Quality Assurance Guidance Document
Quality Assurance Project Plan:
PM2.5 Chemical Speciation Sampling at Trends,
NCore, Supplemental and Tribal Sites
(An Update to the PM2.5 Speciation Trends
Network Field Sampling QAPP, December 2000)
-------
-------
EPA-454/B-12-003
June 2012
Quality Assurance Guidance Document
Quality Assurance Project Plan:
PM2.5 Chemical Speciation Sampling at
Trends, NCore, Supplemental and Tribal Sites
(An Update to the PM2.5 Speciation Trends
Network Field Sampling QAPP,
December 2000)
By:
Ambient Air Monitoring Group,
Air Quality Assessment Division
US EPA, Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina
-------
-------
Foreword
This document is the Quality Assurance Project Plan (QAPP) for all operations involved in the
PM2.5 Chemical Speciation Network (CSN). The original QAPP was issue in 2000 as the CSN
was being deployed. The CSN has evolved over time due to the advent of the NCore monitoring
network prescribed by rulemaking in October 2006 and the most dramatic changes in 2008-2010,
when the national network of samplers were consolidated to Metone SASS or SuperSASS and
URG 3000N. This version of the QAPP reflects that transition and a few new quality control
checks that are recommended to provide a better tracking process for deleterious changes in
sampler flow rates. The approach to utilizing collocated samplers for network precision has also
been expanded to a redeploy the collocated samplers at half of the historic, fixed collocation sites
to annual assignments that rotate among the other NCore sites.
This QAPP was generated using the U.S. Environmental Protection Agency (EPA) Quality
Assurance (QA) regulations and guidance described in EPA QA/R-5, EPA Requirements for
Quality Assurance Project Plans and the accompanying document EPA QA/G-5, Guidance for
Quality Assurance Project Plans. All pertinent elements of the requirements and guidance are
addressed in this QAPP. This QAPP incorporates by reference, the current QAPPs utilized by
analytical and technical support contractors. These are available at the EPA's Ambient
Monitoring Technical Information Center (AMTIC) website,
http://www.epa.gov/ttn/amtic/speciepg.html. The SOPs associated with operating the samplers are
located at http://www.epa.gov/tta/amtic/spectraimng.html. The SOPs associated with operations
performed by the contract service lab are located at: http://www.epa.gov/ttn/amtic/specsop.html.
The date in each section header indicates the last time the particular section was reviewed and/or
revised by EPA.
This 2012 version is the first revision since the publication of the 2000 document for which EPA
is requesting that State, Local and Tribal QAPP review and either (1) adopt this QAPP or (2)
revise their own QAPP. This document should be reviewed by EPA Regional Offices and the
Regional Speciation Coordinators responsible for implementing the CSN in their respective
Regions, the EPA Technical Monitors for the network, and the QA Manager of the contracted
support laboratory for the CSN. The EPA Quality Assurance Policies in EPA Order 5360.1 A2
require agencies that accept federal grant funding for their air monitoring programs to have a QA
program with certain elements including quality management plans (QMPs), QAPPs, and the
identification of a QA management function. It is acceptable for the state, local and/or Tribal
(SLT) air pollution agencies to use this document in lieu of writing their own field QAPP;
however, certain sections may have to be revised to make the information specific to the
monitoring program adopting it. In such case, this document should be modified by the
submitting SLT monitoring organization, and the EPA Regional Offices shall review and
approve the SLT's QAPP, and sign the approval page (see approval page Section 1.2), or if the
SLT has delegated approval authority through a Regionally approved Quality Management Plan,
the Region's signature represents acknowledgement that the QAPP has been "approved."
Note: At various points in this QAPP, the reader is referred to the EPA's AMTIC website for further
information and updated information.
-------
Mention of corporation names, trade names, or commercial products does not constitute
endorsement or recommendation for use.
The following EPA Office of Air Quality Planning and Standards (OAQPS) personnel may be
contacted concerning the contents of this document:
Dennis Grumpier, OAQPS-AAMG C304-06, Research Triangle Park, NC 27711
crumpl er. denni s@ epa. gov
David Shelow, OAQPS-AAMG C304-06, Research Triangle Park, NC 27711
shelow.david@epa.gov
Solomon Ricks, OAQPS-AAMG C304-06, Research Triangle Park, NC 27711
ricks.solomon@epa.gov
-------
Acronyms and Abbreviations
AAMG Ambient Air Monitoring Group
AMTIC Ambient Monitoring Technical Information Center
ANSI American National Standards Institute
APTI (EPA) Air Pollution Training Institute
ASQC American Society for Quality (Control)
AQS Air Quality System
AQAD Air Quality Analysis Division
CAA Clean Air Act
CASAC (EPA) Clean Air Scientific Advisory Committee
CFR Code of Federal Regulations
CMD (EPA) Contract Management Division
COC chain of custody
CSN (PM2.s) Chemical Speciation Network
CV coefficient of variation
DOPO (EPA) delivery order project officer
DQA data quality assessment
DQO data quality objective
EC/OC elemental carbon/organic carbon
EDXRF energy-dispersive X-ray fluorescence
EPA Environmental Protection Agency
FR Federal Register
FRM Federal Reference Method (for PM2.5 sampling)
1C ion chromatography
IMPROVE Interagency Monitoring of Protected Visual Environments (network or sampler)
ISO International Organization for Standardization
IUPAC International Union of Pure and Applied Chemistry
L/min liters per minute
LCD liquid crystal display
M2.5 mass of PM2.5
m3 cubic meter
MASS Mass Aerosol Speciation Sampler (URG Corp.)
max./min. maximum/minimum (thermometer)
mL milliliter
mmHg millimeters of mercury (pressure)
MQO measurement quality obj ective
MSA metropolitan statistical area
MSR management systems review
NAAQS national ambient air quality standard
NAMS national air monitoring station
NAREL National Air and Radiation Environmental Laboratory
NCORE National Core (monitoring network)
NERL (EPA) National Exposure Research Laboratory
NIST National Institute of Standards and Technology
OAQPS (EPA) Office of Air Quality Planning and Standards
in
-------
OAR (EPA) Office of Air and Radiation
ORD (EPA) Office of Research and Development
ORIA (EPA) Office of Radiation and Indoor Air
PAMS photochemical air monitoring station
PM2.5 participate matter, 2.5 micrometer diameter
PO project officer
PQAO primary QA organization
QA quality assurance
QAL quality assurance lead or leader
QAM quality assurance manager
QAPP Quality Assurance Proj ect Plan
Qavg average flow rate
QC quality control
QMP quality management plan
RAAS Reference Ambient Air Sampler (Thermo-Andersen Instruments, Inc.)
R&IE Radiation and Indoor Air
R&P Rupprecht and Patashnick Co. (No longer in existence)
RSC regional speciation coordinator
RTF Research Triangle Park (North Carolina)
SASS Speciation Air Sampling System (Met One Instruments, Inc.)
SHAL sample handling and archival laboratory
SLAMS state and local air monitoring station
SOP standard operating procedure
SPM (laboratory) services program manager
SLT state, local and/or Tribal agency or monitoring organization
STAG state and tribal assistance grants
SVOC semi-volatile organic compound
TSA technical systems audit
VOC volatile organic compound
XRF X-ray fluorescence
C degree Celsius
|j,g/m3 micrograms (of PM) per cubic meter (of air sampled)
IV
-------
Tables
Number Section Page
6-1 Critical Measurements in the PM2.5 CSN 6 3
6-2 Checklist of PM2.5 CSN Site Operator 6 4
Field Activities
6-3 Sequence of Activities for Bringing a 6 9
PM2.5 CSN Field Site On-Line
6-4 Critical Filter and Denuder Holding and 6 11
Use Times
7-1 MQOs for Total Measurement Error 7 4
7-1.a MQOs for Total Measurement Error (2007 7 4
- 2010 Data From Collocated Samplers
Upper Bound 90th Percentile CV From
Values > MDL
9-1 CSN Reporting Package Information 9 2
10-1 Sequence of Activities for CSN Contracted 10 1
Support Laboratory
10-2 Design Criteria for Collection Site 10 2
Surroundings
11-1 Summary of Information Provided by 11 2
Speciation Sampler
11-2 CSN Field Operations Corrective 11 6
Procedures
12-1 Explanation of CSN Custody and Field 12 6
Data Form
14-1 MQOs and Associated QC Activities for 14 3
the PM2.5 CSN
15-1 Testing and Acceptance Criteria Checklist 15 2
for PM2 5 Speciation Samplers
15-2 Preventive Maintenance and 15 3
Recertification of CSN Field Equipment
16-1 Acceptance Criteria and Calibration and 16 4
Maintenance Frequencies for PM2 5
Chemical Speciation Samplers
16-2 Calibration Standards for PM2.5 Chemical 16 7
Speciation Samplers
-------
17-1 Inventory List for CSN Field Equipment 17 2
and Supplies
19-1 Suggested Support Laboratory Data 19 10
Record Archival Summary
19-2 Validation Check Summaries 19 11
19-3 Data Transfer Operations 19 11
19-4 Summary of AQS Data Flags for PM2.5 19 13
19-5 Raw Data Calculations 19 16
20-1 Assessment Summary 20 2
21-1 Summary of Reports to Management for 21 2
CSN
22-1 Data Verification Activities and 22 2
Responsibilities for the CSN
22-2 Quality Control Data for CSN Data 22 7
Verification and Validation
22-3 Statistical Validation Limits for Blanks 22 13
22-4 Statistical Validation Limits for Routine 22 14
Data
22-5 Mapping of Outlier Flags onto AQS Codes 22 14
22-6 Automated Range Checks 22 14
22-7 Data Verification and Validation Summary 22 16
VI
-------
Figures
Number Section Page
4-1 Chemical Speciation Network Coordination Activities 4 2
4-2 Laboratory Technical Management and Staff Organization for 4 11
CSN Filter Analysis and Data Reporting
5-1 Interrelationships Between chemical Speciati on Networks as of 5 2
2011
6-1 Summary of the CSN Project Operations 6 2
6-2 Sample Analysis Delivery Order Process 6 7
6-3 Diagram of Laboratory Filter Processing and Analysis 6 8
Activities, by Filter Type
11-1 Quality Bulletin 11 5
12-1 CSN Custody and Field Data Form for MetOne SASS Sampler 12 5
12-2 CSN Custody and Field Data Form for URG3000N Sampler 12 6
14-1 Monthly Flow Check Decision Tree for Corrective Actions 14 8
14-2 CSN QA/QC Report Form 14 10
19-1 CSN Data Flow Overview 19 2
20-1 Surveillance Report Form 20 8
20-2 Assessment Finding Response Form 20 8
Vll
-------
PM2 5 CSN QAPP
Section No. 1
Revision No.: 1.2.0
Date: 05/12
Page 1 of 7
1.0 QA Project Plan Identification and Approval
Title: Quality Assurance Project Plan: PM2 5 Chemical Speciation Sampling at NCore
Sites, and State, local and Tribal Supplemental Sites
1.1 The EPA-Approved PM2 5 Chemical Speciation Network (CSN) QAPP
Category: Category 1
By signing below, the Environmental Protection Agency's (EPA's) Office of Air Quality Planning and
Standards' (OAQPS) Ambient Monitoring Program and Quality Assurance Leads and Managers have
found the underlying Quality Assurance Project Plan (QAPP) to provide sufficient detail in quality
assurance (QA) and quality control (QC) procedures, sampling and analytical procedures, and data
management as to be directly adoptable by reference by any state, local or Tribal (SLT) monitoring
agency for the purpose of conducting monitoring of chemically speciated particulate matter with a
aerodynamic diameter of 2.5 microns (PM2.s) and reporting same in units micrograms per cubic meter
of air sampled. Signature by a Regional representative, such as a QA manager or monitoring program
manager or their delegated designee, signifies concurrence with all aspects of this QAPP.
An example submittal form for adopting the "EPA-Approved QAPP version 1.2.0," or revising it and
acquiring EPA Regional approval or acknowledgement is provided in Section 1.2. Each monitoring
agency may choose to adopt this QAPP in its entirety, revise it as necessary for its specific needs, or
develop a completely new QAPP to better fit with the implementation of their quality system. In some
Regions, states have been delegated full signature authority for QAPP approvals through a priori
approval of their Quality Management Plans (QMPs). In this case, the SLT would have to sign the
signature page and the Region would retain a copy in their files.
