.85
1.02
0.33
0.61
0.64
4.97
7.63
5.02
8.29
0.25
-------
Table 2. SASS Re-test RTF
SCC-SASS/PQ200FRM Tests- RTF, NC 4 April to 11 April 1999
FRM SCC SCC SCC G. Spiral Dry Spiral
Test PQ200 SASSA1 SASSA2 SASSA3 SASSA4 SASSA5
No. Cone. Cone. Cone. Cone. Cone. Cone.
1
2
3
4
5
18.3
13.3
16.4
11.0
12.1
19.4
12.9
20.1
9.4
12.5
18.7
13.7
17.4
9.8
12.5
19.0
13.5
17.9
9.5
13.1
19.7
13.2
16.6
9.3
14.4
20.9
17.0
19.1
10.7
12.1
SASSB1 SASSB2 SASSB3 SASSB4 SASSB5
19.2
13.6
17.8
9.6
12.3
18.2
13.4
18.1
9.5
12.5
19.5
13.7
17.0
8.6
12.4
18.8
17.8
17.8
10.5
13.0
22.1
17.8
20.0
11.8
13.9
Precision Comparison-|lg/m3
Test No. Six (6) SCC Six(6) SCC+FRM Two (2) Grease Spiral+FRM Two (2) Dry Spiral+FRM
0.49
0.30
1.09
0.42
0.27
0.52
0.28
1.18
0.71
0.30
0.69
2.61
0.76
0.89
1.17
1.92
2.42
1.90
0.55
1.07
-------
Appendix I
Evaluation of PM2 5 Size Selectors Used in Speciation Samplers
Thomas M. Peters, Robert A. Gussman, Lee C. Kenny, and Robert W. Vanderpool
Draft
April 3, 2000
Accepted
PM2000 Special Issue, January 2000
-------
Evaluation of PM2 5 Size Selectors Used in Speciation Samplers
Thomas M. Peters"d, Robert A. Gussmanb, Lee C. Kennyc, and
Robert W. Vanderpoola
a Research Triangle Institute, PO Box 12194, RTF, NC, 27709
b BGI, Inc., 58 Guinan Street, Waltham, MA, 02451
: Health and Safety Laboratory, Broad Lane, Sheffield, S3 7HQ, U.K.
Submitted to Aerosol Science and Technology
Running Title: PM2 5 Separators Used in Speciation Samplers
d Author to whom all correspondence should be addressed
Research Triangle Institute
Center for Environmental Engineering Technology
3040 Cornwallis Rd.
R.T.P., NC, 27709
-------
Abstract
The separation characteristics of the PM2 5 aerosol size selectors used in speciation
samplers developed for the U.S. EPA National PM2 5 Chemical Speciation Trends Network were
evaluated under clean conditions. Measurement of particle penetration versus aerodynamic
diameter was conducted using an APS 3320 in conjunction with a polydisperse test dust. The
resulting penetration curves were integrated with assumed ambient particle size distributions (40
CFR Part 53, Subpart F) to obtain an estimate of measured mass concentration and to predict bias
relative to the PM2 5 reference separator. The cut-point of two SCC cyclones, from the family of
cyclones developed by Kenny and Gussman (1997), compares favorably with the WINS, although
possessing a slight tail that extends into the coarse particle mode. A second cyclone used by the
Andersen Corp., AN 3.68, demonstrated the sharpest cut characteristics of the devices tested;
however, it possesses a D50 cut size of 2.7 jim at its design flow rate. The separation characteristics
of the Spiral separator were observed to be the shallowest and greater than 2.5 jam when greased
or ungreased. Estimated mass concentration bias relative to the FRM was within ± 5% for the
idealized fine and typical assumed ambient distributions for all separators. For the idealized coarse
distribution, estimated bias ranged between + 4% and + 8%. A comparison of these results with
actual field measurements made by the samplers that employ these separators demonstrates that a
laboratory evaluation under clean conditions, while useful for design purposes, is not sufficient to
predict separator behavior in the "real-world".
1. Introduction
The U.S. EPA is in the process of establishing a new monitoring network, the National PM2 5
Chemical Speciation Trends Network, designed to measure the chemical constituents of PM2 5 at
selected sites around the U.S. (Tolocka et al., 2000). To secure samplers for this network, the
U.S. EPA solicited innovative designs via the National PM2 5 Sampler Procurement Contract
leading to the development of several new samplers, all of which aspirate, separate, and collect
atmospheric PM2 5 in slightly differing manners. An extensive method intercomparison, the 4-City
-------
Study, was conducted from January through March 1999 to determine similarities and differences
between PM2.5 mass and speciated chemical concentrations measured with these new samplers and
the designated reference method samplers based on the Federal Reference Method (FRM).
This study provides a laboratory evaluation of the collection characteristics of the final aerosol
size selectors used in PM2 5 speciation samplers that have not been documented in peer-reviewed
literature. In addition, a comparison is provided between these laboratory results and the field
measurements made with their respective samplers during the 4-City Study and similar tests
conducted immediately following the 4-City Study. The laboratory work was carried out at the
Health and Safety Laboratory in Sheffield, England to take advantage of their well-characterized
and well-documented rapid particle size separator testing apparatus.
2 Materials and Methods
2.1 Size Selectors Tested
A photograph of each of the particle size separators tested is presented in Figure 1 (A)
through (D). Figure 2 presents a schematic diagram that identifies the critical dimensions of
each device. Further identification details are presented in Table 1.
Three of the separators tested comprise a form of round entry cyclones, shown in Figure 1
(A), (B), and (C) and in Figure 2 (A), originally presented by Smith et al. (1979). They are a
developmental variation of the original round entry cyclones developed by Southern Research
Institute (SRI) under U.S. EPA contract (Smith et al., 1979) and further refined by Kenny and
Gussman (1997). Kenny and Gussman (2000) describes the model adopted for the design of
the two sharp cut cyclones (SCC) presented in this research. The SCC 2.141 cyclone
operated at 6.7 Lpm serves as the inlet and the final separator in the Met One Instruments
SASS sampler. The SCC 1.829 is not currently associated with a speciation monitor, but R&P
Corporation plans to include this design in their real-time sulfate and nitrate monitors (Meyer,
2000). Their respective companies manufacture these two cyclones.
The AN 3.68 cyclone, shown in Figure 1(C), is similar to the SRI H (Smith et al., 1979)
and to the selector described by John and Reischl (1980). Table 2 provides a comparison of
the SRI II cyclone dimensions and the AN 3.68 cyclone as manufactured by Andersen
-------
Samplers Inc. The similarities lie in the majority of the cylindrical diameter dimensions but differ
in the height of the cylinder and cone which are longer. Aspiration in the Andersen-RAAS
speciation sampler is achieved using a PM2 5 FRM design inlet with the interior PM10 impaction
section removed and operated at a flow rate of 54 Lpm. A small portion of this flow, 6 Lpm, is
used to maintain the interior components at near ambient temperatures. The remaining 48 Lpm
bifurcates into two identical AN 3.68 cyclones operated at 24 Lpm and then passes to a
manifold from which the airflow is directed to the sample collection filters.
The Spiral collector, originally developed by Yaffee et al. (1952), is presented in Figure
1(D) and Figure 2 (B). The operating principle of this device is based upon the selective
removal of particles by centrifugal forces induced as the air passes through a spiral channel.
Further development of this device was reported by Macher and First (1984). The revision
tested in this work was developed by Aerosol Dynamics Inc. and Particle Science Inc. for the
Met One Instruments SASS sampler (Met One User's Manual, 1998). During the 4-City
Study, the Met One-SASS used five parallel Spirals to aspirate and separate the fine mode
aerosol from the environment. As a result of field measurements presented in this paper, Met
One Instruments discontinued the use of the Spiral in their speciation sampler (April 1999) in
favor of the SCC 2.141, as mentioned above.
The Met One SASS manual dated December, 1998 provided performance curves for 5
different Spiral inlets, each tested at 7.0 Lpm. Also shown is data from one of these inlets
operated at a lower flow rate of 6.5 Lpm. The manual states that "within the range from 6.5 to
7.0 Lpm, small changes in flow rate do not measurably alter the cut-point." In this work, the
Spiral separator was evaluated at 7.0 Lpm due to the fact that five of the six performance
curves in the manual were determined at this flow rate.
These size selectors are compared in the results and discussion section with the Well
Impactor Ninety-Six (WINS), described in the US Federal Register (40 CFR Part 50
Appendix L), because of its adoption by EPA as the descriptor penetration curve for PM2.5.
The WINS is an impactor with a non-standard impaction plate geometry that is used in all
reference method samplers for PM2 5. It has been described in detail and tested by several
-------
authors (Kenny et al., 2000, Peters et al., in press). The curve used for the analysis in this
paper is presented in Peters et al. (in press).
2.2 Determination of Aerosol Penetration Curves
The experimental methods used to test the cyclones were similar to those described in detail
by Maynard and Kenny (1995). The tests were carried out in an aerosol chamber with
working cross section of 1 m2. The test aerosol consisted of solid, spherical glass microspheres
(Whitehouse Scientific) with physical diameters up to 25 jim, and density 2.45 g/cm3. The
aerosol was dispersed using a rotating brush generator into the separate mixing section at the
top of the chamber. An aluminum honeycomb layer was used to remove eddies from the
aerosol which was transferred into the working section by a slow (<2 cm sec"1) steady
downflow of air. The generated aerosol typically had a number median diameter around 1 jim
and a mass median diameter around 4 jim. The number concentration was typically 100-200
particles per cubic centimeter, and was generally stable over the time scales necessary for the
test (10 minutes per separator).
The test sampling lines were situated close to the center of the chamber's working section,
connected to an Aerodynamic Particle Sizer (APS 3320, TSI, Incorporated, St. Paul, MN) via
two 15 mm diameter vertical metal tubes. The APS was situated directly below the working
section, outside the chamber. Access to the working section was gained through sealed glove
ports in the side of the chamber, which allowed the flow through each cyclone to be measured
accurately using a Gilibrator bubble flow meter placed inside the chamber. The flow through
the system was maintained using a mass flow controller, calibrated and set before each test
using the Gilibrator.
The test procedure involved placing a separator on one of the two sampling lines. Both
sampling lines to the APS shared identical geometry and switching from one to the other was
accomplished by means of ball valves. The size selection characteristics were measured by
taking five 60-second samples of the poly disperse aerosol alternately from the two sampling
lines. The ratio of the aerosol size distributions measured through each line gives the size
selective aerosol penetration through the selector alone, all other effects (including any
-------
aspiration and transfer losses) being identical in both lines. The entire test sequence was (in
most cases) repeated on two different days, in order to provide a true replicate result for each
sampler. Note that earlier tests on the WINS were carried out using an APS 3310 system,
otherwise the methods were identical.
2.3 Data Analysis
Files from the APS were exported and processed using an Excel spreadsheet to
calculate the penetration curves, taking into account the appropriate corrections for particle
density and phantom particle counts. The average particle number counted in each size bin of
the APS with the selector present was divided that without the selector present to determine the
aerosol penetration for that diameter range. The raw data were normalized so that the
penetration values tended to unity for dae= 0, where necessary.
TableCurve 2D (Jandel Scientific Software, San Rafael, CA), a computer software
program for fitting curves to experimental data, was used to model the discrete penetration
measurements with a reverse asymmetric sigmoid equation (equation 8092 in TableCurve 2D).
This equation is expressed as:
_!
/n n J(x + dln(2e -l)-c),,...
.y = a+6[l-[l+exp£ L- }-^}\e} (1)
d
where, y is the penetration represented as a fraction of one, x is particle aerodynamic diameter,
and a through e are curve parameters. This equation has been shown to fit cyclone and
impactor penetration measurements with r2 values of greater than 0.99 where symmetric
equations demonstrate poorer fits. The aerodynamic diameter associated with 84%, 16%, and
50% penetration or D84, D16, and D50, respectively, are then determined using the modeled
curve. The curve is then summarized with the D50 value (cut-point diameter) and the GSD,
which is equal to:
-------
The fitted curve was then integrated with assumed ambient particle size distributions, as
specified for equivalent methods for PM2.5 (40 CFR Part 53 Subpart F), using a Visual Basic
(Microsoft Corporation, Redmond, WA) program specifically written for this purpose. For each
separator, an estimate of mass concentration measurement was calculated for particle distributions
that represent idealized fine, typical, and coarse ambient aerosols. The various separation devices
were compared with one another by expressing each device's percent bias from the FRM in
estimated measured mass concentration.
2.4 Field Performance
Four sites in the U. S. (namely Philadelphia, PA; Phoenix, AZ; Rubidoux, CA; and RTF,
NC) were identified as appropriate sites to challenge methods of fine aerosol collection, owing to
individual differences in chemical atmospheres (Tolocka et al, 2000). The prominent objective of
the 4-City Study was to perform an intercomparison between a variety of prototype chemical
speciation samplers and the PM2 5 reference method samplers. This study was conducted during
the period of January thru March, 1999.
Sample collection details and analysis methods were presented in detail by Tolocka et al.,
2000. Briefly, teflon sample collection filters were analyzed for mass and energy dispersive x-ray
fluorescence (EDXRF) analysis. Mass concentration was determined using FRM procedures
found in 40 CFR Part 50 Appendix L. Crustal material was calculated using the EDXRF analysis
of three elements: Si, Ca, and Fe by the methods presented by Malm, et al (1994). Sulfates were
measured by ion chromatography.
Immediately following the 4-City Study, Met One Instruments conducted tests in Phoenix, AZ
and RTF, NC to compare the mass concentration measured with a reference method sampler to
their speciation sampler operated with alternative PM2 5 separators (Merrifield, 2000). During
these tests, a reference method sampler was collocated with two Met One speciation samplers. On
each Met One sampler, three of the five channels were operated with SCC 2.141 cyclones. The
remaining channels were operated with a Spiral greased with Apeizon M and a Spiral ungreased as
operated in the 4-City Study. Teflon sample collection filters were used in all channels. The post
-------
sampling analysis was limited to determination mass concentration by gravimetric methods outlined
in the FRM. The tests were conducted from March to April 1999.
3 Results and Discussion
3.1 Separator Penetration and Bias Estimates
The size-selection curves for each of the separators tested are presented in Figure 3, Figure 4,
and Figure 5. Table 3 presents the descriptive parameters of the curve fitted to the penetration
measurements for each separator. Further, this table presents the D50, D84, D16, and GSD to
summarize the separation characteristics of each device. Table 4 presents the predicted mass
concentration, difference from the FRM ideal mass concentration, and percent bias for the idealized
fine, idealized typical, and idealized coarse particle size distributions. In some instances, more than
one flow rate was tested in order to aid in identifying the flow rate that provides a D50 cut-point of
2.5 urn.
Figure 3 displays the separation curves measured for the SCCs. The SCC 1.829
possesses a cut-point diameter of 2.44 jim with a GSD equal to 1.23. The SCC 2.141 has a D50
value of 2.52 jim and 2.35 jim for flow rates of 6.7 Lpm and 7.0 Lpm, respectively. The GSD of
the SCC 2.141 was measured to be 1.24 at both flow rates. In comparison with the WINS, the
SCCs are slightly sharper below the respective device's cut-point, but possess a tail that extends
slightly into the coarse particle mode. This finding is reflected in the positive bias estimated when
the SCCs are integrated with the idealized coarse particle distribution: +6.1%, +2.1%, and +4.2%
for the SCC 2.141 operated at 6.7 Lpm, the SCC 2.141 operated at 7.0 Lpm, and the SCC
1.829 operated at 5.0 Lpm, respectively. The percent bias for the idealized fine and idealized
typical distributions ranged from +0.1% to 2.3% for both SCC designs at all flow rates tested.
The D50 cut-point of the AN 3.68 cyclone, presented in Figure 4, was measured to be 2.72
|im at its design flow rate of 24.0 Lpm. This cyclone demonstrated a slightly sharper cutoff curve,
GSD = 1.15, than the WINS (GSD = 1.18). Although the cutoff is sharp, the percent bias estimate
for the typical coarse aerosol is estimated to be +7.4% greater than the ideal FRM because the cut-
point is greater than 2.5 jim.
The AN 3.68 separation characteristics were again measured at a higher flow rate, 28.1 Lpm,
to aid in identifying the D50 cut-size for 2.5 jim. At this flow rate, the cut-point was measured to be
-------
2.33 jim with a slightly less sharp GSD, equal to 1.18, than that measured at 24 Lpm. Interpolation
between these points, using the log-normal relationship presented in Kenny and Gussman (1997)
provides an estimated D50 of 2.5 at a flow rate of 26.6 Lpm. The bias estimated for the coarse
aerosol, presented in the previous paragraph, should be eliminated by operating the AN 3.68 at this
larger flow rate; however, caution is recommended because of the tendency of cyclonic separators
to become more efficient as they are loaded with particles (Kenny et al., 2000). Further tests
should be conducted to determine the magnitude of cut-point shift after being loaded with coarse
mode aerosol before committing to flow rate alterations or design modifications.
The size-selection characteristics of the Spiral at 7.0 Lpm showed significant variability in the
measured cut-point diameter. For example, the cut-point of the ungreased Spiral was measured to
be 2.7 |im, 2.2 jim, and 1.9 jim for three discrete tests. We conjecture that, as the unit is tightened,
the gasket material deforms causing a slight, but important, dimensional change in the height of the
spiral channel which serves to increase air velocity, increase collection of smaller particles, and,
hence, shift the cutsize of the device to a lower size. The variability in separation characteristics of
the Spiral from run to run was minimized by eliminating the gasket and lapping the joining surfaces.
The tests presented in this paper are with this refined Spiral.
The performance of the ungreased Spiral demonstrated the least sharp cutoff characteristics of
all the separators evaluated (GSD = 1.30) with a cut-point of 2.69 jim. The Spiral was also
evaluated by greasing the spiral channel with a thin film of Apeizon M grease because of reports of
coarse particle breakthrough (Tolocka et al., 2000). The grease coating did not change the D50
cut-size (2.67 jim), but it did serve to sharpen the characteristic performance curve slightly to yield
a GSD of 1.24. The bias estimate for the coarse mode ambient aerosol was reduced by a small
magnitude from +6.1% when ungreased to 5.5% when greased.