In those Regions where final approval authority has not been delegated, the Regional Monitoring
Program and QA managers should approve any SLT adoption or revisions to the QAPP on the
signature page that is provided. In the latter case, the monitoring agency should submit the revised
QAPP or a list of modifications to the "EPA Approved" QAPP, to their respective EPA Regional
Monitoring Program Office with supporting documentation that the revisions fully meet the quality
system objectives set out in this network QAPP. When signed, these forms are to be kept in the
permanent records of the EPA Regional Office.
-------
PM2 5 CSN QAPP
Section No. 1
Revision No.: 1.2.0
Date: 05/12
Page 2 of 7
CSN QAPP Version 1.2.0 Approvals
OAQPS
Date
David
Shelow, CSN Program Lead
Date
Lewis
Weinstock, Group Leader, AAMG
Date
Dennis Grumpier, CSN QA Lead
Regional Concurrence/Approvals (See Attached pages)
Region 1
Date
Print name/title below
Region 3
Date
Print name/title below
Region 5
Date
Print name/title below
Region 7
Date
Print name/title below
Region 9
Date
Print name/title below
Region 2
Date
Print name/title below
Region 4
Date
Print name/title below
Region 6
Date
Print name/title below
Region 8
Date
Print name/title below
Region 10
Date
Print name/title below
-------
PM2 5 CSN QAPP
Section No. 1
Revision No.: 1.2.0
Date: 05/12
Page 1 of 7
PM, , CSN QAPP
Section No. 1
Revision No.: 1.2.0
Dale: OS.' 12
Page 1 of 7
CSN QAPP Version 1.2.0 Approvals
OAQPS
1 Date
1 David Shelow, CSN Program Lead
Date
Lewis Weinstock, Group Leader, AAV1G
Date
Dennis Crumpler, CSN QA Lead
Regional Concurrence/Approvals
Region 1
1 " / r,.i ., J\. to Date-'/' //•>
Print name'title below ,/
1 KV'-'" r .,. //. tC ,.t .'V _r-..,..,.| ., ;-!
Region 3
Date
Pnn! name'tltle below
1 Region 5
Date
1 Prin! nanie/Jitic below1
Region 7
Date
Print namc'tnic below
Region 9
1 Date
FnrH name^'titie below
Region 2
Date
Print name/title below
Region 4
Date
Print narne'ltUt; below
Region 6
Date
Print name/title below
Region 8
Date
Print name/title below
Region 10
Date
Print name-title below
r ,vj , , i .^;v i^rtrr
Scciion No I
Revision No.: 1.2.0
Date: 05' 12
Page 1 of?
CSN QAPP Version 1.2.0 Approvals
OAQPS
Date Dale
David Shelow, C.'SN I'rouram Lead Dennis Crumpler, C'SN QA Lead
Date
Lewis Wcinsux-'k. (iroup 1 .CiKlei. AAVKi
Regional Concurrence/Approvals
• Rcuion 1 Region 2
/? /^7 r\ /
Dale 7?&frJL-^/,tf%£ Hale him 4.2012
i'riol n.jnic'litSc ix'!o\s I rrin! mmic.'Utie below
1 i Mark Winter, Environmental Engineer
Region 3 i Region 4
Date i Dale
i Prim ituino'titlc tK-hn\ i Pfin! niiHlc'tide bcknv
Region 5 Region 6
1 )aie 1 )ate
Region 7 Region 8
Dale Date
I'rint iiitmciitlc bch>u Prsni nainc/tilSc bciow
Region 9 Region 10
-------
PM2 5 CSN QAPP
Section No. 1
Revision No.: 1.2.0
Date: 05/12
Page 2 of 7
PM;, CSN QAPP
Section No. i
Revision No.: 1.2.0
Dale: 05/12
Page I of 7
CSN QAPP Version 1.2.0 Approvals
OAQPS
Date
David Shelow
CSN Program Lead
Date
Lewis Weinstock. Group Leader. AAMG
Date
Dennis Grumpier, CSN QA Lead
Regional Concurrence/Approvals
Region 1
Date
Print name/title below
Region 3
Date
Print name/title below
Region 5
Date
Print name/title below
Region 7
Date
Print name/title below
Region 9
Date
Print name/title below
Region 2
Date
Print name/title below
Region 4
^lAA^Uck-C -"- Date 0(e 04/|;
liauru \\ckerman, SAS Chief
Region 6
Date
Print name/title below
Region 8
Date
Print name/title heluvv
Region 10
Dale
Print name/title below
PMU CSN QAPP
Section No. 1
Revision No.: 1.2.0
Date: 05/12
Page 1 of 7
CSN QAPP Version 1.2.0 Approvals
OAQPS
Date
David Shelow, CSN Program Lead
Date
Lewis Weinstock, Group Leader, AAMG
Date
Dennis Grumpier, CSN QA Lead
Regional Concurrence/Approvals
Region 1
Date
Print name/title below
Re»i«a-_j
Wi_7k /,/
yC^- / W-r- eJL Date (e/S/13-
Kia Hence - 'QA Coordinator, Air Protection Division
Region 5
Date
Print name/title below
Region 7
Date
Print name/title below
Region 9
Date
Print name/title below
Region 2
Date
Print name/title below
Region 4
Date
Print name/title below
Region 6
Date
Print name/title below
Region 8
Date
Print name/title below
Region 10
Date
Print name/title below
-------
PM2 5 CSN QAPP
Section No. 1
Revision No.: 1.2.0
Date: 05/12
Page 3 of 7
PM2, CSN QAPP
Section No. 1
Revision No.: 1.2.0
Dale: 05/12
Page 1 of 7
CSN QAPP Version 1.2.0 Approvals
OAQPS
Date
David Shelow, CSN Program Lead
Date _
Dennis Grumpier, CSN QA Lead
Date
Lewis Wcinstock, Group Leader, AAMG
Region I
Print name/title below
Region 3
Print name/title below
Region 5
Region 7
Regional Concurrence/Approvals
Region;
Date
Print narne'Eitle below
Region. 4
Date
Print iiaincftillc below
Region 6
Print namc/title below
Region
Date
Date
Date
CSN QAPP Version 1.2.0 Approvals
OAQPS
PM,,CSN QAPP
Section No. I
Revision No : 1.2.0
Date: OS/12
Page 1 of"
Date-
David Shelow, CSN Program Lead
Date
Lewis Wcinstock
Group Leader. AAMG
Date
Dennis Grumpier, CSN QA Lead
Regional Concurrence/Approvals
Region i
Region 2
Date
Print name/title below
Date
Prim name/tilic below
Region 3
! Region 4
Dale
Print name/title below
Region 5
Prim name/title below-
Date
Region 6
Date
Print name/title below
Region 7
Date
Print name/title below
-'. • , Date
Print name tll'c Mow
Maria L. Martinez/Chief, Air Quality Analysis Section
Region 8
Date
Print name/title below
Region 9
Region 10
Date
Print name.'title below-
Print name title below
Dale
-------
PM2 5 CSN QAPP
Section No. 1
Revision No.: 1.2.0
Date: 05/12
Page 4 of 7
PMj 5 CSN QAPP
Section No. I
Revision No.: 1.2.0
Date: 05/12
Page I of?
CSN QAPP Version 1.2.0 Approvals
OAQPS
Date
David Shelow, CSN Program Lead
Date
Lewis Weinstock,
Group Leader, AAMG
Date
Dennis Crumpler, CSN QA Lead
Regional Concurrence/Approvals
Region 1
Date
Print name/title below
Region 3
Date
Print name/title below
Region 5
Date
Print name/title below
Region/*"1^
Q^K0}^—- <&/<&/lL
' V ^-/&- Date
Print name 'title below
/MICHA&- FDAVi'*< CARb favctGitep
Region 9
Date
Print name/title below
Region 2
Print name/title below
Region 4
Print name/tHle below
Region 6
Print name/title below
Region 8
Print name/title below
Region 10
Print name/title below
Date
Date
Date
Date
Date
CSN QAPP Version 1.2.0 Approvals
OAQPS
_ _ _ Date.
Das id Shelow, CSN Program Lead
Bait-
Lewis Weinstoek. Group Leader, AAMG
Dale _
Dennis Crumpler. CSN QA Lead
Regional Concurrence/Approvals
Reuion 2
Print name title below
Reuion 3
1_ _
Region 4
Region 5
Reuion 7
Date-
Date
Dale
Date
Date
Region 6
Print name.'title below
Reuion 9
Prim name.tit!e bekm
Carl Daly
Director, Air Program
Region 10
Print name title beiou
(/
Date
Date'
Date
-------
PM2 5 CSN QAPP
Section No. 1
Revision No.: 1.2.0
Date: 05/12
Page 5 of 7
PM2.s CSN QAPP
Section No. I
Revision No.: 1.2.0
Date: 05/12
Page I of 7
CSN QAPP Version 1.2.0 Approvals
OAQPS
Date
David Shelow, CSN Program Lead
Date
Lewis Weinstock, Group Leader, AAMG
Date
Dennis Crumpler, CSN QA Lead
Regional Concurrence/Approvals
Region 1
Date
Print name/title below
Region 3
Date
Print name/title below
Region 5
Date
Print name/title below
Region 7
Date
Print name/title below
Region 9_/~^__x ~n . /
V^^VRV-U^Jfo Date t/4/>2.
ff
Print name/title below
fa EU^e^c.^ ;«^Wa^k~h:^-i
Region 2
Date
Print name/title below
Region 4
Date
Print name/title below
Region 6
Date
Print name/title below
Region 8
Date
Print name/title below
Region 10
Date
Print name/title below
CSN QAPP Version 1.2.0 Approvals
OAQPS
Date
Da\jd Shelow, CSN Program Lead
Date _
I Dennis Grumpier. CSN QA Lead
Date
Lewis Weinslock. Group Leader, AA.MG
Regional Concurrence/Approvals
Region 1
Region 2
Date
Dale
Region 3
Region 4
Date
Region 5
Date
Region "
Riuon 8
Date
Date
Date
Region 9
Region 1Q/
Date
-------
PM2 5 CSN QAPP
Section No. 1
Revision No.: 1.2.0
Date: 05/12
Page 1 of7
1.2 Adoption of the EPA-Approved CSN QAPP Revision 1.2.0 or Adoption of a
Revision of the CSN QAPP by a Monitoring Agency (This page is retained by the
EPA Regional Office)
By signing below, the Monitoring Agency commits to implement either: (check one)
The CSN Quality Assurance Project Plan Revision 1.2.0, dated approved by EPA; or
The CSN Quality Assurance Project Plan submitted herein, with all
Agency Name
deviations from the "EPA-Approved CSN QAPP Revision 1.2.0" approved by EPA clearly
identified, and documentation supporting the proposition that all revisions provide equivalent or
better quality monitoring data.
Delegated QAPP Approval Authority
The has been delegated full QAPP Approval and signature authority
Agency Name
by EPA Region .