3.2 Field Performance
The mass, sulfate, and crustal material concentrations measured by the reference method
sampler and by the other methods can be compared to draw conclusions about the ability of the
FRM to separate fine mode aerosol from coarse mode aerosol. Furthermore, a valuable
comparison between the predicted bias from the FRM and that actually measured at a variety of
-------
sites around the U.S. can be made. It must be kept in mind, however, that any observed
differences between samplers may be attributed to any of the physical processes involved in the
collection of a representative sample (aspiration, separation, and transport to the sample collection
filter). Chemical analysis results from the FRM and the other samplers operated at the four sites
indicated that the criteria for varied chemical atmospheres was satisfied: [Philadelphia, high sulfate;
Phoenix, high carbon and crustal material; Rubidoux, high nitrate; and RTF, all low concentrations]
(Tolocka et al., 2000).
Ammonium sulfate particles typically have an aerodynamic diameter between 0.1 jim and 1.0
|im (John et al., 1990 and Hering and Freidlander, 1982). Therefore, no difference, or very small
differences, should be observed in the measured sulfate concentrations between methods if the
characteristic separation allows particles of less than 1.0 jam to pass to the sample collection filter
with unit efficiency. On the other hand, a predominant fraction of crustal material particles have
aerodynamic diameters greater than 2.5 |im (Lundgren and Paulus, 1975); therefore, the crustal
material concentration measured with the various samplers penetrating to the sample collection filter
is a good indicator of large particle intrusion.
A summary of selected results from the 4-City Study, presented in Table 5, provides a
comparison of PM2.5 mean mass, sulfate, and crustal material concentrations measured by the
FRM, the Andersen RAAS, and the Met One SASS (with the Spiral) at the four cities. The mass,
sulfate, and crustal material concentrations measured with the FRM agree closely to those
measured with the Andersen RAAS speciation sampler in all cities. This indicates that the WINS in
the reference method sampler provides similar separation characteristics as the AN 3.68 cyclone in
the Andersen RASS speciation sampler. The bias estimates for the AN 3.68 cyclone were positive
for the fine, coarse, and typical assumed ambient size distributions; however, at all four sites, the
mean mass concentration for the Andersen RAAS was slightly less than that measured by the
FRM.
The mean sulfate concentration measured with the Met One SASS was similar to the other
methods in all cases indicating proper aspiration and collection of the fine aerosol. In contrast, the
mean mass concentration measured with the Met One SASS was greater than that measured by the
FRM at all sites. In RTF and in Philadelphia, this difference, although slight, may be attributed to an
-------
increased magnitude of crustal material reaching the sample collection filter. This difference is
observed to be markedly more pronounced at the Phoenix and Rubidoux sites where the mean
mass concentration was 2.0 and 1.4 times greater than the FRM, respectively. A bias of this
magnitude was not predicted based on the laboratory evaluation of the clean Spiral presented in the
previous section.
The tests conducted by Met One immediately following the 4-City Study to resolve these
coarse particle intrusion issues are summarized in Table 6. In both Phoenix and RTF, the mass
concentration measured by the SASS channels employing the SCC 2.141 cyclone and the greased
Spiral were observed to be very near, but slightly greater than, that measured by the reference
method sampler. This slight positive bias is consistent with the predicted bias based on the
performance characteristics of these separators. In Phoenix, the mass concentration measured by
the SASS channels operated with the ungreased Spiral was a factor of two greater than the FRM
indicating that there was sufficient coarse mode aerosol present to test for coarse particle intrusion.
As a result of these tests, Met One incorporated the SCC 2.141 into the SASS design.
4 Conclusions
The final size-selection devices for several PM2 5 chemical speciation monitors were evaluated
under laboratory conditions. The cut-point of the SCC cyclones compare favorably with the
WINS although possessing a slight tail that extends into the coarse particle mode. The AN 3.68
demonstrated the sharpest cut characteristics of the devices tested; however, it has a D50 cut size of
2.7 jim at its design flow rate. The Spiral impactor possesses the shallowest cutoff curve and has a
cut-point size that is greater than 2.5 jam when greased or ungreased.
The measured penetration curves were integrated with assumed ambient particle size
distributions (fine, typical, and coarse) to obtain an estimate of measured mass concentration and to
predict bias relative to the PM2 5 reference separator (following procedures specified in 40 CFR
Part 53, Subpart F). For all of these separators, the bias in mass concentration relative to the FRM
was estimated to be within ± 5% for the fine and typical distributions and to be within 4% and 8%
for the coarse distribution.
-------
These bias estimates were then compared to field measurements made with the speciation
samplers carried out during the 4-Cities Study. The Andersen RASS, employing the AN 3.68
cyclone, compared well with the FRM for all cities; however, the bias was observed to be slightly
negative rather than slightly positive as predicted. The laboratory-based bias predicted for the Met
One SASS operated with the Spiral inlet was severely underestimated for the cities with high crustal
material (Phoenix and Rubidoux). Subsequent tests carried out by Met One demonstrated that this
problem could be eliminated by using a SCC 2.141 cyclone or by greasing the Spiral channel.
This evaluation, with the separators clean and only tested with solid aerosol, represents a single
component of the overall understanding of the collection of ambient paniculate matter with these
samplers. The comparison of laboratory predicted bias and "real-world" observations
demonstrates the necessity to evaluate performance of these separators after becoming dirty from
field operation or with some variation of the loading tests described in 40 CFR Part 53, Subpart F.
Additionally, evaluation of the inlet aspiration and aerosol transport through the sampler is needed
to ensure that the collected sample is representative of the atmospheric aerosol.
-------
5 References
Hering, S. V., and Friedlander, S. K. (1982) Origins of Aerosol Sulfer Size Distributions in the Los
Angeles Basin, Atmos. Environ., 16,2647-2656.
John, W. and Reischl, G. (1980). A Cyclone for Size-Selective Sampling of Ambient Air. APCA
Journal 30(8):872-876.
John, W., Wall, S. M., Ondo, J. L., and Winklmayr, W. (1990). Modes in the Size Distributions
of Atmospheric Inorganic Aerosols, Atmos. Environ. 24:2349-2359.
Kenny, L.C. and Gussman, R.A. (1997). Characterization and Modelling of a Family of Cyclone
Aerosol Preseparators. J. Aerosol Sci. 28(4):677-688.
Kenny, L.C and Gussman, R.A (2000). An Empirical Approach to the Design of Cyclones for
Airborne Particulate Monitoring Applications. In press: J. Aerosol Science.
Kenny, L.C., Gussman, R.A., and Meyer M. (2000). Development of a sharp-cut cyclone for
ambient aerosol monitoring applications. Aerosol Science and Technology. 32(4):338-358.
Lundgren, D. A. and Paulus, H. J. (1975). The Mass Distribution of Large Atmospheric Particles,
JAPCA 25:1227-1231.
Macher, J.M. and First, M.W. (1984). Personal Air Samplers for Measuring Occupational
Exposures to Biological Hazards, Am. Ind. Hyg. Assoc. J. 45(2):76-83.
Malm W.C., Sisler J.F., Huffman D., Eldred R.A., Cahill T.A. (1994). Spatial and seasonal trends
in particle concentration and optical extinction in the United States. J. ofGeophys. Res.
99,1347-1370.
Maynard, A.D. and Kenny, L.C. (1995). Performance Assessment of Three Personal Cyclone
Models, Using an Aerodynamic Particle Sizer, J. Aerosol Sci. 26(4):671-684.
Merrifield, T. (2000). Private communication.
Meyer, M. (2000). Private communication.
Met One User's Manual (1998). Spiral Aerosol Speciation Sampler - SASS Operators Manual,
October, 1998. Met One Instruments, Grants Pass, OR, 97526
Peters, T.M., Vanderpool, R.W., and Wiener, R.W. (in press). Design and Calibration of the EPA
PM2 5 Well Impactor Ninety-Six (WINS). Aerosol Sci. and Technol. Anticipated
publication date: Fall 2000 in a Special FRM Issue.
-------
Smith, W.B., Wilson, R.R., and Harris, D.B.(1979). A Five-Stage Cyclone System for in Situ
Sampling. Environ. Sci. and Technol. 13:1387-1392.
Tolocka, M.P., Solomon, P.A., Mitchell, W., Norris, G., Gemmill, D.B., Vanderpool, R.W., and
Homolya, J. (2000). East Vs West in the US: Chemical Characteristics of PM2 5 During the
Winter of 1999 presented at: PM2000 Paniculate Matter and Health - The Scientific
Basis for Regulatory Decision making Specialty Conference & Exhibition. Air &
Waste Manage. Assoc., Charleston, SC.
Yaffee, C.D., Hosey, A.D., and Chambers, J.T. (1952). The Spiral Sampler- A New Tool for
Studying Paniculate Matter. Arch. Ind. Hyg. Occ. Med. 5:62-70.
-------
Table 1. Identification of size selectors tested.