State/Local/Tribal Agency
Print Name: Signature:
Date: PM2.5 Speciation Monitoring Coordinator
Print Name: Signature:
Date: Quality Assurance Officer
Print Name: Signature:
Date: Air Program Director
EPA Regional Office Acknowledgement or Approval
Print Name: Signature:
Date: PM2.5 Speciation Monitoring Coordinator
Print Name: Signature:
Date: Quality Assurance Manager
Print Name: Signature:
Date: Grants, Audits & Procurement Program Director
-------
PM2 5 CSN QAPP
Section No. 2
Revision No.: 1.2.0
Date: 05/12
Page 1 of 7
2.0 Table of Contents
Section
Foreword
Acknowledgments
Acronyms and Abbreviations
Tables
Figures
Page Revision Date
i
ii
iii
vi
viii
1.2.0
1.2.0
1.2.0
1.2.0
1.2.0
10/11
10/11
10/11
10/11
10/11
1.0 QA Proj ect Plan Identification and Approval
1/1
1.2.0 10/11
2.0
Table of Contents
1/7
1.2.0 10/11
3.0
Distribution
1/4
1.2.0 10/11
4.0 Project/Task Organization
4.1 Chemical Speciation Network Coordination
Activities
4.2 Chemical Speciation Network — State, Local, and
Tribal Organizations
4.3 CSN Network Laboratory Activities Organization
1/10 1.2.0 10/11
2/10
7/10
8/10
5.0 Background and Problem Definition
5.1 Background
Problem Definition
5.2
5.3
PM2.5 Speciation Sampling Techniques and Ongoing
Research
5.4 Monitoring Network Design Considerations
1/3
1/3
1/3
2/3
3/3
1.2.0 10/11
6.0 Project/Task Description
6.1 Description of Work to be Performed
6.2 Field Activities
1/11
1/11
3/11
1.2.0 10/11
-------
Section
6.3 Laboratory Activities
6.4 Schedule of Activities
6.5 Project Assessment Techniques
6.6 Proj ect Records
6.7 References
PM2 5 CSN QAPP
Section No. 2
Revision No.: 1.2.0
Date: 05/12
Page 2 of 7
Page Revision Date
4/11
8/11
11/11
11/11
11/11
7.0 Quality Obj ectives and Criteria For Measurement Data
7.1 Data Quality Obj ectives Process
7.2 Development of DQOs for the PM2.5 Chemical
Speciation Network
7.3 Measurement Quality Objectives
7.4 References
1/4
1/4
1/4
2/4
4/4
1.2.0 10/11
8.0 Special Training Requirements/Certification
8.1 Training
8.2 Certification
1/3
1/3
3/3
1.2.0
10/11
9.0 Documentation and Records 1/6
9.1 Information in the Management and Organization 1/6
Reporting Package
9.2 Information in the Field Operations Reporting 2/6
Package
9.3 Information in the Laboratory Operations Reporting 4/6
Package
9.4 Information in the QA Reporting Package 4/6
9.5 Reports to Management 5/6
9.6 Archival and Retrieval of Data Reporting Packages 6/6
1.2.0 10/11
10.0 Sampling Process Design 1/4
10.1 Scheduled Proj ect Activities, Including Management 1/4
Activities
10.2 Rationale for the Design 2/4
10.3 References 4/4
1.2.0 10/11
-------
PM2 5 CSN QAPP
Section No. 2
Revision No.: 1.2.0
Date: 05/12
Page 3 of 7
Section
Page Revision Date
11.0 Sampling Methods Requirements 1/8
11.1 Sample Collection and Preparation 1/8
11.2 Sampling/Measurement System Corrective Action 3/8
Process
11.3 Avoiding Sample Contamination; Temperature and 4/8
Holding Time Requirements
1.2.0 10/11
12.0 Sample Handling and Custody Requirements 1/6
12.1 Introduction 1/6
12.2 Filter Handling and Custody Procedures Prior to 1/6
Sampling Event
12.3 Sample Handling and Custody Procedures 3/6
12.4 Filter and Sample Archival in the CSN Laboratory 5/6
1.2.0 10/11
13.0 Analytical Methods Requirements
1/1
1.2.0 10/11
14.0 Quality Control Requirements
14.1 Quality Control Checks
14.2 QC Samples
14.3 Collocated Samplers
14.4 Calculations of Accuracy, Bias, Precision, and
Completeness
14.5 References
1/11
3/11
3/11
5/11
5/11
11/11
1.2.0 10/11
15.0 Sampler/Equipment Testing, Inspection, and Maintenance 1/4
Requirements
15.1 Testing and Acceptance Criteria 1/4
15.2 Maintenance 1/4
15.3 Critical Spare Parts 3/4
1.2.0 10/11
16.0 Instrument Calibration and Frequency
16.1 Overview
1/6
1/6
1.2.0
10/11
-------
PM2 5 CSN QAPP
Section No. 2
Revision No.: 1.2.0
Date: 05/12
Page 4 of 7
Section
16.2 Calibration and Verification of Field Instrumentation
16.3 Calibration and Verification of Laboratory
Instrumentation
Page Revision Date
1/6
3/6
17.0 Inspection/Acceptance for Supplies and Consumables 1/2
17.1 Purpose 1/2
17.2 Critical Supplies and Consumables for Field Site 1/2
Operations
17.3 Acceptance Criteria 2/2
1.2.0 10/11
18.0 Data Acquisition Requirements (Nondirect Measurements) 1/2
18.1 Acquisition of Nondirect Measurement Data 1/2
1.2.0 10/11
19.0 Data Management 1/16
19.1 Overview 1/16
19.2 Data Management Activities at the CSN Contracted 3/16
Support Laboratory
19.3 Data Management Activities at the RO 3/16
19.4 Recommended Data Management Practices 6/16
19.5 Data Validation 8/16
19.6 Data Transformations 14/16
19.7 Data Transmittal 14/16
19.8 Data Reduction 14/16
19.9 Data Analysis 15/16
19.10 Data Storage and Retrieval 15/16
1.2.0 10/11
20.0 Assessment and Response Actions
20.1 Types of Assessments
20.2 Assessment Frequency
20.3 Acceptance Criteria
20.4 Assessment Schedules
20.5 Assessment Personnel
1/9
1/9
1/9
2/9
3/9
3/9
1.2.0 10/11
-------
PM2 5 CSN QAPP
Section No. 2
Revision No.: 1.2.0
Date: 05/12
Page 5 of 7
Section
20.6
20.7
20.8
Assessment Reports
Implementation of Response Actions
References
Page
5/9
6/9
9/9
Revision Date
21.0 Reports to Management 1 /5
21.1 Annual QC Summary Report (Laboratory) 3/5
21.2 Annual QA Report to Management (Network 3/5
Review)
21.3 Precision, Bias, and Accuracy Quarterly Reports 3/5
21.4 Laboratory Performance Evaluation and Technical 4/5
Systems Audit Results
21.5 Site TSAs (External) 4/5
21.6 Routine Quality Control Records 4/5
21.7 Data Validation Summaries 4/5
21.8 Corrective Action Reports and Performance-Related 5/5
Records
1.2.0 10/11
22.0 Data Review, Validation, and Verification Requirements 1/16
22.1 Data Verification and Validation Responsibilities 1/16
22.2 Corrective Action Reporting Process 6/16
22.3 Use of QC Information for Verification and 6/16
Validation
22.4 Use of Calibration Information for Verification and 6/16
Validation
22.5 Level 0 Verification and Validation 9/16
22.6 Level 1 Data Validation 10/16
22.7 Data Screening Techniques used by the Contracted 11/16
Support Laboratory
22.8 Treatment of Deviations from Requirements 14/16
22.9 Verification and Validation Criteria: Field 14/16
Component
1.2.0 10/11
23.0 Validation and Verification Methods
1/5
1.2.0 10/11
-------
Section
23.1 Inter-organizational Responsibilities for Data
Validation
23.2 Personnel Responsibilities Within the Reporting
Organization
23.3 Completion of Level 0 Data Verification and
Validation
23.4 Identification of Outliers and Data Flagging
Techniques
PM2 5 CSN QAPP
Section No. 2
Revision No.: 1.2.0
Date: 05/12
Page 6 of 7
Page Revision Date
1/5
2/5
2/5
3/5
24.0 Reconciliation with Data Quality Objectives
24.1 DQO for Chemical Speciation Trends
24.2 Interim Evaluations of Data Quality
24.3 Assessing and Reporting Chemical Speciation
Trends
24.4 Reconciling Other CSN Research Objectives
24.5 References
1/7
1/7
2/7
6/7
7/7
7/7
1.2.0 10/11
-------
PM2 5 CSN QAPP
Section No. 2
Revision No.: 1.2.0
Date: 05/12
Page 7 of 7
Appendix A
Section Page Revision Date
A-l SOP: CSN Standard Operating Procedure for Field 1.2.0 10/11
Installation of PM2.5 Speciation Samplers
A-2 SOP: CSN Condensed Standard Operating Procedure for 1.2.0 10/11
the MetOne SASS
A-3 SOP: CSN Standard Operating Procedure for the URG 1.2.0 10/11
3000N Carbon Speciation Sampler
A-4 SOP: CSN Packing Instructions for the Speciation Sampler 1.2.0 10/11
Modules
A-5 XL Worksheets and Instructions for Recording and 1.2.0 10/11
Reporting CSN Sampler Performance Verifications
A-6 XL Worksheets and Instructions for Recording and 1.2.0 10/11
Reporting TSA's of CSN Monitoring Sites Including
Sampler Performance Audits
Note: Standard Operating Procedures (SOPs) for the Andersen RAAS and the URG MASS 400
and MASS 450 speciation samplers (used in the CSN until 2009) are not included as appendices.
None of these samplers remain in the CSN network and vendor support has been suspended.
SOPs for these samplers are available on the AMTIC website in the December 2000 QA
Guidance Document, EPA 454/DO1-001.
-------
PM2 5 CSN QAPP
Section No. 3
Revision No.: 1.2.0
Date: 11/11
Page 1 of 1
3.0 Distribution
An electronic version of this Quality Assurance Project Plan (QAPP) will be available on the
Ambient Monitoring Technical Information Center (AMTIC) Speciation web page at
http://www.epa.gov/ttn/amtic/specguid.html. Availability of the QAPP will be announced on the
Monitoring List Serve maintained by the Ambient Monitoring Program within the
Environmental Protection Agency's (EPA) Office of Air Quality Planning and Standards
(OAQPS) at Research Triangle Park, North Carolina. The OAQPS will email a copy to the
Regional Speciation Coordinators (RSCs) at the National Contract Laboratory, and the EPA's
Office of Radiation and Indoor Air (ORIA) National Air and Radiation Environmental
Laboratory (NAREL), Montgomery, Alabama. The RSCs will be responsible for ensuring that
the QAPP is distributed through state, local, and/or Tribal (SLT) agency Monitoring Program
Manager or Director of each field site and analytical laboratory (if different than the national
contract laboratory) in the environmental data operations of the Chemical Speciation Network
(CSN). The RSCs should also provide a copy of this QAPP to their Regional Quality Assurance
(QA) Managers.
Required Distribution
• Regional RSCs
o SLT Agency Monitoring Program Manager or Director
• Every monitoring site and site supervisor
o Analytical support laboratories for specific SLTs who have opted to analyze their
own filters or who have contracted a laboratory who is not the national contract
laboratory
• Delivery Order Project Officers
• Regional QA Managers
• National Contract Laboratory
-------
PM2 5 CSN QAPP
Section No. 4
Revision No.: 1.2.0
Date: 01/12
Page 1 of 10
4.0 Project/Task Organization
This section is intended to provide all parties involved in the Chemical Speciation Network
(CSN) with a clear understanding of their roles and the lines of authority, communication, and
reporting for the project.
The organization of the CSN involves multiple interacting entities:
(1) The EPA's Office of Air Quality Planning and Standards (OAQPS), Office of Radiation and
Indoor Air (ORIA), Office of Research and Development (ORD), and Regional Offices; (2)
state, local and/or Tribal (SLT) air monitoring and reporting organizations; and (3) one or more
contracting laboratories to service monitoring sites with supplies, sample analysis, and data
reporting. To make the best use of available resources and to meet demanding timelines for
collection and analysis of samples, the flow of information and samples must be optimally
organized. The deployment and operation of the network are responsibilities shared among all
the involved parties. This section describes the roles of all parties and establishes the lines of
communication and reporting, with the goal of facilitating a smoothly operating network.
Figure 4-1 provides a unified view of all agencies and contractors involved in the network's
operating structure as well as lines of communication and reporting. Block A of Figure 4-1
points out the CSN's coordination activities, led by EPA's OAQPS. The Program Quality
Assurance (QA) Manager, the network's Project Officer (PO), the EPA-Regional Speciation
Coordinators (RSCs), and Delivery Order Project Officers (DOPOs), are also included in the
network's planning and coordination activities component. Block B of Figure 4-1 shows the field
activities, the SLT agency, CSN sampling, and QA/Quality Control (QC) personnel will perform
and their lines of communication to the DOPOs, the CSN PO, and the contracted support
laboratory. Block C of Figure 4-1 shows the contracted support laboratories. The EPA ORIA
technical assistance and QA assessment laboratories are included in Blocks B and C.
The Speciation Contacts Table found on the Ambient Monitoring Technical Information Center
(AMTIC) website provides the current names and contact information for the persons in each of
these positions. Sections 4.1 through 4.3 discuss details of the roles and interactions between and
among the many participants in this maze of activity.
-------
PM2 5 CSN QAPP
Section No. 4
Revision No.: 1.2.0
Date: 01/12
Page 2 of 10
EPA PM2.s Chemical Speciation
Workgroup
OAQPS Group Leader
Regional Office
- Regional Speciation
Coordinators (RSCs)
- Delivery Order Project Officers
(DOPOs)
Program Manager
PM2.s Laboratory Services
Contracts Project Officer (PO)
Advisors
- CASAC
- ORD/Panel NERL
Program QA Manager or Lead
Block A
\ /
CSN Field Activities
State. Local, Tribal Agencies
-!j|&Cl!
flip
Field QA Assessments
ORIA, Las Vegas, NV
Block
CSN Laboratory Support Activities
01,
Chemical Speciation Contracted
Support Laboratories (CSLs)
Block C
Lab QA Auditing and
Technical Assistance
ORIA, Montgomery, AL
Figure 4-1. Chemical Speciation Network Coordination Activities.