Separator
SCC 2.141
SCC 1.829
AN 3.68
Spiral
WINS
Type
Cyclone
Cyclone
Cyclone
Spiral
Impactor
Design Flow
Rate
(Lpm)
6.7
5.0
24.0
7.0
16.7
Notes
Serves as the inlet and final separator in
the Met One SASS
Proposed to be used in the R&P Real-
time Sulfate / Nitrate Monitors
Serves as the final separator in the
Andersen RAAS, internally coated
Teflon
Previously served as the inlet and final
separator in the Met One SASS
Final separator in the FRM, refer to 40
CFR Part 50 Appendix L, previously
tested
Table 2. Detailed dimensions of the AN 3.68 cyclone (all dimensions are presented in centimeters).
Dimension
Cyclone diameter, DC
Inlet diameter, Din
Exit diameter, De
Cone diameter, B
Height, H
Cylinder height, h
Cone height, z
Vortex finder depth, s
Cup height, Hcup
Cup diameter, Dcup
SRI II
3.66
1.01
1.05
1.29
6.62
2.11
4.71
1.57
2.22
3.73
John and Reischl
(1980)
3.658
1.008
1.052
1.270
5.923
1.173
4.750
1.570
AN 3.68
3.68
1.01
1.09
1.28
7.07
2.33
4.74
1.55
2.26
3.10
-------
Table 3. Summary of modeled curve parameters, D84, D16, and GSD for the separators evaluated.
Separator
-------
Table 4. Estimated measured mass concentration, deviation from FRM ideal, and percent bias from
FRM ideal.
Separator
-------
Table 6. Comparison of PM2.5 mass concentrations measured with a reference method sampler and
the Met One speciation sampler operated with a SCC 2.141, a greased Spiral, and an ungreased
Spiral. (Mar-Apr 1999).
Phoenix (28 Mar 1999 to 1 Apr 1999)
Sampler
FRM
WINS
Met One
SCC 2.141
Met One
Greased
Spiral
Met One
Ungreased
Spiral
N
5
30
10
10
Mass Concentration
Mean ± St. Dev
ug/m3
7.6 ±3.0
9.0 ±4.0
8.2 ±3. 8
16.5 ±7.6
RTF (4 Apr 1999 to 1 1 Apr 1999)
Sampler
FRM
WINS
Met One
SCC 2.141
Met One
Greased Spiral
Met One
Ungreased
Spiral
N
5
30
10
10
Mass Concentration
Mean ± St. Dev
ug/m3
14.2 ±2.7
14.5 ±3. 6
15.1±3.2
16.5 ±3. 9
-------
Figure 1. Photograph of particle separators evaluated: A) SCC 1.829, B) SCC 2.141, C) AN 3.68, D) Spiral.
-------
De
Din
S
T
DC
H
T
Hcup
1
B.
B
Figure 2. Schematic diagram of the particle separators tested illustrating the important component dimensions: A) Cyclone and B) Spiral.
-------
•SCO 2.141 -7.0 Lpm
• SCC 1.829-5.0 Lpm
WINS
234
Aerodynamic Diameter, |jm
Figure 3. Size-selection curves for the SCC 2.141 and SCC 1.829.
-------
AN 3.68 - 24.0 Lpm
- -AN 3.68 - 28.1 Lpm
WINS
234
Aerodynamic Diameter, pm
Figure 4. Size-selection curve for the AN 3.68 cyclone at two flow rates.
-------
01
c
Ol
a.
1.1
0.9
0.7
0.6
0.5
0.4
0.3
0.2
0.1
234
Aerodynamic Diameter, |j m
Figure 5. Size-selection curve for the Spiral with and without a greased coating.
-------
Appendix J
Comparison of Particulate Organic and Elemental Carbon Measurements
Made with the IMPROVE and NIOSH Method 5040 Protocols
Gary A. Norris, M. Eileen Birch, Michael P. Tolocka,
Charles W. Lewis,
James B. Homolya, Paul A. Solomon
US EPA
Research Triangle Park, NC
July 2000
To be submitted to Environmental Science and Technology
-------
Comparison of Particulate Organic and Elemental Carbon Measurements
Made with the IMPROVE and NIOSH Method 5040 Protocols
Gary A. Norris1' *, M. Eileen Birch0, Michael P. Tolocka*, Charles W. Lewis*, James B. Homolya¥,
Paul A. Solomon*
* U.S. EPA, ORD/NERL, Research Triangle Park, NC 27711
D National Institute for Occupational Safety and Health, Cincinnati, Ohio 45226-1998
¥U.S. EPA OAQPS, Research Triangle Park, NC 27711
tAuthor to whom correspondence should be addressed.
To be submitted to Environmental Science and Technology.
ABSTRACT
Two primary thermal optical methods are used to measure particulate carbon collected on quartz-fiber
filters: thermal optical transmittance (TOT), and thermal optical reflectance (TOR). These two
particulate carbon measurement methods were evaluated using two different protocols: National
Institute of Occupational Safety and Health (NIOSH) Method 5040 that uses TOT; and Interagency
Monitoring of Protected Visual Environments (IMPROVE) that uses TOR. The methods are based on
the same measurement principle but differ in the method used to adjust for pyrolysis (transmittance or
reflectance), temperatures for evolving carbon, and the amount of time at each temperature step. A
combination of quality assurance, source, and ambient samples were used to evaluate the differences
between the two protocols. Elemental carbon (EC) measured by the IMPROVE protocol is
approximately twice that measured by the NIOSH protocol for urban particulate samples. The
difference in the EC concentration measured by the two methods is primarily due the highest
temperature used to evolve organic carbon (OC) in He, with the NIOSH and IMPROVE protocols
using 550 and 850 »C, respectively. Adding the OC evolved in the NIOSH method from 550 to 850
•C to EC explained the difference between the protocols for urban samples. Total carbon
concentrations measured using IMPROVE were approximately 10 percent higher than NIOSH.
Measurement precision for the two protocols also was evaluated using triplicate analyses and the
precision was similar for both methods.
INTRODUCTION
Carbonaceous material is a major component of particulate matter less than 2.5 m in aerodynamic
diameter (PM2 5) accounting for 25 to over 50 percent of the mass for samples collected in the United
States (EPA, 1996; EPA, 1999). Thermal optical transmittance and thermal optical reflectance are the
two primary thermal optical techniques used for the measuring particulate carbon. These two methods
were evaluated using two different protocols: National Institute of Occupational Safety and Health
(NIOSH) Method 5040 (Birch and Gary, 1996) that uses thermal optical transmittance; and
Interagency Monitoring of Protected Visual Environments (IMPROVE) (Malm et al., 1994) that uses
J-2
-------
thermal optical reflectance. The NIOSH Method 5040 (Eller and Cassinelli, 1996) was evaluated
using a thermal optical transmittance instrument (Sunset Laboratory , Forest Grove, OR) (Birch and
Gary 1996, Birch 1998). Interagency Monitoring of Protected Visual Environments uses thermal
optical reflectance developed by Chow et al. (1993) at Desert Research Institute (DRI) (Reno, NV).
These two methods are based on a carbon analyzer designed by Huntzicker et al (1982). Both
analyzers quantify carbon by heating filters to volatilize (or combust) the carbon, which is then oxidized
in a granular bed of MnO2, reduced to CIL, in a Ni methanator, and quantified as CF^ with a flame
ionization detector. Organic carbon (OC) is evolved in an He atmosphere, while elemental carbon is
evolved in an He/O2 atmosphere. The methods use different techniques to correct for the formation of
pyrolysis products and temperature programs for evolving organic and elemental carbon. In this
analysis the two methods will be referred to as the NIOSH and IMPROVE protocols.
NIOSH and IMPROVE protocols have been used to analyze a large number of samples with 100,000
samples analyzed with the IMPROVE protocol (Chow et al., 1993) and at least 25,000 samples by the
NIOSH protocol (Gary, Sunset Laboratory, personal communication). Currently DRI and other
researchers have 6 thermal optical reflectance or thermal optical reflectance/transmittance instruments
(5 at DRI, 1 at EPA (NERL)), and Sunset Labs and other researchers have 31 thermal optical
transmittance instruments (2 at Sunset Laboratory, 2 at EPA (NERL/NRML), 7 international).
Carbonaceous paniculate matter is classified into three main categories: organic carbon (OC),
elemental carbon (EC), and carbonate carbon (CC) (Chow et al., 1993). These categories are defined
operationally by application of the thermal/optical techniques. Organic carbon sources include
combustion, biogenic, and photochemical process. Graphic-like EC sources are mainly combustion
related. Carbonate carbon is typically associated with soil related sources (Appel et al., 1983) and
usually constitutes less than 5 % of the TC (total carbon = EC + OC + CC) (Chow et al, 1993). The
ratio of OC to EC has been determined for a large number of sources with the IMPROVE method, and
the ratio depends on the source of the particulate matter: agricultural burning (3.2); unleaded gasoline
vehicle exhaust (2.2), and diesel motor vehicle exhaust (1.2) (Watson et al,1994). For these
combustion sources, particulate carbon is a large fraction of the PM2 5 mass: agricultural burning (TC =
45.4 %), unleaded vehicle exhaust (TC = 43.6 %); diesel motor vehicle exhaust (TC = 73 %).
A number of comparisons of particulate carbon methods have been conducted using various methods
including thermal optical techniques. Cadle and Groblicki (1982) compared several measurement
methods for elemental carbon, including organic extractions, nitric acid digestion, vacuum stripping,
thermal methods, and the integrating plate method. Systematic but correlated differences were found
between the methods. Hering and colleagues (Hering et al, 1990) conducted a comparison of seven
variations of thermal carbon analysis methods. Similar results were found for OC and TC with an
average ratio of method mean to sample mean ranging from 0.84 to 1.05, but the EC concentrations
had a larger variation with ratios varying from 0.69 to 1.42. Cadle and Mulawa (1990) conducted a
J-3
-------
round-robin methods study with 11 laboratories and an automated thermal technique (Cadle et al,
1980). Elemental carbon concentrations had the largest range in the ratio between the thermal
technique with a low of 0.90 and a high of 4.3. Total carbon and organic carbon showed better
agreement with the thermal technique: TC range from 0.84 to 1.39 and OC range from 0.60to 1.16.
Birch (1998) conducted a study that compared NIOSH, IMPROVE, and thermal techniques with
coulometric detection of CO2. Similar TC results (within 14 %) were observed, in the study, but the
EC concentrations for the coulometric techniques biased high when compared to the thermal-optical
techniques: 120 % for the urban sample, 69 % for the truck exhaust, and 80 percent for the fire station
sample. Elemental carbon concentrations measured using the IMPROVE protocol were about twice
the EC concentrations obtained from MOSH for urban samples. The difference between the
IMPROVE and NIOSH protocols was less for diesel with the IMPROVE reporting 31 percent more
elemental carbon than the NIOSH method. Chow et al (2000) also have also found an average
difference of 107 percent in the EC concentrations measured by NIOSH and IMPROVE that they
attribute to tarry, apparently light absorbing, volatile compounds from that are evolved above 650 »C in
the NIOSH protocol. Total carbon concentrations were 0.7 % different between the two protocols run
on the same analyzer. Chow and colleagues have not compared their NIOSH protocol that was run
on a GreenTek (Douglasville, GA) TOT/TOR analyzer to the NIOSH method run on a Sunset
Laboratory TOT analyzer. The various carbon analysis methods showed more consistent agreement
for TC, however, the measurement of EC is operationally defined. Since TC is the sum of OC and EC,
a difference in the EC concentration between methods also effects the OC concentration, but by a
smaller percentage since OC typically is 80 - 90 percent of the TC.
Differences in EC measured by the NIOSH and IMPROVE methods may be due to the different
definitions of EC used by the IMPROVE and NIOSH methods: IMPROVE defines EC as graphitic
carbon and light absorbing OC (Chow et al, 1993; Chow et al, 2000), and NIOSH defines EC only
as graphitic-like or refractory carbon. These two definitions are used to justify the highest temperature
step in the He atmosphere of 550 or 850 «C for IMPROVE or NIOSH, respectively. Higher
temperatures provide a more selective measure of graphitic carbon by evolving higher molecular weight
hydrocarbons during the OC step with particular relevance for quantifying diesel exhaust, which is
mostly EC. The lower maximum temperature in IMPROVE results in an EC value that includes a
mixture of graphitic carbon and light absorbing organic carbon, which is relevant to the IMPROVE
Network objective of evaluating visibility (Chow et al, 2000; Malm et al, 1994).
EXPERIMENTAL DESIGN
The IMPROVE and NIOSH particulate carbon analysis techniques were compared using ambient
PM2.5 samples from 3 cities with different sources of carbon, and with quality assurance standards of
sucrose, wood smoke, diesel, diesel spiked with carbonate, and blank filters. A summary of the
samples used in this comparison is shown in Table 1. The laboratories did not know the composition of
the samples or the spiking solutions
J-4
-------
Sunset Laboratory evaluated the samples with NIOSH Method 5040 and DRI evaluated the samples
with the IMPROVE Method. Temperature programs for each laboratory are shown in Table 2. The
NIOSH and IMPROVE methods use a 1.45 and 0.516 cm2 punch from a quartz-fiber filter for
analysis, respectively. The IMPROVE method does not use a constant time for each ramp, but instead
allows the FID to return to baseline before moving to the next step and quantifies the carbon evolved in
each step (OC1, OC2, OC3, OC4, EC1, EC2, and ECS). The split between organic and elemental
carbon is based on when the reflectance or transmittance in the oxidizing atmosphere returns to the
value at the start of the analysis in the He atmosphere. An example of the two protocols from the
analysis of a Phoenix urban PM2.5 sample is shown in Figure 1. The top plot shows the temperature
programs, the middle shows the FID response, and the bottom plot shows the laser signal and indicates
the point where the reflectance or transmittance returns to the initial value. As can be seen from the
transmittance and reflectance plot, with the addition of the He/O2 mixture the EC continues to evolve
until it the laser signal reaches the level for a blank or clean filter. Calibration procedures for the
NIOSH and IMPROVE techniques are not the same. The NIOSH protocol uses sucrose dissolved in
water as the primary standard while IMPROVE uses a combination of sucrose dissolved in water,
potassium hydrogen phthalate dissolved in water, and National Institute of Standards and Technology
(NIST) traceable carbon dioxide/balance He, and methane^alance He gases.
A NIOSH laboratory (Cincinnati, OH) provided quality assurance samples. Sunset Laboratory and
DRI were each sent a quality assurance sample set: sucrose solution and a 47-mm quartz fiber filter,
carbonate solution, quartz-fiber filter with diesel paniculate, and a quartz-fiber filter with wood smoke
particulate matter. To determine the potential interference of carbonate when present in a
carbonaceous particulate sample (in this case diesel), the sample portion was analyzed with and without
a carbonate spike. However, EC results should be the same in both cases if carbonate does not
interfere. All of the standards were provided in triplicate to allow for calculation of the measurement
precision of the calibration and source samples. In the NIOSH method, CC is routinely determined
above 620 C in He, however, in IMPROVE CC is not normally quantified, but is included in the EC2
peak. Quantification of CC by IMPROVE is performed by acidifying a second filter punch before
analysis and obtaining the CC by difference.
Ambient samples were selected from the Philadelphia and Phoenix PM2 5 Chemical Speciation Sampler
Evaluation sites (Solomon et al, 2000; Tolocka et al., 2000), and the Spokane Particulate Matter and
Health Study (Haller et al.,1999). Philadelphia particulate carbon is a mix of automobile and industrial
sources (Dzubay et al., 1988), while Phoenix is a mix of automobile and photochemical organic
aerosols (Chow et al., 1991; Norris et al., 2000). The Philadelphia and Phoenix samples were
evaluated with and without XAD denuders upstream of the quartz-fiber filter to remove organic gases
(Gundel et al., 1995). Spokane filters were collected in November and December when the particulate
carbon is predominately from wood smoke (Haller et al.,1999). Two additional wood stove emissions
source samples (dry oak cordwood) from the EPA's, National Risk Management Laboratory
J-5
-------
(Research Triangle Park, NC) also were evaluated. Samples from the 3 cities and the source samples
represent a wide range of OC concentrations and EC/OC ratios. Differences between the methods are
highlighted by using the combination of samples from cities with different sources of carbon, and source
samples.
J-6
-------
RESULTS
Samples were provided first to DRI and then shipped to Sunset Laboratory. This comparison used a
combination of spiked samples, source samples, and PM2 5 samples from Phoenix and Philadelphia.
Measurement precision was evaluated using source samples provided by MOSH. Results from the
analyses are divided into calibration, wood smoke, diesel and spiked diesel exhaust, and ambient
samples from Phoenix and Philadelphia.
Sample blanks
The level of detection for the MOSH method is 0.2 g/cm2 carbon. Additional detail is provided on
the LOD for the IMPROVE method: 0.5 - 1.0 g/cm2 for OC, 0.0 - 0.2 g/cm2 for EC, and 0.0 - 0.4
g/cm2 for carbonate. The pre-cleaned blank sample used for evaluating the calibration had TC
concentrations of 0.39 and 0.30 g/cm2 for the MOSH and IMPROVE methods, respectively. Three
field blanks were evaluated from the PM2 5 Chemical Speciation Sampler Evaluation Study: MOSH
TC average = 0.93, range = 0.75 - 1.23 g/cm2; IMPROVE TC average =1.11, range = 0.87 - 1.55
g/cm2.
Calibration
Calibration of the MOSH and IMPROVE methods was evaluated using a sucrose standard (Chow et
al., 1993; Birch, 1998). Each lab was provided with a pre-cleaned quartz fiber-filter and the labs
spiked a punch of the filter with 10 L of a supplied sucrose solution with 30.1 g C per 10 L. Both
techniques reported TC and OC concentrations within 4 percent of the spiked carbon concentration.