4.1 Chemical Speciation Network Coordination Activities
The coordination of CSN network activities is accomplished by several groups, as illustrated in
Figure 4-1. Block A of this figure includes the PM2.5 Chemical Speciation Workgroup, the
OAQPS Program Manager for the CSN advisory groups (Clean Air Scientific Advisory
Committee (CASAC) and ORD), the Program QA Manager, and the Laboratory Services
Contracts PO, the RSCs, and the DOPOs. These groups assist in coordinating, advising,
planning, and managing the activities of the CSN field and laboratory activities. Roles of the
-------
PM2 5 CSN QAPP
Section No. 4
Revision No.: 1.2.0
Date: 01/12
Page 3 of 10
workgroup, advisory panels and committees, and the QA lead person are described below. Not
all committees and personnel are shown in Figure 4-1.
4.1.1 PMi.5 Chemical Speciation Workgroup
The PM2.5 Chemical Speciation Workgroup was formed at the inception of the CSN in 1998 to
provide input and the review of all aspects of the chemical speciation program, which
encompasses sampler design, network design and QA procedures. Members of the workgroup
included personnel from EPA OAQPS, EPA Regional Offices, the EPA ORD's National
Exposure Research Laboratory (NERL), and SLT air monitoring organizations. This workgroup
is currently inactive but may be reconvened if critical issues needing broad technical and policy
input arise.
4.1.2 Office of Air Quality Planning and Standards Ambient Monitoring Program
The OAQPS Ambient Monitoring Program has oversight concerning design of the nation's
ambient monitoring networks and the quality of the nation's ambient air data these networks
generate. The OAQPS has published specific regulations for the development of a quality system
which are found in 40 Code of Federal Regulations (CFR) Part 58, Appendix A. One specific
element of this quality system is the development and coordination of the PM2.5 ambient
monitoring networks, one of which is the CSN. The OAQPS will ensure the orderly operation of
the CSN through the following activities:
• Coordinating and overseeing the Chemical Speciation Trends Sampling at NCore
sites which are part of the CSN.
• Working with the EPA Regional Offices and SLT organizations to determine the best
sampling locations.
• Providing a contract vehicle for one (or more) laboratories to support the field sites
and provide filter analysis and associated functions.
• Developing documents for the chemical speciation monitoring network, CSN,
including the strategic plan for the quality system, standard operating procedures
(SOPs) for field sites and laboratory operations, and the QAPP for the NCore
chemical speciation sites.
• Providing, through this document, a model QAPP for state and local air monitoring
station (SLAMS) and Tribal Networks to adapt and adopt.
• Developing guidance for field and laboratory personnel requirements as well as
training activities.
• Securing national experts and advisors to answer specific technical questions and
review the network; responding to recommendations provided by national experts and
advisory committees.
• Assessing the species concentration information entered into the Air Quality System
(AQS) database.
-------
PM2 5 CSN QAPP
Section No. 4
Revision No.: 1.2.0
Date: 01/12
Page 4 of 10
• Developing an information management system and other means to archive data,
assess data sets, and release trends information to stakeholders, data users, and the
general public by use of the EPA's AMTIC QA website.
• Interacting with Regional, SLT agency personnel and the contracted support
laboratory concerning the setup and operation of the chemical speciation networks
and their data results.
• Ensuring the success of the network by coordinating quality assurance oversight
activities such as management systems reviews (MSRs) and technical systems audits
(TSAs).
Most budgetary and technical planning activities will be coordinated through the OAQPS. The
Ambient Air Monitoring Group (AAMG) within the Air Quality Analysis Division (AQAD) is
ultimately responsible for facilitating the implementation of the CSN. These responsibilities
include most of the technical components (with support from ORD, Regional Offices and
laboratories, and SLT agencies) and the resource estimates underlying program implementation.
The CSN Program Lead and QA Lead provide guidance and oversight for the development of
the CSN quality system. They also oversee the periodic review and revision of this QAPP.
Resource guidance necessary for the State and Tribal Assistance Grants (STAG) distribution is
coordinated through core staff within OAQPS. In addition, the EPA Stationary Source
Compliance Division (SSCD) is responsible for the AQS data management system.
4.1.2.1 The PM2.5 Chemical Speciation Network Quality Assurance Lead (QAL)
The CSN-QAL, at OAQPS, is responsible for reviewing and assessing the quality system of the
CSN. The QAL will:
• Review the national network's QAPP and other quality-related documents and
coordinate their approval.
• Ensure that contracted support laboratory's SOPs are reviewed and updated as
required.
• Coordinate with the QA Workgroup, the Regional QA laboratories, ORIA, and others
to ensure that periodic systems reviews and performance reviews of the field and
laboratory activities are conducted and completed.
4.1.2.2 The PM 2.5 Chemical Speciation Network Program Lead
The OAQPS Program Lead for the CSN is responsible for the activities that are implemented as
part of normal data collection activities. His/Her responsibilities include:
• Communication with EPA POs and EPA QA personnel on issues related to routine
sampling and QA activities;
• Understanding EPA monitoring and QA regulations and guidance, and ensuring all
key personnel understand and follow these regulations and guidance;
-------
PM2 5 CSN QAPP
Section No. 4
Revision No.: 1.2.0
Date: 01/12
Page 5 of 10
• Viewing acquisition packages (contracts, grants, cooperative agreements, inter-
agency agreements) to determine the necessary QA requirements;
• Developing budgets and providing program costs necessary for EPA allocation
activities;
• Ensuring that all personnel involved in environmental data collection have access to
any training or QA information needed to be knowledgeable of QA requirements,
protocols, and technology;
• Recommending management-level corrective actions.
4.1.3 Advisory Panels
Three advisory panels will consult with OAQPS on technical matters related to the CSN in its
initial development and deployment. These panels were the CASAC, the Expert Panel on
speciation, and a panel of ORD/NERL technical experts.
CASAC—This committee is a subcommittee of EPA's Science Advisory Board (SAB) and
reports directly to the SAB administrator. It serves as the principal review body for the PM2.5
monitoring program and continues to review technical issues associated with operation of the
chemical speciation network. A third level subcommittee, the Air Monitoring subcommittee,
interacts with the OAQPS and, in particular, with the AAMG on all technical and policy issues
associated with ambient monitoring.
Expert Panel on Speciation—The Expert Panel on Speciation was formed during the initial years
of deployment to advise OAQPS on structural and technical matters related to the PM2.5
speciation network. It was composed of recognized technical experts in the fields of network
monitoring strategy, sampling and monitoring methods for PM2.5, physical and chemical
characterization of fine particles, and data analysis and interpretation. External technical review
and input will continue to be provided through the CASAC.
Office of Research and Development—The NERL provided expertise for helping to review the
original requirements for the speciation samplers and the analytical methodologies selected for
quantifying species. This office continues to provide advice on sampling technology, and Federal
Reference and "Equivalent" methods.
4.1.4 Chemical Speciation Network Contracts Project Officer
The CSN Laboratory Services Contracts PO is on the Central Operations and Resources (CORE)
staff of EPA/OAQPS. He/She is the liaison between the EPA Contracts Management Division,
the various contractors and vendors, other program staff in OAQPS related to PM 2.s Chemical
Activities, the DOPOs, and the contract laboratory's Services Program Manager. The primary
role of the PO is to approve and forward delivery orders prepared by the DOPOs to the lab
services contractor and ensure contract details are followed, including the submittal and review
of required draft and final annual data summary reports from the contracted laboratory.
-------
PM2 5 CSN QAPP
Section No. 4
Revision No.: 1.2.0
Date: 01/12
Page 6 of 10
4.1.5 Regional Speciation Contacts
The PM2.5 RSC in each Region interacts with the SLT agencies and provides assistance on field-
related network design and QA issues. The Regional contacts also assist SLTs with procuring
special equipment or non-routine analyses. The names of the RSCs are listed in a table of CSN
contacts on the AMTIC website at
4.1.6 Delivery Order Project Officer
The DOPO receives PM2.5 speciation analytical orders directly from SLT agencies and will
consolidate the requests for sampling media and sample analysis received from the RSCs. A
DOPO is assigned to the East (Regions 1-4 and Puerto Rico), Midwest (Regions 5-7), and West
(Regions 8-10) sections of the United States. The DOPO will communicate with the CSN
Contracts PO and with the contracted support laboratory for analysis of filters and other
sampling media. The DOPO will also review the monthly analytical data packages from the
contracted support laboratory, ensure they are complete and, after approval, make arrangements
for payment of the invoice for the various delivery orders. The names of the DOPOs are listed in
a table of CSN contacts on the AMTIC website at
4.1.7 EPA Contracts Management Division
The Contracts Management Division (CMD) is located within the Office of Acquisition
Management (OAM). The CMD is responsible for all communications with vendors and
extramural contract organizations. For the CSN, the CMD will:
• Provide a Contracts PO to represent the government;
• Award national contracts for filter purchases, sample shipping, and laboratory support
and analyses of speciation sampler filters;
• Approve work assignments, delivery orders and task orders written by OAQPS or, in
some cases, EPA Regional Offices, for activities supporting the operation and QA
programs for the CSN;
• Communicate with OAQPS to provide the above services.
4.2 Chemical Speciation Network—State, Local and Tribal Organizations
Personnel from SLT agencies (Block B of Figure 4-1) install speciation samplers atNCore sites
as well as at other SLAMS sites, operate the speciation samplers at the sites, and recover and
ship samples to the contracted support laboratory by predetermined schedules. It is the
responsibility of the SLT personnel to implement quality control (QC) procedures as given in the
SOPs for sampler operation and sample shipping. The SLT personnel will consult with the
DOPO and the RSC regarding issues concerning sampling equipment and laboratory supply or
sample analysis. Interested SLT personnel may also participate in Chemical Speciation
Workgroup discussions.
The SLT site operators and supervisors should communicate directly with the contracted support
laboratory only when there are concerns about timely shipment and receipt of supplies, sampling
-------
PM2 5 CSN QAPP
Section No. 4
Revision No.: 1.2.0
Date: 01/12
Page 7 of 10
media, and data packages. Field sites will also communicate with OAQPS and with the EPA
Regional QA laboratories for technical assistance on QA issues. The SLT site operations and
records will be subject to systems and performance reviews. The EPA Regional CSN contacts
are expected to conduct full program and technical systems audits of every SLT monitoring
program at least once every 3 years.
4.2.1 Organization for Routine Field Sampling Operations
The SLT agencies are responsible for day-to-day operation of CSN sites. The management and
technical organizations already in place at these agencies will be used to implement operations.
Personnel will need to be assigned and organized to accomplish the following tasks, among
others:
• Site selection and platform and utility installations and maintenance.
• Ensure these data are posted in proper fields of AQS for QA reports.
• Purchase, receipt, acceptance testing, and installation of sampler, calibration
equipment, and meteorological equipment and maintenance/replacement of the above.
• Operator hiring and training.
• Communication with the RSC and DOPO regarding sample analysis task orders.
• Scheduled operation of the sampling site, including internal QA/QC activities.
• Monthly validation of draft data sets received from the contract laboratory within
specified time constraints.
• Interactions with CSN network management personnel.
• Participation in external QA activities such as TSAs and site audits.
4.2.2 QA/QC Organization for State, Local and Tribal Agency Field Site Operations
Quality assurance activities supporting the CSN sites will be arranged through the QA Manager
at each SLT agency. The QA personnel will be identified prior to field data collection and will
be assigned and organized to accomplish the following tasks, among others:
• Implementation of the quality system for the CSN.
• Review and approval of the network's field QAPP or acceptance of this QAPP as
the field QAPP.
• Site inspections (and/or audits) and review of procedures to ensure specified
QA/QC checks are being made and measurements systems are in control.
• Email prescribed audit reports to the support contractor for posting flow audit data
on AQS and other audit results in the speciation audit data-base.
• Issuance of corrective action memoranda and monitoring of follow-up actions.
• Participation in monthly validation of draft data sets received from the contracted
support laboratory within specified time constraints.
-------
PM2 5 CSN QAPP
Section No. 4
Revision No.: 1.2.0
Date: 01/12
Page 8 of 10
• Arrangement for and participation in QA activities called for by EPA Regional
Offices.