Total carbon concentrations were 4 percent higher for the IMPROVE compared to MOSH. Since
sucrose does not contain EC, any EC generated in the analysis is due to pyrolysis that is adjusted for
using optical techniques. Each method yielded similar EC concentrations with the labs reporting 0.26
and 0.27 g/cm2 for MOSH and IMPROVE, respectively. The measured EC/TC results are less than
1%, suggesting the pyrolysis adjustment is highly accurate.
Wood smoke
Wood smoke source and wood smoke dominated ambient samples were evaluated in this comparison.
The IMPROVE and MOSH OC and TC were on average, within 5 % of each other. OC and EC
precision by both method were within 5 %. However, these methods reported large variations in the
EC concentrations. The IMPROVE EC was on average 175 percent greater than MOSH EC for a
triplicate analysis of the smoldering wood smoke sample shown in Table 3a. Both techniques had
relatively low measurement precision with a coefficient of variation of 21 and 23 percent for MOSH
and IMPROVE, respectively. This low precision is expected since the average EC concentration of
was near the EC LOD of 0.2 g/ cm2 for MOSH and IMPROVE (Chow et al, 1993; Birch and Cary,
1996). Highly loaded wood smoke source samples from a wood stove burning dry oak cordwood had
average percent differences of 5, 6, and 5 for OC, EC, and TC, respectively. These results shown in
J-7
-------
Table 3b suggest both methods give similar results for the heavily loaded wood smoke source sample in
this comparison. Spokane wood smoke samples shown in Table 3c, showed a smaller difference in the
EC concentrations relative to the smoldering wood, but not the highly loaded wood smoke samples
with an average difference of 21 percent. Differences in the EC concentrations for the Spokane
samples were dependent on the loading, with the greatest difference of 71 percent with samples SP3
that had the highest OC loading of 26 g/cm2 (based on NIOSH).
Diesel Exhaust
A diesel truck source sample provided by NIOSH was evaluated in triplicate. Average OC, EC, and
the coefficient of variation for the samples are shown in Table 4a. The IMPROVE EC was 30 percent
higher than the NIOSH EC. Total carbon measured by IMPROVE was also 7 percent higher than the
NIOSH TC. The precision for the NIOSH EC (1 percent) was slightly lower than the IMPROVE EC
precision (4 percent). These results are similar to those presented by Birch (1998) for truck exhaust:
IMPROVE 2 percent, NIOSH 2 to 11 percent (4 labs reported precision for the NIOSH method).
The diesel truck exhaust was spiked with Trona solution, a sodium carbonate ore containing both
carbonate and bicarbonate. A 10 L carbonate spike of 1.60 g/ L C was added to the diesel sample.
Results for the diesel and diesel spiked with carbonate expressed in g C are shown in Table 4b.
Neither analysis showed a significant change in EC. Based on TC, the spike recovery was 105 and 98
percent for the NIOSH and IMPROVE protocols, respectively. The NIOSH protocol quantifies CC
as part of the routine analysis since CC evolves at 620 »C. In IMPROVE CC evolves as part of EC2
but is not usually quantified. A CC of 14.9 g C was reported by Sunset, which is close to the 16.0 g
C that was added to the diesel sample. Sunset Laboratory noted in their results that the carbonate
might form "glasses" that would cause the laser transmittance to increase, causing an over-reporting of
the EC. For the IMPROVE protocol, CC evolved in OC3 (450 »C). This finding is different than
previous reported by Chow and colleagues (1993) who reported CC would show up in EC2. This
may indicate the composition of the sample may have changed with the addition of the Trona solution.
Phoenk and PhUadelphia
Samples were selected from the Phoenix and Philadelphia that represented a range of loadings. For
both cities, a set of 6 days of collocated samples was selected from those collected with a Versatile Air
Pollutant Sampler (VAPS, Pinto et al, 1998) with an XAD denuder, an IMPROVE sampler, and a
Anderson Reference Ambient Air Sampler (RAAS) (Solomon et al., 2000). The VAPS, IMPROVE,
and RAAS collected 21.6, 32.8, and 10.5 cubic meters of sample on a 47 mm quartz filter,
respectively. Use of the different samplers provided range of loadings for each day. One additional
sample from a Met-One SASS from Phoenix was included in this comparison since DRI indicated the
sample had a non-white punch after the analysis, which may indicate the presence of minerals. An
analysis by Tolocka et al. (2000), discusses the comparison between the XAD denuded and non-
denuded samples.
J-8
-------
A summary of the average sample EC and OC loadings for Phoenix and Philadelphia are given in
Tables 5a and 5b, respectively. No significant quantities of carbonate were measured on the samples
by the NIOSH method. Phoenix EC concentrations were significantly different with averages of 3.75
and 6.73 g/cm2 for NIOSH and IMPROVE, respectively. Philadelphia EC concentrations also
showed a similar discrepancy with averages of 3.10 and 6.13 g/cm2 for NIOSH and IMPROVE,
respectively. Organic carbon concentrations were on average 9 percent higher for the NIOSH vs. the
IMPROVE method. Total carbon was on average 9 percent lower for the NIOSH vs. the IMPROVE
method. Carbonate at a concentration greater than the detection limit of 0.5 g/cm2 (Bob Gary, Sunset
Laboratory, personal communication) determined by the NIOSH method was not observed on one
Phoenix sample that had a non-white punch after the IMPROVE analysis. The IMPROVE analysis
comments for non-white samples suggest that carbonate may be present in the sample. The EC, OC,
and TC concentrations were significantly different (p < 0.001) when the methods were compared using
a paired t-test and the Wilcoxon Signed Rank Test. Tables 5a and 5b summarize the statistics from the
paired data. Linear regression was used to evaluate the linear relationship between IMPROVE and
NIOSH protocols in Phoenix and Philadelphia. Slopes for EC, OC, and TC were not significantly
different for each city, and the intercepts were not significantly different from zero. Overall regression
results are summarized in Figure 1. Measurements of the EC with NIOSH and IMPROVE were
significantly different: IMPROVE EC = NIOSH EC * (1.95) - (0.25), r>= 0.92. Total carbon
measurements were also significantly different, which may be due to difference calibration techniques.
The difference in EC between the NIOSH and IMPROVE protocols was investigated further by
integrating the carbon evolved from 550 »C to the addition of oxygen in the NIOSH method using
software developed by Sunset Laboratory (CalcSOO). This fraction (labeled as OC5) was added to
the EC and subtracted from the OC, generating adjusted NIOSH parameters OC" and EC+. The OC5
fraction was calculated using thermogram data provided by Sunset Laboratory for 33 of the 37
samples. On average, this fraction accounts for 20 % of the OC measured by the NIOSH protocol.
The adjusted NIOSH EC is in excellent agreement with IMPROVE EC as shown in Figure 2. The
linear relationship for EC is IMPROVE EC = NIOSH EC * 1.03 - 0.61, r>= 0.90, and the
relationships for OC and TC are shown in Table 6. A similar adjustment was also tried by Chow et al.
(2000), however they added the OC evolved from 650 to 850 «C and found NIOSH EC was 18
percent higher than the IMPROVE. This difference is most likely due to the carbon that evolves
between 550 and 650 »C. We consider our adjustment to be the preferable approach to relate
NIOSH and IMPROVE EC measurements because the method proposed by Chow et al (2000) does
not include the carbon evolved from 550 and 650 »C.
RESULTS AND CONCLUSIONS
Paniculate carbon typically represents a large fraction of PM2 5, therefore the measurement of OC and
EC is critical. The methods are based on the same measurement principle but differ in the method used
to adjust for pyrolysis (transmittance or reflectance), temperatures for evolving carbon, and the amount
J-9
-------
of time at each temperature step. These result in the reporting of different concentrations of EC and,
OC by each method. Two conclusions can be made based on this methods comparison: IMPROVE
EC > NIOSH EC, and IMPROVE TC > NIOSH TC. The difference in EC concentrations appear to
depend on the source of the sample, with the percent differences ranging from 175 percent for a
smoldering wood smoke sample to 5 percent for a wood smoke impacted sample with low a loading.
Consistent differences were found with urban PM2 5 samples, with IMPROVE EC being a factor of
two higher than the NIOSH EC. Measurement precision for the two protocols was similar, with the
poorest precision for the smoldering wood smoke sample with a low EC loading. Additional samples
from non-urban locations need to be evaluated to determine if the association is the same for urban
motor vehicle dominated and non-urban biogenic/secondary aerosol samples. More research is
needed to understand the composition of carbon evolved at temperatures greater than 550 »C (Chow
et al, 2000) since the OC evolved at greater than 550 »C may explain differences in the EC
concentrations for urban samples in this study. This high temperature organic carbon may be pyrolysis
products from heating the sample from ambient to 550 »C, real components formed in the atmosphere,
or a combination of both.
Acknowledgements:
The authors would like to thank Bob Vanderpool and Eva Hardison at Research Triangle Institute, NC
and David Gemmil at US EPA, NERL for coordinating the sample transfer between the laboratories.
We also like to thank Dennis Finn and Candis Claiborn at Washington State University (Pullman, WA)
for providing the samples from Spokane; and Dean Smith at the U.S. EPA, National Risk Management
Laboratory, Research Triangle Park, NC, for providing the wood smoke source samples.