4.3 CSN Network Laboratory Activities Organization
Block C of Figure 4-1 illustrates the organization of the CSN laboratory activities. Laboratory
activities include QA-related work by the EPA ORIA and the contracted support laboratory
which provides field site filter supply and filter analysis services for the NCore (previously the
Speciation Trends Network) sites. There are other service laboratories contracted by individual
SLT agencies that are part of the chemical speciation program; these are also audited by the
EPA.
4.3.1 EPA QA Laboratories
Two of the EPA's ORIA laboratories work with OAQPS to provide QA oversight of CSN field
and laboratory activities. These two laboratories are the National Air and Radiation
Environmental Laboratory (NAREL) in Montgomery, Alabama, and the Radiation and Indoor
Air (R&IE) laboratory in Las Vegas, Nevada. These laboratories will:
• Provide speciation laboratory QA support to the state and local agencies and to the
contracted support laboratories.
• Provide QA auditing and technical assistance to the field sites and to the contracted
support laboratory.
• Assist OAQPS in development of training materials and staffing training activities.
• Communicate with OAQPS regarding QA issues in performance of the Contract
Laboratory and field operations of the CSN (and supplemental networks). The
OAQPS will in turn communicate ORIA's findings on the performance of the
network to the PO and the RSCs.
4.3.2 Contracted Support Laboratory Organization
Figure 4-2 illustrates the typical organization of management, QA, and technical staff for the
PM2.5 contracted support laboratory.
Management Organization—The PM2.5 contracted support laboratory is headed by a Laboratory
Services Program Manager and a Deputy Manager. The Laboratory Services Program Manager:
• Receives and acts on site servicing delivery orders from the EPA DOPOs.
• Is responsible for returning analyzed filters to SLT agencies that request them.
• Oversees activities of the technical laboratories and the data management office.
• Receives and responds to quality systems findings provided by the contracted support
laboratory QA Manager and staff as well as to findings in external audit reports.
Quality Assurance Organization—The contracted support laboratory's QA Office is staffed by a
QA Manager, a Deputy QA Manager, and support staff. The QA Manager interacts with each of
-------
PM2 5 CSN QAPP
Section No. 4
Revision No.: 1.2.0
Date: 01/12
Page 9 of 10
the technical area laboratories or offices to conduct scheduled and follow-up systems and
performance evaluations (PEs). Findings from internal QA activities are reported to the
laboratory services program manager for review and action. This office summarizes QA
activities for inclusion in monthly and annual reports to EPA. The QA Office also assists in the
preparation and update of the laboratory QAPP and associated SOPs.
Technical Organization—The present contracted support laboratory is composed of seven
technical components that provide the following analytical services and data report packages to
CSN sites:
• Sample Handling and Archival Laboratory (SHAL). Supplies each CSN site with
all necessary sampling supplies to include coated denuders, sampling filter media,
assembled sampling modules, shipping containers, and documentation paperwork.
(The hardware is purchased by the state or local organization; the contractor will
supply only the filters.)
• Gravimetric and Microscopy Laboratories PM2s mass by microbalance
determination.
• Elemental Analysis Laboratory. Elements and metals by energy-dispersive X-ray
fluorescence (EDXRF) determination and other methods such as ICP-AES.
• Cations/Anions Laboratory. Cations and anions by ion chromatography (1C)
analysis.
• Carbon Species and Semi-volatile Organic Carbon Compounds Laboratory
Carbon species by elemental carbon/organic carbon (EC/OC) analysis.
Determination of semi-volatile organic compounds (SVOC's) as requested.
• Denuder Refurbishment Laboratory. Denuder refurbishment with magnesium
oxide or other prescribed coating material.
• Data Management Office. Track sampling media components and manage all data.
This includes issuance of monthly and quarterly data sets to SLT agencies and to the
EPA DOPOs and entry of validated data to the AQS database system.
Details on these CSL activities can be found in the laboratory QAPP and associated SOPs for
this program, which are available on the AMTIC website.
-------
PM2 5 CSN QAPP
Section No. 4
Revision No.: 1.2.0
Date: 01/12
Page 10 of 10
Subcontractors
Technical Advisors
Consultants
Sample
Handling and
Archiving Lab
(SHAL)
Manager
Support Staff
Gravimetric and
Microscopy
Labs
Technical
Supervisor
Support Staff
Elem
EPA Project Officer
and
Delivery Order
Project Officers
CSL Services Program
Manager
CSL Deputy Services
Program manager
t-i
Analysis Lab
Supervisor
Support Staff
and
Subcontractors
Cations/Anions
Lah
Technical
Supervisor
Support Staff
Quality Assurance Manager
Deputy Quality Assurance
Manager
QA Support Staff
Species and
SVOC Lab
Technical
Supervisors
Support Staff
and
Subcontractors
Denuder
Refurbishment
Lab
Technical
supervisor
Support Staff
Data
Management
Office
Technical
Supervisor
Support Staff
Figure 4-2. Laboratory Technical Management and Staff Organization for CSN Filter
Analysis and Data Reporting.
-------
PM2 5 CSN QAPP
Section No. 5.0
Revision No.: 1.2.0
Date: 11/11
Page 1 of6
5.0 Background and Problem Definition
For environmental pollutants, the term particulate matter (PM) is used to describe a broad class
of chemically and physically diverse substances that are either natural in origin, emitted directly
from stationary and mobile sources, or formed in the atmosphere by reactions of gaseous
emissions such as nitrogen oxides, sulfur dioxide, and volatile organic compounds (VOCs).
5.1 Background
The Clean Air Act (CAA) requires EPA to revise or update the air quality standards based on
review of the latest scientific information on known and potential human health effects
associated with PM levels typically found in the ambient air. In fulfilling these obligations, the
EPA in the mid-1990's, reviewed the air quality criteria and National Ambient Air Quality
Standards (NAAQS) for PM. The epidemiological evidence revealed an association between
ambient concentrations of PM and a range of serious health effects. Based on the results of its
review, the EPA revised and promulgated two new primary standards for the fine fraction of PM
(i.e., particles with aerodynamic diameters less than or equal to 2.5 jim, referred to as PM2.s) and
issued regulatory requirements for monitoring the chemical composition of these particles. The
EPA, with state, local, and Tribal (SLT) agencies and EPA Regions, designed a chemical
speciation network projected to consist of approximately 300 monitoring sites, to monitor and
gather data on the chemical makeup of PM2.5. The EPA designated 54 of the sites for tracking
temporal trends in concentration levels of selected ions, metals, carbon species, and organic
compounds in PM2.5. The implementing regulation further required that approximately 25 of the
trends samplers be placed at existing Photochemical Assessment Monitoring Stations (PAMS).
The locations for the remaining chemical speciation trends sites were primarily in or near larger
Metropolitan Statistical Areas (MSAs); data from these sites were to enhance the required trends
network and to provide information for developing effective State Implementation Plans (SIPs).
The balance of the 300 chemical speciation sites, dubbed "supplemental," were to be deployed at
either PM2.5 State and Local Air Monitoring Sites (SLAMS) for better spatial representation of
community scale or rural background air sheds, or for specially selected source attribution sites.
Selection of the SLAMS and other locations for supplemental chemical speciation was left up to
the particular monitoring organization. Approximately 50 of the supplemental speciation sites
selected for rural background were permitted to utilize Interagency Monitoring of Protected
Visual Environments (IMPROVE) samplers as described below. These are called IMPROVE
protocol sites. The combined Speciation Trends Network (STN) and supplemental network,
including IMPROVE protocol sites, reached its zenith in 2004 at around 325 sites.
The IMPROVE samplers were originally designed to help quantify visibility impairment in our
nation's Class 1 areas. When the Regional Haze rules were promulgated in 1999, the IMPROVE
and IMPROVE protocol network were adopted as the network to measure the baseline year
concentrations and subsequent trends of haze-forming aerosols in Class 1 areas. The IMPROVE
sampler measures essentially the same attributes of fine particulate as the Chemical Speciation
Network (CSN) samplers. However, it uses a flow rate of 22.7 liters per minute and it employs a
smaller filter (25 mm). A typical IMPROVE station also incorporates a PMio collection channel.
-------
PM2 5 CSN QAPP
Section No. 5.0
Revision No.: 1.2.0
Date: 11/11
Page 2 of6
The primary analytical protocols for total mass, trace metals and anions are very similar, but
carbon quantification in the CSN (a Thermal Optical Transmittance (TOT) method) until 2007
was procedurally different, and the data reflected a small but noticeable bias. The URG 3000N
replaced all the legacy carbon samplers of the STN and supplemental CSN in 2007-2009. It was
deployed to make carbon measurement in the CSN the same as the IMPROVE network. It uses
the sample filter media, sampler module and flow rate as the IMPROVE sampler. The only
significant difference is that the URG uses a mass flow controller to assure the flow remains
constant over the 24-hour sampling period. The same analytical procedure used by IMPROVE
(Thermal Optical Reflectance (TOR) "Method A") is now used by the CSN to quantify the
carbon constituents on the URG 3000N quartz filter. In a rulemaking by EPA (71 FR 61236,
October 17, 2006), a new designation of monitoring sites was adopted as part of a revised
national monitoring strategy. The NCore strategy called for PM2.5 chemical speciation trends
sites to be incorporated into NCore sites; however, a few adjustments were anticipated. NCore
sites will employ filter-based PM speciation sampling and analyses along with a myriad of other
monitoring protocols including new generation trace gas measurements for ozone (Os), carbon
monoxide (CO), sulfur dioxide (802), and nitrogen oxide (NOy) (all reactive oxygenated
nitrogen gases). NCore sites will continue to emphasize ambient concentration trends. The total
size of the filter-based PM2.5 CSN has shrunk from its initial 300 plus sites to between 150 and
200 sites which includes approximately 80 NCore sites and still more than 40 IMPROVE
protocol sites. The overall numbers will change further in the future, but the NCore sites should
operate indefinitely. Figure 5-1 illustrates the interrelationships of the CSN and the
IMPROVE/IMPROVE protocol networks.
IMPROVE,
168 sites
Supplemental sites,
119 From original CSN
41 IMPROVE Protocol
Class 1 Area
119 sites
Special and QA,
6 sites
IMPROVE Protocol
43 sites
I Ncore,
80 sites from
38 STN
1+ 2 IMPROVE Protocol
I+40 Converted original
V , Supplemental CSN
Figure 5-1. Interrelationships between Chemical Speciation Networks as of 2011
-------
PM2 5 CSN QAPP
Section No. 5.0
Revision No.: 1.2.0
Date: 11/11
Page 3 of6
Note that this Quality Assurance Project Plan (QAPP) still pertains to the filter-based
collection and measurement of speciated PMi.5. IMPROVE protocol sites that are part of
the CSN network are subject to siting criteria of the CSN and IMPROVE, but their
operating procedures and data management are subject to the IMPROVE QAPP and
standard operating procedures (SOPs). The IMPROVE network operates under its own
QAPP for the purposes of supporting the assessment of visibility impairment and regional
haze. Those documents may be found on the IMPROVE website
http://vista.cira.colostate.edu/improve/Publications/publications.htm; or through a link on
AMTIC at http://www.epa.gov/ttn/amtic/qapollutant.html.
5.2 Problem Definition
The CSN is a complementary network to the National PM2.5 Mass Monitoring Network (whose
goal is to measure ambient concentrations to be compared against the PM2.5 NAAQS). The CSN
data have not been used for attainment or nonattainment decisions related to PM2.s mass but its
use for other regulatory purposes, such as visibility and SIP development, are possible in the
future. The programmatic objectives of the CSN network are to provide:
• Annual and seasonal spatial characterization of aerosols;
• Air quality trends information for analysis and tracking the progress of SIP control
programs;
• Data to assist in development of emission control strategies; and
• A chemical speciation data set for comparison to the data collected from the
IMPROVE network.
Primary stakeholders in the CSN are decision-makers of SLT agencies, who will use the data as
input to models and for development of emission control strategies and determination of their
long-term effectiveness. Secondary stakeholders will include EPA analysts who will use CSN
data to determine trends of PM2.s chemical species over a period of 3 or more years, to relate the
data to health effects, and to develop and evaluate air quality models. Other users will be public
health officials and epidemiological researchers. However, expectations for data sets from the
CSN must be put in context. A number of limitations are recognized, one being the 24-hour
integrated sample approach with samples taken every 3rd day, which cannot allow determination
of diurnal patterns and may have limited use for those who study acute health effects. The EPA
recognizes these data use limitations as well as limitations of the sampling and analysis
methodologies. Thus, the EPA does not rule out the possibility that objectives, requirements, and
methods for speciation sampling may need to be adjusted in the future.
The measurement quality objectives stated in the IMPROVE QAPP are more stringent than those
of the CSN. To the extent that the IMPROVE data quality is found comparable to or better than
the CSN data, EVIPROVE's data may also be combined with CSN data in the analyses conducted
for the trends and other monitoring objectives stated in Chapter 7.
-------
PM2 5 CSN QAPP
Section No. 5.0
Revision No.: 1.2.0
Date: 11/11
Page 4 of6
5.3 PM2.s Speciation Sampling Techniques and Ongoing Research
PM2.5 speciation samplers are designed to collect 24-hour integrated samples on three different
filter media (polytetrafluoroethylene [PTFE], nylon, quartz). The CSN network samplers covered
by this QAPP are the MetOne SASS™ (spiral ambient air sampler) and Super SASS, and the
URG 3000N carbon sampler.1 The design and operation of the 3000N sampler is modeled after
IMPROVE network samplers. Appendix A to this QAPP provides SOPs for the SASS™ and a
detailed SOP for setup and operation of the URG 3000N sampler.
The MetOne SASS and SuperSASS samplers use PTFE, nylon, and quartz filter media for the
collection of target analytes. The PTFE filter is used to collect particles for the analysis of mass
and metals composition; samples for ion analysis are collected on nylon. A few MetOne
samplers still collect carbon components on quartz, but the URG 3000N is the primary carbon
sampler. It is limited to only carbon. Each sampler's operating manual and the corresponding
field operations SOP (refer to Appendix A) should be consulted for further details. New
prototype semi-continuous samplers have been developed for sulfates, carbon containing species
and trace elements. Their comparability to the filter-based sampling techniques have not been
established to date; consequently this QAPP will not apply to their utilization.
5.4 Monitoring Network Design Considerations
The initial design of the old Trends network was influenced by the need to place sites primarily
in populated areas of the country and to link PM2.5 speciation data sets to data collected at
collocated PAMS and PM2.s mass sampling sites. Appendix D of the Part 58 PM2.s regulations
(62 FR38763) provides the general criteria to apply in choosing new monitoring stations for
PM2.5. General requirements for the Trends network specified that approximately 25 sites must
be located at existing "type 2 PAMS sites." (PAMS network design and monitoring objectives
are explained in Code of Federal Regulations (CFR), Part 58 Appendix D.)
The initial selection of the remaining (trends) sites was based on EPA recommendations, with
review and advice from state and local agencies. Most of the remaining sites were located in
MSAs. Their specific locations were based on factors such as:
• Location of existing PAMS.
• Geographic locations of MSAs using the latest available population statistics.
• Ozone nonattainment areas.
nonattainment areas.
1 Historical speciation samplers included the R&P Model 2300, the URG MASS (mass aerosol speciation sampler)
400 and 450, and the Andersen RASS™ (reference ambient air sampler) 40land 410. These are no longer supported
by their manufacturers and have been replaced by the SuperSASS and URG 3000N. SOPs for the MASS and
RASS™ samplers are found in the December 2000 version of the QA Project Plan for PM25 Speciation Field
Sampling which is available from the EPA Ambient Monitoring Technical Information Center (AMTIC) website.
-------
PM2 5 CSN QAPP
Section No. 5.0
Revision No.: 1.2.0
Date: 11/11
Page 5 of6
CSN supplemental sites were initially placed at the one MSA community-oriented PM2.5 federal
reference method (FRM) mass site that has exhibited the maximum concentration in the area.
Sites placed at population centers in the central, midwest, and southeast portions of the country,
brought the total number of PM2.5 speciation trends sites to a maximum of 54. This number
decreased to 52 in 2009 and may change further as the national monitoring strategy evolves.
The strategy for the NCore network was to incorporate as many existing PM2.5 speciation trends
sites as feasible. Locating speciation trends sites at NCore provides PM2.5 paniculate matter
speciation data to complement and supplement NCore gaseous pollutant and meteorological
measurements. By mid-2009, about 2/3 (38) of the original 52 trends sites were included in the
NCore network design. It is likely that a few of the original trends sites may change location.
The remaining 14 STN sites will continue with their designation monitoring objective, but they
will be considered part of the CSN. The national air monitoring station (NAMS) designation was
eliminated by the October 2006 rulemaking.
For further information on the existing locations of PM2.5 speciation sites in the CSN, including
trends sites that are to be located at the NCore sites, refer to the EPA AMTIC website.
Refer to Section 10.0 for further information on the sampling process design for the PM2.5 CSN.
5.4 Monitoring Site-Specific Design Considerations
The CSN sampler siting criteria is based on that established for the PM2.5 FRM network. Key
requirements are set out in 40 CFR part 58 as follows
APPENDIX A Section 3.2.5.6 states
"The two collocated monitors must
be within 4 meters of each other and at least
2 meters apart for flow rates greater than 200
liters/minute or at least 1 meter apart for samplers
having flow rates less than 200 liters/
minute to preclude airflow interference."
APPENDIX E: PROBE AND MONITORING PATH SITING CRITERIA FOR
AMBIENT AIR QUALITY MONITORING, in addition to the Appendix A requirements,
are summarized below.
1. Inlets should be >20 meters from the drip line of tree(s) and must be 10 meters from the
drip line when the tree(s) act as an obstruction.
2. Distance from sampler, probe, or 90 percent of monitoring path to obstacle, such as a
building, must be at least twice the height the obstacle protrudes above the sampler,
probe, or monitoring path.
3. Inlet must have unrestricted airflow 270 degrees around the probe or sampler; 180
degrees if the probe is on the side of a building.
-------
PM2 5 CSN QAPP
Section No. 5.0
Revision No.: 1.2.0
Date: 11/11
Page 6 of6
4. The probe, sampler, or monitoring path should be away from minor sources, such as
furnace or incineration flues. The separation distance is dependent on the height of the
minor source's emission point (such as a flue), the type of fuel or waste burned, and the
quality of the fuel (sulfur, ash, or lead content). This criterion is designed to avoid undue
influences from minor sources.
5. Distance from roadways may be an important decision based on the monitoring
objectives for the particular site. Appendix E Section 6.3 covers these criteria.
All siting criteria should be discussed in the Monitoring Agency's monitoring plan and approved
by the EPA Regional Monitoring Program. The Region may also grant exemptions from some
siting criteria.
-------
PM2 5 CSN QAPP
Section No. 6
Revision No.: 1.2.0
Date: 11/11
Page 1 of 12
6.0 Project/Task Description
This section provides PM2.5 Chemical Speciation Network (CSN) participants with a background
understanding of the CSN program and the types of activities to be conducted which include:
acquiring the samples; performing chemical analysis; carrying out quality assurance/quality
control (QA/QC) procedures to achieve data quality goals; and meeting the schedules for
continued network implementation, operation, and data reporting.
6.1 Description of the Work to be Performed
6.1.1 Overview of CSN Operations
The operation of the CSN is a series of interrelated field and laboratory activities. Figure 6-1
depicts ten steps involved in implementing and operating a site in the network and the
delineation of responsibilities for required tasks. Each step or set of activities is briefly described
below:
1. The PM2.5 CSN site contact person makes arrangements for purchase and delivery of
sampling equipment to the site. Information about field site contact names, mailing
and shipping addresses, telephone numbers, supplies, and the sampling schedule is
sent to the Regional Speciation Coordinator (RSC) by the site contact person or other
state, local and/or Tribal (SLT) coordinator. The requirements for number and type of
sample filters, denuders, and sample analyses are sent to the RSC as information only.
2. The site contact conveys the site's needs with respect to the requirements for number
and type of sample filters, denuders, and sample analyses to the Delivery Order
Project Officer (DOPO). The DOPO consolidates several requests and informs the
contract laboratory's Services Program Manager (SPM) of each site's address, point
of contact, sampling schedule, needed sampling equipment and filter media, and suite
of analytes. The lab support contract project officer at the Office of Air Quality
Planning and Standards (OAQPS) authorizes the contract laboratory to begin
supplying the site, analyzing samples received, and sending analytical results.
3. The analytical support laboratory ships sampling supplies to the site contact address.
4. Site personnel conduct sampler quality control checks (e.g., time and date, leakage,
temperature, barometric pressure, flow rates, and cleanliness), conduct QC checks on
the meteorological sensors, if present (e.g., examine anemometer and wind vane for
damage; check real-time data display for wind speed, wind direction, temperature,
and relative humidity against independent observations), collect the PM2.5 samples,
deploy and retrieve field or trip blanks, pack all samples, complete the custody and
field data forms, and download data stored in the sampler's memory to an electronic
storage device.
5. Site personnel ship routine samples and field data to the contracted support
laboratory. (Sometimes non-routine and quality assurance samples are sent to the
designated EPA QA laboratory. The RSC informs the site when there are non-routine
or QA samples and how and where to send them.)
-------
PM2 5 CSN QAPP
Section No. 6
Revision No.: 1.2.0
Date: 11/11
Page 2 of 12
7.
The contracted support laboratory analyzes filter samples and conducts level 0 and
level 1 data validation. The data are sent to the state and locals who review and accept
the levels 0 and 1 validated data or provide edits. The information is sent back to the
contracted support laboratory which corrects data. The contracted support laboratory
makes final Air Quality System (AQS) database entry.
The contracted support laboratory archives filter extracts for 6 months. The
contracted support laboratory archives the polytetrafluoroethylene (PTFE) and carbon
filters and will dispose as directed by EPA at the end of their contract. Upon request,
the laboratory sends filters back to the reporting organizations that wish to store
filters for a longer period of time.
8. The contracted support laboratory prepares monthly analysis and QA/QC activity
reports.
9. The contracted support laboratory submits the monthly analysis report to CSN site
personnel for review, further validation, and verification. This report is also submitted
to the DOPO.
10. The SLT personnel conduct levels 2 and 3 data validations; questions about results
are directed to the laboratory through the DOPO. Data users and stakeholders
interpret and summarize the validated data sets to detect trends according to their own
established protocols.
•Regional Spec Contact
•DOPO
•EPA Lab Contract PO
•PMj 5 CSN Site
Contact Person
Site Operator/ technician
-Sampler QC checks
n- I.JI sampling media
"Retrieve filter samples
"Pack and ship samples with
COC and Field data forms
Ships Filters to SLT
monitoring sites
Contract
Support and
Sample Analysis
Laboratory
•Performs sample analysis
•Conduct level 0 & 1 data validations
•Provide data for review by SLT agencies
•Compile monthly QA/QC Data
•Load Audil and QC Check data
to AQS
•File annual reports to OAQPS
SLT Data Analyst
•Conduct Level 2 data
validation
•Coordinate revisions with
Lab Service Contractor
Loads Sample Data into AQS
i
EPA QA labs
•Coordinates PE Tests
•Conduct Lab TSAs
•Reports to SLTs and EPA/QAQPS
SLT QA personnel
•Perform Site Audits
•Submit reports to Data
Analysts and Contract
Service Lab
Figure 6-1. CSN Activities and Responsibilities
-------
PM2 5 CSN QAPP
Section No. 6
Revision No.: 1.2.0
Date: 11/11
Page 3 of 12
Table 6-1 lists critical field and laboratory measurements for the PM2.5 CSN and gives the
methods to be used to acquire the data. Consult the EPA Ambient Monitoring Technical
Information Center (AMTIC) website for updates.
Table 6-1. Critical Measurements in the PM2.s CSN
Parameter and/or Analyte
Methodology
Field Site
Site name, sample date, and sample ID
Temperature, ambient (Celsius)
Temperature, ambient (Celsius)**
Relative humidity (percentage)**
Pressure, atmospheric (mmHg)
Date and elapsed sample time (h)
Flow rate, sampler (L/min)***
Total volume sampled (m3)
Wind speed (m per s)**
Wind direction (degrees of compass)**
Free-form notes on sampling difficulties and unusual
conditions at the site
Commercial speciation sampler's sensor
Meteorological package
Meteorological package
Commercial speciation sampler's sensor
Commercial speciation sampler's clock and timer
Commercial speciation sampler's flow rate sensor
Commercial speciation sampler's display
Anemometer of meteorological package
Wind vane of meteorological package
Analytical Support Laboratory
Temperature, shipment container (Celsius)
Mass, PM25 (PTFE filter) (ug/filter and ug/m3 of air)
Elements (PTFE filter) (ug/m3)
Cations (various filters) (ug/m3)
Anions (various filters) (ug/m3)
Carbon species (quartz filter) (ug/m3)
Digital thermometer
Balance, microgram
Energy -dispersive X-ray fluorescence (EDXRF) (Na
through Pb)
Ion chromatography (1C) (NH4+, Na+, K+)
1C (nitrate, sulfate)
Thermal/optical analysis (total, organic, elemental, and
carbonate carbon)
** Meteorological measurements required at NCore sites; optional elsewhere.
*** Flow rates are measured by samplers over 15 minute periods and is averaged for 24-hour events
6.2 Field Activities
6.2.1 Checklist of Field Activities
From the perspective of field QA there are two key areas of concern. Making sure the sampler
operates at the design flow rate with little variability and minimizing sample contamination.