Disclaimer:
The U.S. Environmental Protection Agency through its Office of Research and Development funded
and managed the research described here under contract 68-D5-0040 to Research Triangle Institute.
It has been subjected to Agency review and approved for publication. Mention of trade names or
commercial products does not constitute an endorsement or recommendation for use.
J- 10
-------
REFERENCES
Appel, B. R., Y. Tokiwa, J. Hsu, E. L. Kothny, E. Hahn, and J. J. Wesolowski. 1983. Visibility
reduction as related to aerosol constituents. Final Report to the California Air Resources Board,
Agreement No. A1081-32. NTIS Report PB 84 243617.
Birch M.E., Gary R.A. (1996) Elemental carbon-based methods for monitoring occupational exposures
to particulate diesel exhaust. Aerosol Sci. and Technol. 25: 221- 241.
Birch M.E. (1998) Analysis of carbonaceous aerosols: inter-laboratory comparison. Analyst. 123: 851
-857.
Cadle S.H., Groblicki PJ. 1982. An evaluation of methods for the determination of organic and
elemental carbon in particulate samples. In Particulate Carbon: Atmospheric Life Cycle, eds. G. T.
Wolff and R. L. Klimisch, pp. 89 - 109. New York: Plenum.
Cadle S.H., Mulawa P.A. (1990). Atmospheric carbonaceous species measurement methods
comparison study: General Motors results. Aerosol Science and Technology. 12: 128-41
Chow J.C., Watson J.G., Crow D., Lowenthal D.H., Merrifield T., Comparison of IMPROVE and
NIOSH Carbon Measurements, AS&T.
Chow J.C., Watson J.G, Richards D.W., Haase D.L., McDade C., Dietrich D.L., Moon D., Sloane
C. (1991) The 1989-90 Phoenix PM10 Study, Volume II. Source Apportionment Appendices, DRI
Document No. 8931 6F2.
Chow, J.C.; Watson, J.G; Pritchett, L.C.; Pierson, W.R; Frazier, C.A.; Purcell, RC. (1993) The
DRI Thermal/Optical Carbon Analysis System: Description, evaluation, and application in U.S. air
quality studies. Atmos. Environ. 27A: 1185 - 1201.
Chui W., Eatough D.J., Eatough N.L. (1998) Fine particulate organic material in the Los Angeles Basin
-1: Assessment of the High-Volume Brigham Young University Organic Sampling System, BIG BOSS.
J. Air & Waste Manage. Assoc. 48:1024-1037.
Dzubay T.G. (1988) Development and evaluation of composite receptor methods. EPA/600/3-88/026.
Eller, P.M.; Cassinelli, M.E. Elemental Carbon (DieselParticulate): Method5040.
NIOSH Manual of Analytical Methods, 4th ed. (1st Supplement) National Insitute for
Occupational Safety and Health, DHHS (NIOSH), Cincinnati, OH; Publication No. 96-135, 1996.
J- 11
-------
Gundel, L.A.; V.C. Lee; K.R.R. Mahanama; R.K. Stevens; J.M. Daisey; (1995) Direct determination
of the phase distributions of semi-volatile poly cyclic aromatic hydrocarbons using annular denuders.
Atmos. Environ., 29: 1719-1733.
Haller L., Claiborn C., Larson T., Koenig J., Norris G., Edgar R. (1999) Airborne particulate matter
size distributions in an arid urban area," J. Air & Waste Manage. Assoc. 49: 161-168.
Hering S.V., Appel B.R., Cheng W., Salaymeh F, Cadle S.H., Mulawa P.A., Cahill T.A., Eldred
R.A., Surovik M., Fitz D., Howes I.E., Knapp K.T., Stockburger L., Turpin B.J., Huntzicker J.J.,
Zang X., McMurry P.H. 1990. Comparison of sampling methods for carbonaceous aerosols in
ambient air. Aerosol Sci. Technol. 12:200-13.
Huntzicker, J.J.; Johnson, R.L.; Shah, J.J.; and Gary, R.A. (1982). Analysis of organic and elemental
carbon in ambient aerosol by a thermal-optical method. In Particulate Carbon: Atmospheric Life
Cycle, Wolff, G.T. and Klimisch, R.L., Eds. Plenum Press, New York, NY, pp. 79-88.
Malm W.C., Sisler J.F., Huffman D., Eldred R.A., Cahill T.A. (1994) Spatial and seasonal trends in
particle concentration and optical extinction in the United States. J. Geophys. Res. 99: 1347 - 1370.
McDow S.R., Huntzicker JJ. (1990) Vapor adsorption artifact in the sampling of organic aerosol: face
velocity effects. Atmos. Environ. 24A:2563-2571.
Pinto J.P., Stevens R.K., Willis R.D., Kellog R., Mamne Y., Novak J., Santroch J., Benes I, Lenicek
J., Bures V. (1998) Czech Air Quality Monitoring and Receptor Modeling Study, Environ. Sci.
Technol. 32:843-854.
Solomon, P.A., Mitchell, W., Gemmill, D., Tolocka, M.P., Norris, G., Wiener, R., Homolya, J.,
Eberly, S., Rice, J., Vanderpool, R., Murdoch, R., Natarajan, S., and Hardison, E., (2000) Methods
intercomparison of samplers for EPA's National PM2 5 Chemical Speciation network. Submitted for
Publication, JAWMA.
Solomon, P.A., Mitchell, W., Gemmill, D., Tolocka, M.P., Suggs, J., Norris, G., Wiener, R.,
Homolya, J., Vanderpool, R., and Natarajan, S. East vs West in the US: (2000) Chemical
Characteristics of PM2 5 during the Winter of 1999. Submitted for publication, AS&T.
Watson J.G, Chow J.C., Lu Z., Fugita E.M., Lowenthal D.H., Lawson D.R (1994) Chemical mass
balance source apportionment of PM10 during the Southern California Air Quality Study. Aerosol Sci.
Technol. 21:1-36.
J- 12
-------
J- 13
-------
Table 1. Samples analyzed with the NIOSH and IMPROVE protocols
Sample
W/O XAD Denuder
Philadelphia
Phoenix
Spokane
Wood smoke source samples
Number
12
13
4
2
Description
Carbon from motor vehicles and
industrial processes
Carbon from motor vehicles and
secondary aerosols
Carbon from motor vehicles and
wood smoke
2 wood smoke source samples from
the EPA, Research Triangle Park,
NC
| With XAD Denuder
| Philadelphia
| Phoenix
| Quality Assurance
| Sucrose
1 1 1
1 1 1
1 1 1
1 1 1
|3 | Triplicate evaluation of the calibration |
Wood Smoke
Diesel
Diesel spike with carbonate
O
3
3
Triplicate evaluation of a wood
smoke sample
Triplicate evaluation of a diesel
sample
Triplicate evaluation of spiked sample
Table 2. Temperature programs used for the NIOSH and IMPROVE protocols.
Carrier gas
He
He
He
He
2 % O2 balance He
2 % O2 balance He
2 % O7 balance He
2 % O7 balance He
NIOSH
250 'C, 30 sec
500 'C, 30 sec
650 'C, 60 sec
850 'C, 90 sec
Cool to 650 'C
650 'C, 60 sec
750 'C, 60 sec
850 'C, 60 sec
940 'C, 120 sec
IMPROVE
OC1: 120 'C
OC2: 250 'C
OC3:450'C
OC4: 550 'C
EC1: 550 'C
EC2: 700 'C
EC3: 800 'C
J- 14
-------
Table 3 a. Smoldering wood smoke sample (g/cm2)
Mean
Standard
Deviation
N
CV%
Difference %
***
Ratio****
N OC*
51.7
0.5
3
0.9
I_OC*
*
53.4
1.1
3
2.1
-3
1.0
N_EC
0.1
0.0
o
20.8
I_EC
1.3
0.3
3
22.9
-175
0.1
N TC
51.8
0.5
O
0.9
I_TC
54.7
1.3
3
2.3
-5
0.9
EC/TC
0.01
...
___
...
...
...
* N = NIOSH, ** I = IMPROVE, *** 100 * (NIOSH minus IMPROVE, divided by their average),
**** NIOSH/IMPROVE
Table 3b. Wood smoke source sam
EPAEL1
EPAEL2
EPAEL3 (blank)
EPAEL1 Difference %
EPAEL2 Difference %
EPAEL3 Difference %
EPAEL1 Ratio
EPAEL2 Ratio
EPAEL3 Ratio
NO
C
386.7
260.5
0.3
le ( g/cm2)
I OC
345.5
265.7
0.2
11.2
-2.0
___
1.1
1.0
...
N_EC
8.0
10.7
0.1
I_EC
7.9
9.5
0.0
0.4
11.8
___
1.0
1.1
...