Strict attention to procedures and constant awareness of the possibility for sample contamination
are keys to obtaining a valid sample. Table 6-2 lists the activities the field site operator is
expected to accomplish for the CSN. The table also gives the frequency at which the activity
should be performed.
-------
PM2 5 CSN QAPP
Section No. 6
Revision No.: 1.2.0
Date: 11/11
Page 4 of 12
Table 6-2. Checklist of PM2.5 CSN Site Operator Field Activities
Activity
Attend hands-on training sessions. Update training via
website or other means.
Assist with equipment selection, contact with RSC,
and ordering of laboratory analytical services.
Equipment (sampler, calibration or verification
devices, and so on) receipt, inspection, inventory, and
operability checkout. Maintain spare parts inventory.
Install sampler(s) at site. De-installation of sampler(s)
at site.
Sampler calibrations.
QC checks of sampler operation (checks of time and
date display, leaks, ambient and interior temperature
sensors, pressure sensor, and flow rate).
Sampler operation. Includes preventive maintenance,
maintaining sampler cleanliness, installing denuders
and sampling media, programming sampler start/end
times, and keeping field activities notes in site logbook
and/or on field data forms.
Retrieval and packaging of samples, field blanks, and
denuders into shipment container. Completion of
custody and field data forms, and shipment of samples
to laboratory via overnight express service.
Data download and retrieve data memory card from
sampler. Store download data from MetOne samplers
to the compact flash card of the URG 3000N.
Participate in data validation as needed. Review initial
data from laboratory; conduct levels 2 and 3 data
validation1. Inform DOPO and the lab SPM of data
acceptability.
Communicate with state or local management, CSN
management, and contract laboratory (through
DOPO).
Frequency
Once, prior to site operation.
Updated training as required.
Once, prior to beginning site
operation or after request of
new equipment.
Once, initially, and whenever
new or replacement parts
arrive.
Once, at beginning of CSN
participation. Again if site is
moved or closed.
Prior to first sampling event;
annually thereafter or whenever
out-of-tolerance checks occur
that cannot be corrected;
following repairs affecting flow
rate.
Check dependent; see Table
14-1
NCore and Trends sites every
3rd day with dates and start/end
times specified by EPA. Some
non-trends sites sample every
3rd day and others every 6th
day.
Per frequency of sampler
operation or as specified in
Section 6.4 to coincide with
laboratory supplied filters and
deunders.
At time of each filter retrieval
or as soon thereafter as
practicable
Initial training. Follow-up
sessions. Monthly review of
data sets.
As required and appropriate.
Comment or Reference
See Section 8.0, this
document
Quality Assurance
Guidance Document
2.12, Section 4
Refer to Appendix A-l of
this QAPP
Refer to Appendices A-2,
A-3, and A-4 of this
QAPP
Refer to Table 14-1
Appendices A-2, A-3, and
A-4
Cleaning process must be
thorough and avoid
contamination of sample
pathways
Refer to
Appendix A-l, & A-2
Pay close attention to card
insertion and removal
procedures in the URG
3000N SOP, Appendix A-
2 and Operator's Manual.
For the MetOne
SASS/SuperSASS refer to
Appendix A-l and
MetOne SASS COM AQ
instructions
-------
PM2 5 CSN QAPP
Section No. 6
Revision No.: 1.2.0
Date: 11/11
Page 5 of 12
Activity
Frequency
Comment or Reference
Participate in scheduled CSN QA activities (for
example, on-site and field office inspections, handling
of special QA samples, installation and periodic
operation of a collocated sampler).
As scheduled by Monitoring
Agencies QA program or EPA
Office of Air Quality Planning
and Standards (OAQPS) or
EPA Regional QA Officers.
See Section 14, Table 14-1 of
this QAPP for specific
frequency
t The contract support lab will conduct a level 0 and level 1 validation but the field operators must
participate by accurately entering data on the COC and field data form (CAFDS), and then they must
review the data that have been recorded to accompany the results to make sure there are no
transcription errors. They would participate in levels 2 and 3 validations when the SLT data analyst
inquires about specific sampling results and associated meta data.
6.3 Laboratory Activities
The CSN requires extensive laboratory activities. A single contracted network support laboratory
(with subcontractors) is employed for analysis of routine samples from the network. The
following subsections and figures summarize the laboratory activities that support the CSN.
6.3.1 Pre-sampling Activities
1. PTFE, nylon, and quartz filters are received from various vendors and examined for
integrity and background analyte concentrations.
2. Filters (or their containers) are numbered or otherwise identified to allow tracking and
accurate data entry.
3. Filters are conditioned, tested, equilibrated, and weighed as required and stored
awaiting use.
4. Filters are packaged in various filter holders or sampling modules that are specific to
the type of sampler to be used at the field site.
5. The laboratory maintains shipping/receiving supplies to include shipment containers,
ice substitute packs, and custody and field data forms.
6. Assembled sampling modules including denuders are shipped to sampling sites on a
pre-determined schedule.
6.3.2 Post-sampling Activities
1. Shipments of sampling modules containing filters bearing PM2.5 deposits are received
in the laboratory, checked for integrity (damage, shipment temperature, and so on),
and logged in. Information on the custody and field data form is reviewed.
2. Filters are placed in labeled containers which are stored (refrigerated or frozen) until
ready for analysis.
-------
PM2 5 CSN QAPP
Section No. 6
Revision No.: 1.2.0
Date: 11/11
Page 6 of 12
3. Filters and denuders intended for analysis of acidic or basic gases are promptly
distributed to individual laboratories for weight determination and other analyses so
that strict time frames for completion of activities are met.
4. Results of analyses are entered into the laboratory database.
5. Sampled volume data are entered into the data entry system in order to calculate
concentrations of species in terms of a mass per unit volume of air sampled (|ig/m3).
6. Filters and filter extracts are archived (frozen or refrigerated) for 6 months or longer
if requested.
7. Data are electronically posted for examination by the DOPO and the designated SLT
contact for each CSN site. After edits and approval, the service laboratory uploads
data to the national AQS database. The reporting schedule is set-up with the
laboratory through the contract, but it is consistent with the reporting schedule for the
PM2.5 FRM network reporting requirements at 40 CFR Part 58.16.
8. All paperwork, including custody and field data forms, chromatograms, thermograms,
analyzer data reports, results of QA/QC studies, and free-form notes, are filed for
ready retrieval and inspection as required for at least a 3-year period.
-------
PM2 5 CSN QAPP
Section No. 6
Revision No.: 1.2.0
Date: 11/11
Page 7 of 12
The details for these activities are included in the laboratory SOPs, which are part of the
contracted support laboratory's QAPP available on the EPA AMTIC website. Figure 6-2 is a
simplified flow diagram of the sample analysis delivery order process. Figure 6-3 is a flow
diagram of handling and analysis steps for the filters.
:™^k
5LT forwards s'rte ^^
ID'S and sampling ^V
frequencyta f
DOPQ S
DOPOcansa lidate s and
forwards de Iriwry orders from
theirzaneto cintract lab PO at
QAQPS
" " "" R5CccpiEdcnc-:rr-sF-:n:
Sites collect samples and return
fitters to cantract bh
Contract PO forwards all derrrtr>-
crdsrs ta cantract lab
tree
i
r
Contract lab prepares and ships
filters and supfliesto srtss
Routine
Field OA/nanrcutine
1
Contract lab psrfanrB
an arises '
V
-s,^^^ type' _^r-
Contract lab psrfsnrs
valid at bn
2G Business D»
^ Data forwarded
i
EPAQAiabperfenn
analyses
1
r
I Dstate stats. /
/ ;" /
Figure 6.2. Sample analysis delivery order process.
-------
PM2 5 CSN QAPP
Section No. 6
Revision No.: 1.2.0
Date: 11/11
Page 8 of 12
Mass and Elements
Sodium-Lead
Carbon: Total, Organic,
Elemental and Carbonate
Cations: Ammonia Sodium,
Potassium
Anions: Nitrate, Sulfate
PTFE Teflon
Quartz
Nylon
Prefiring
Acceptance Testing
Acceptance Testing
Washing
Equilibration J
Pre-sampling
Gravimetry
> Transfer to the Field <
Acceptance Testing
Field Sampling
Transfer to the
Laboratory
PTFE Teflon
Post-sampling
Gravimetry Mass
X-Ray Fluorescence
Elements Sodium-
Lead
Quartz
Thermal Optical Analysis
Total, Organic, and
Elemental, Carbonate
Carbon
Nylon
(Teflon in case of
URG samples)
±
Filter Extraction
I
Ion Chromatography
Ammonium, Sodium,
Potassium, Nitrate, and
Sulfate
Filter and extract
handling
efngerated Storage
and Archive for possible
Re-analysis
/ Database: AIRS-AQS
Y Input
Figure 6-3. Diagram of laboratory filter processing and analysis activities, by filter type.
-------
PM2 5 CSN QAPP
Section No. 6
Revision No.: 1.2.0
Date: 11/11
Page 9 of 12
6.3.3 Critical Laboratory Processes and Measurements
In order to generate a mass concentration, the most critical measurements of the laboratory are
the unexposed filter preweights and the exposed filter postweights or concentrations in terms of
jag/filter. The difference between these two measurements provides the net weight of particles (or
particle components) in micrograms (ng) that, when divided by the field sampler air volume in
cubic meters (m3), provides a final concentration in micrograms per cubic meter (|ag/m3). Table
6-1 lists all the analytes to be quantified for the CSN samples.
6.4 Schedule of Activities
Whenever a new or relocated CSN or other type PM2.5 speciation sampling site comes on-line,
many aspects of the program must be completed in a timely, efficient fashion in order to meet
goals for data quality and timelines for data review and reporting.
6.4.1 Planning Time Lines
Table 6-3 outlines the sequence of activities for bringing a PM2.5 CSN field site on-line.
Table 6-3. Sequence of Activities for Bringing a PM2.5 CSN Field Site On-Line
Activity
Select sites. Arrange for space, electrical power,
personnel.
Order speciation sampler(s) and accessories.
Acquire 10-m meteorological tower and sensor
package, if appropriate.
Review conditionally approve CSN quality assurance
project plan (QAPP) and pertinent standard operating
procedures (SOPs).
Receive and inventory sampler(s) and spare parts.
Confer with Regional Speciation Coordinator (RSC)
and others.
Attend training sessions and/or review training video.
Complete site preparation and sampler installation.
Conduct safety and security checks.
Establish communications with contracted support
laboratory to set date for first series of sample
collections.
As needed, conduct hands-on training at site. Collect
one or more 24-hour test samples and complete data
transcription and documentation.
Begin routine, every 3rd-day speciation sampling at
CSN trend sites and frequencies for other SLAMS
CSN sites as adopted by the monitoring agencies.
Conduct routine field QA activities and participate in
independent audits and program assessments.
Notes
Ensure siting criteria are met.
Confer with EPA Regional Office and OAQPS prior to
ordering to confirm sampler selection
Confer with EPA prior to ordering.
Order only if needed at site(s).
Contingent on vendor schedule.
Finalize mechanism and schedule for delivery order
process for routine, non-routine, and QA samples.
Consult EPA's OAQPS and EPA Regional Offices for
plans.
Conduct acceptance tests per Table 15-1. Obtain site
documentation and photographs/slides for site file.
Request filters, data sheets, etc., from contracted support
laboratory.
Involve field staff, RSC, support laboratory, and OAQPS
as required.
Refer to every 3rd, 6th, 12th -day sampling schedules on
AMTIC website.
As scheduled or requested, interact with RSC.
-------
PM2 5 CSN QAPP
Section No. 6
Revision No.: 1.2.0
Date: 11/11
Page 10 of 12
Activity
Review data and conduct level 2 and level 3
validations of monthly data reports received from
laboratory.