N TC
394.7
271.2
0.4
I_TC
353.7
275.2
0.2
11.0
11.0
___
1.1
1.0
—
EC/TC
0.0
0.0
0.1
—
___
___
___
___
—
Table 3c. Spokane, Washington ambient air samples (g/cm2)
SP1
SP3
SP4
SP1 Difference %
SP3 Difference %
SP4 Difference %
N OC
4.0
26.6
3.3
I OC
4.1
25.8
3.3
-2.7
3.0
1.2
N EC
0.6
3.4
0.4
I EC
0.6
7.2
0.4
-5.3
-71.0
14.4
N TC
4.5
30.0
3.8
I TC
4.7
32.9
3.9
-3.1
-9.5
-2.4
EC/TC
0.1
0.2
0.1
—
—
—
J- 15
-------
SP1 Ratio
SP2 Ratio
SP3 Ratio
1.0
1.0
1.0
0.9
0.5
1.2
1.0
0.9
1.0
___
—
—
J- 16
-------
Table 4a. Diesel Truck (g/cm2)
Mean
Standard
Deviation
N
CV%
Difference %
Ratio
N OC
16.4
0.1
3
0.9
I OC
16.3
0.7
3
4.1
0.9
1.0
N EC
4.5
0.0
3
0.5
I EC
6.1
0.3
3
4.9
-29.6
0.7
N TC
20.9
0.2
3
0.7
I TC
22.4
1.0
3
4.3
-6.6
0.9
EC/TC
0.2
—
___
___
___
—
Table 4b. Diesel filters spiked with 10 L of Trona solution (g C) *, **
Diesel
Diesel + CC
Spike
N
CV%
N OC
23.8
23.8
0.0
3
...
I OC
8.4
25.8
17.4
3
3.3
N EC
6.5
7.6
0.9
3
37.2
I EC
3.1
2.6
-0.5
3
-28.6
N TC
30.3
31.5
0.8
3
2.5
I TC
11.5
28.4
16.8
3
4.3
N CC
0.0
jg 9***
^4 9****
3
0.0
N*1.45 cm2 or 1*0.516 cm2 (punch area) ** 10 L Trona solution contains 16 g C, respectively,
*** value high since one spike was 20 L instead of 10, **** Adjusted for spike volume
J- 17
-------
Table 5a. Philadel
C
Mean
Standard
Deviation
N
Percent Diff
N/I
Paired t-test*
Wilcoxon **
phia samples ( g/cm2)
N OC
11.7
6.5
18
I OC
10.0
5.4
18
15.8
1.2
< 0.001
< 0.001
N EC
3.1
1.6
18
I EC
6.1
3.5
18
-65.7
0.5
< 0.001
< 0.001
N TC
14.8
8.0
18
I TC
16.1
8.6
18
-8.5
0.9
< 0.001
< 0.001
* p-value, ** Wilcoxon signed rank test
Table 5b. Phoenix samples (g/cm2)
C
Mean
Standard
Deviation
N
Percent Diff
N/I
Paired t-test*
Wilcoxon **
N OC
16.3
8.6
19
I OC
15.4
8.1
19
5.5
1.1
0.001
0.001
N EC
3.8
2.0
19
I EC
6.8
3.8
19
-57.3
0. 6
< 0.001
< 0.001
N TC
20.1
10.4
19
I TC
22.2
11.7
19
-10.3
0.9
< 0.001
< 0.001
* p-value, ** Wilcoxon signed rank test
J- 18
-------
Figure Captions
Figure 1. Comparison of the NIOSH and IMPROVE protocols with an urban sample from Phoenix.
Figure 2. Comparison of urban PM2 5 EC, OC, and TC measured with the IMPROVE and NIOSH
protocols.
Figure 3. Comparison of urban PM2.5 EC, OC, and TC measured with the IMPROVE and adjusted
NIOSH protocols.
J- 19
-------
1050
- 2000
- 600
- 400
- 200
200 400
600
800 1000 1200 1400 1600 1800 2000
Time (seconds)
§
op
GO
Q
E
ffi
GO
o
300
- 200
- 100
- 0
- 1200
- 1000 E
- 800 -
CO
CO
O
J-20
-------
50
40 -
6
o
~
O
W
30 -
o
£j 20
10 -
NIOSH C (Hg/cm'
J-21
-------
o
O
PH
50
40 -
30 -
20 -
10 -
10 20 30
NIOSHC(ng/cm2)
40
50
J-22
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1. REPORT NO.
EPA-454/R-01/005
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Evaluation of PM2.5 Chemical Speciation Samplers for Use in the EPA
National PM2.5 Chemical Speciation Network
5. REPORT DATE
30 March 2000
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Paul Solomon*, William Mitchell, Michael Tolocka, Gary Norris,
David Gemmill, Russell Wiener, Robert Vanderpool, Robert Murdoch,
Sanjay Natarajan, Eva Hardison
(Corresponding author)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-D5-0040 with RTI
12. SPONSORING AGENCY NAME AND ADDRESS
Director
Office of Air Quality Planning and Standards
Office of Air and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
This was a joint venture between OAQPS and ORD.
-------
16. ABSTRACT To develop improved source-receptor relationships and for better understanding the causes of high PM2.5
concentrations in the atmosphere, it is necessary to not only determine concentrations of PM2.5 mass, the NAAQS indicator, but
also the chemical components of PM2.5. A sampling program of this type, which will consist of up to 300 sites nationwide has
been initiated by EPA (Speciation Guidance Document, 1999 athttp://www.epa.gov/ttn/amtic/ pmspec.html'l. Since the PM2.5
Federal Reference Method (FRM) using only Teflon filters is not suitable for determining the chemical composition of the collected
aerosol, since carbon can not be directly measured (Speciation Guidance Document, 1999), EPA solicited innovative designs for
speciation samplers, based on performance specifications. This led to the development of three slightly different candidate
samplers manufactured by Andersen Samplers, MetOne, and University Research Glassware (URG). These samplers are designed
to allow for a nearly complete mass balance of the collected aerosol, while minimizing sampling artifacts for nitrate and allowing
flexibility for minimizing organic carbon artifacts in the future. Due to the need to have consistency across this national network,
the Speciation Expert Panel (Recommendations of the 1998 Expert Panel, 1998 at http://www.epa.gov/ttn/amtic/ pmspec. html)
recommended a methods comparison field study among the new speciation samplers, historically used samplers, and the PM2.5
FRM. The program plan for EPA's Chemical Speciation Sampler Evaluation Study (1999, http://www.epa.
gov/ttn/amtic/casacinf.html'l details the approach and implementation of the study. This report presents the approach and results
from the 4-City intercomparison study; Phase 1, of the full evaluation of these samplers.
Because of potential sampling artifacts when using filters and potential differences in inlet cutpoints and sample fractionators, the
chemical speciation samplers must be able to properly determine the chemical components of PM2.5 under a variety of atmospheric
and environmental conditions. Four locations, with different atmospheric chemical and meteorological conditions were chosen and
included: Rubidoux, CA (high nitrate and carbon and low sulfate), Phoenix, AZ (high crustal material and moderate carbon and
nitrate), Philadelphia, PA (high sulfate, moderate carbon, and low nitrate), and Research Triangle Park (RTF), NC (low PM2.5
concentrations). The latter site also allowed for a more thorough evaluation of the samplers' in-field operational performance as it
was located near EPA offices in RTF. In addition to the three candidate samplers, a Versatile Air Pollution Sampler (VAPS), an
IMPROVE sampler, and an FRM were collocated at each site. Replicate samplers were located at Rubidoux. Samples were collected
for up to 20 days during January and February, 1999 using state personnel (Rubidoux and Phoenix) or EPA contractors
(Philadelphia and RTF). All sampling periods were 24-hrs in duration. Mass and trace elements were determined on Teflon filters;
sulfate, nitrate, and ammonium were determined on either Teflon, pre-fired quartz-fiber, or nylon filters depending on the sampler;
and OC/EC were determined on pre-fired quartz-fiber filters. To minimize variability, all filter preparation, filter changing, and
chemical analyses for a particular species were preformed by one contractor. Quality assurance/quality control followed EPA
guidelines (QAPP for the Four-City PM2.5 Chemical Speciation Sampler Evaluation Study, January, 1999 Research Triangle
Institute, Project Number 07263-030).
In general, the performance of the candidate samplers is reasonable for their first use in the field. All samplers had operational
problems that increased their variability, most of which have been addressed by the manufactures. Tradeoffs exist among the
samplers for ease of use, flexibility for sampling, and cost. Performance of the samplers was excellent for sulfate and reasonable for
other stable species. However, real differences among the samplers exist for nitrate and organic carbon and possibly ammonium as
collected in the IMPROVE sampler. These differences are significant and can possibly affect design of compliance strategies for
controlling PM2.5 mass concentrations in air, as total differences as high as 3-5 • g/m3 are observed among the samplers for these
two species.
17.
KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
PM Chemical Speciation, nitrate, sulfate, organic carbon,
elemental carbon, trace elements, Rubidoux, Phoenix,
Philadelphia, PM2.5 mass and composition
18. DISTRIBUTION STATEMENT
Release Unlimited
b. IDENTIFIERS/OPEN ENDED TERMS
Air Pollution control
19. SECURITY CLASS (Report)
Unclassified
20. SECURITY CLASS (Page)
Unclassified
c. COSATI Field/Group
21. NO. OF PAGES
22. PRICE
EPA Form 2220-1 (Rev. 4-77)
PREVIOUS EDITION IS OBSOLETE
------- |