Report problems and suggestions for improvement to
SLT management.
Notes
Each month. Complete review within 30 days.
RSC and OAQPS to resolve problems.
Interact with
As required, interact with RSC and OAQPS.
6.4.2 Implementation Time Lines
Other important dates must be met during implementation activities at a new site. They involve
both laboratory and field activities. One aspect of the field site implementation process that is
critical is the time-efficient use and return of denuders and filter media to the support laboratory.
As shown in Table 6-4 and stipulated in the Code of Federal Regulations (CFR), PTFE filters
must be used at the field site within 30 calendar days of presampling weighing or they must be
reconditioned and reweighed. Furthermore, the contracted support laboratory must submit
validated speciation concentration data to the EPA within 20 business days following receipt of
any sample from the field. Therefore, it is critical that the laboratory and the field sites develop,
agree on, and consistently follow a schedule that will satisfy the requirement that the mass of
PM2.5 on PTFE filters be determined within 10 days after the sampling period ends and that
determination of all analytes be completed, validated, and submitted to the DOPO and site
contact person for review and acceptance within 20 business days.
6.4.3 Field Time Lines
Table 6-4 indicates that sample filters collected by a single event sampler or a sampler run in a
single event mode should be collected and be prepared and ready for shipment to the laboratory
within 48 hours of the end of the sample period in order to prepare for the next l-in-3 day
sampling period. In some circumstances, this may not be practical or possible. If the agency
cannot collect the samples on a weekend, then an alternate schedule must be adopted. This
should be noted on the chain of custody form. Any other instance where a sample is not shipped
within 48 hours should also be noted. Refer to the AMTIC website for information on alternate
schedules. The EPA is also allowing a 110-hour retrieval and shipping schedule for filters used
in a sampler that is operated in a sequential mode typically due to an inability to perform week-
end or holiday retrievals. This applies to the first sampling event filter generated in the sequential
two event series. The second event filter is still subject to the normal 48-hour retrieval and
shipping requirement. If the monitoring agency has concerns over loss of semi-volatile mass, the
operator may retrieve the first set of MetOne PTFE and nylon filters on the first work day
following the first sequential sampling event. This can be done while the sampler is engaged in
the second of the sequential events. Data should be downloaded or removed via memory card or
other storage device from the speciation samplers on the day of filter sample retrieval. Data
should be stored on two media, one serving as the backup. In addition, the most critical data
values must also be hand-recorded from each sampler's display screens onto the sample custody
and field data form (supplied by the contracted support laboratory) and sent back to the
laboratory with the samples. The CAFDS is the form to be completed for the CSN; examples of
this form for the MetOne SASS and SuperSASS and the URG 3000N are found in Section 12.
-------
PM2 5 CSN QAPP
Section No. 6
Revision No.: 1.2.0
Date: 11/11
Page 11 of 12
Table 6-4. Critical Filter and Denuder Holding and Use Times
Filter or Denuder Type
PTFE (*)
Quartz (*)
Nylon (*)
Nitric acid denuder (magnesium
oxide) MetOne SASS sampler .(**)
Field Deadlines
Use within 30 days of preweighing;
retrieve and ship within 48 hours of
sample completion or 1 10 hours
following first event of sequential
sampling.
Retrieve and ship within 48 hours of
sample completion or 1 10 hours
following first event of sequential
sampling.
Retrieve and ship within 48 hours of
sample completion or 1 10 hours
following first event of sequential
sampling.
Replace each denuder after 3 months
use.
Laboratory Deadlines
Condition and re weigh within 10
business days of receipt from field
site.
Analyze filter deposit within 20
business days of receipt of sample.
Analyze filter deposit within 20
business days of receipt of sample.
Refurbish or replace as required.
(*) Special deadlines for use and shipment may apply to field blank and collocated sampler filters.
(**) MetOne SASS sampler denuder is integral to the ion sampling module and are serviced by the contracted
support laboratory.
6.4.4 Data Assessment Time Line
Data Availability—In order to compare the PM2.5 speciation and PM2.5 routine FRM or FEM
samplers' mass data, data from the routine FRM or FEM sampler must also be available in AQS.
"Routine sampler" refers to the gravimetric only network sampler which is always present at a
NCore or CSN Trends site. State/local requirement for data upload to AQS is 90 days after the
quarter in which the data were collected. However, the time frame for pre- and post-sampling
weighing is now more rigid than the PM2.5 total mass network, as indicated in Table 6-4. The
reason for this is to minimize any losses of semivolatiles or formation of secondary PM that was
not on the filter at the time it is retrieved. This requirement is written into the support
laboratory's contract with OAQPS. The ability to include a comparison of data acquired with
speciation monitors with that of FRM or FEM samplers at the same monitoring site in a level
1 validation is contingent on the availability of the FRM or FEM routine sampler data within the
20 business day window. The SLT may have a better opportunity to make this comparative
assessments during levels 2 and 3 validations. Assessments—after both routine (gravimetric mass
only) data and speciation data for a site are in the AQS database, OAQPS, Regions, and SLT
agencies can use the AQS data evaluation programs, based on data quality assessment (DQA)
techniques, to assess this information. This assessment is part of the level 2 and level 3 data
validation process. Consult the AMTIC website for information on use of the Special Data
Validation Tool (SDVAT) program for statistical assessment of data sets.
-------
PM2 5 CSN QAPP
Section No. 6
Revision No.: 1.2.0
Date: 11/11
Page 12 of 12
6.4.5 OAQPS Reporting Time Lines
QA Reports—The OAQPS prepares a yearly QA Final Report and an interpretive QA report
every 5 years to coincide with the data quality objectives. The yearly report beginning in 2012
will be based on the past calendar year but will include a 2-year look back on key QA and QC
elements. It is completed 6 months after the last valid entry of routine data in the preceding
calendar year's data base. The 5-year QA report is generated within 12 months after the last valid
entry of routine data in the fifth year by the state and local agencies.
6.5 Project Assessment Techniques
An assessment is an evaluation process used to measure the performance or effectiveness of a
system and its elements. As used here, assessment is an all-inclusive term used to denote any of
the following: management systems reviews, network reviews, technical system audits,
performance evaluations, and audits of data quality. Table 20-1 specifies the regions,
laboratories, and agencies responsible for these assessments.
6.6 Project Records
The field and contract laboratory programs will establish and maintain procedures for the timely
preparation, review, approval, issuance, use, control, revision, and maintenance of documents
and records. Refer to Table 9-1, CSN reporting Package Information, for the categories and types
of records and documents applicable to document control for PM2.5 sampling and analysis.
Information on key documents in each category is explained in more detail in Section 9.0 of this
QAPP.
6.7 References
1. Quality Assurance Guidance Document 2.12. Monitoring PM2.5 in Ambient Air Using
Designated Reference or Class I Equivalent Methods. US EPA, November 1998, Available at
http://www.epa.gov/ttn/amtic/files/ambient/pm25/qa/m212covd.pdf.
2. SDVAT Speciation Data Validation Tool. Information and program available at:
http ://www. epa.gov/ttn/amtic/sdvat.html.
-------
PM2 5 CSN QAPP
Section No. 7
Revision No.: 1.2.0
Date: 03/12
Page 1 of 5
7.0 Quality Objectives and Criteria for Measurement Data
7.1 Data Quality Objectives Process
The data quality objectives (DQO) process is a strategic planning approach used to prepare for a data
collection activity in order to achieve data of adequate quality to support decision-making. The DQO
process helps to ensure that the type, quantity, and quality of environmental monitoring data will be
sufficient for the data's intended use, while simultaneously ensuring that resources are not wasted
collecting unnecessary, redundant, or overly precise data. The formal DQO process consists of the
following seven steps that allow an experimental design to be developed to meet specific decision
criteria specified by stakeholders in the decision, as described in EPA QA/G-4, Guidance on Systematic
Planning Using the Data Quality Objectives Process (U.S. Environmental Protection Agency, 2006):
• State the problem.
• Identify the decision.
• Identify the inputs to the decision.
• Define the boundaries of the study.
• Develop a decision rule.
• Specify tolerable limits on decision errors.
• Optimize the design.
A Chemical Speciation DQO Workgroup was established to develop and document DQOs for the PM2.5
Chemical Speciation Trends Network (STN). The STN was subsumed into the, now called the
Chemical Speciation Network (CSN). The DQO process that the workgroup employed is fully
documented in its report (U.S. EPA, 1998), which is available online at the Ambient Monitoring
Technical Information Center (AMTIC) Web page for speciation:
http://www.epa.gov/ttn/amtic/files/ambient/pm25/spec/dqo3.pdf
7.2 Development of DQOs for the PM2.s Chemical Speciation Network
The primary DQO, detection of trends in the chemical speciation data, was defined as follows by the
EPA workgroup, whose members acted as stakeholders for the program:
"... to be able to detect a ±3 to 5 percent annual trend [i.e., ± 3 to 5 percent change in one
year] in the concentrations of selected chemical species with 3 to 5 years of data on a
site-by-site basis after adjusting for seasonality, with power of 0.80." (U.S. EPA, 1999a)
It should be noted that the DQO statement says "±3 to 5 percent" with "3 to 5 years" of data. The
default assumptions in this Quality Assurance Project Plan (QAPP) will be detection of an average ± 5
percent trend after 5 years at 80 percent probability. This presumes that 5 at least years of data are
collected at any given site with no fundamental change or changes in the way data are collected. A
fundamental change would reset the baseline year of the data collection period. However, a change in
the trend may or may not be noticeable depending on whether the change or changes affect the
uncertainty of data (hopefully improves it) or they create a systematic bias in the results, which would
be noticed as a discreet change in the values of the data.
-------
PM2 5 CSN QAPP
Section No. 7
Revision No.: 1.2.0
Date: 03/12
Page 2 of 5
Statistical power is defined as the likelihood that a particular statistical test will correctly reject the null
hypothesis when it is false. Because the null hypothesis is that a trend does not exist, the DQO
statement means that there must be an 80 percent probability of detecting a trend of
5 percent after 5 years at any particular site, for any single analyte although only sulfate, calcium, total
carbon, and nitrate were shown to have met these criteria (EPA 1999a).
Several secondary objectives for data collected at the STN sites (and now the additional NCore sites)
and other chemical speciation sites were identified, but these were not evaluated quantitatively by the
workgroup. Five important secondary data uses are as follows:
• Model evaluation, verification, and/or validation;
• Emission inventory;
• Source attribution;
• Spatial and seasonal characterization of aerosol distributions;
• State Implementation Plan attainment strategy development.
The desirable data quality characteristics for these secondary uses are probably significantly different
from those applicable to trend assessment. This QAPP section only considers the needs of the primary
objective for trend detection.
The DQO study also concluded that by sampling every 3rd day for 5 years, trends greater than
± 5 percent per year can be detected for sulfate, calcium, and total carbon, on a single-site basis. For
nitrate, however, the annual trend must exceed ± 6.3 percent to be detected with a power of 80 percent.
The decision-makers concluded that this was not sufficiently different from the 5 percent goal to require
adjustment to the sampling design. Sampling daily instead of every 3rd day provides little improvement
in the ability to detect trends; however, the model showed that cutting the sampling rate to every 6th day
begins to impair the ability to detect concentrations trends within 5 years.
Significant changes at several sites have occurred in the network's history. Andersen samplers were
left unsupported by their manufacturer and were pulled from the network. The URG 400 & 450 MASS
samplers were replaced in 2007-2009 due to a decision to outfit all CSN sites—most especially trends
and NCore with the same samplers—the Metone SASS or SuperSASS for mass, trace elements, and
ions; and the URG 3000N for carbon and carbon compounds.
7.3 Measurement Quality Objectives
Once a DQO is established, the quality of the data must be evaluated and controlled to ensure that it is
maintained within the established acceptance criteria. Measurement Quality Objectives (MQOs) are
designed to evaluate and control various phases (sampling, preparation, analysis) of the measurement
process to ensure that total measurement uncertainty is within the range necessary to achieve the DQOs.
The MQOs can be defined in terms of the following data quality indicators described below. However,
these MQOs can evolve into DQOs by aggregating over longer time periods, e.g., 5 years vs. a single
event or a month; and over a broader population of data sources, e.g., a geographic region vs. one or
two sites.
Precision - A measure of mutual agreement among individual measurements of the same property,
usually under prescribed similar conditions, and expressed generally in terms of the percent deviation
-------
PM2 5 CSN QAPP
Section No. 7
Revision No.: 1.2.0
Date: 03/12
Page 3 of 5
from an average of all the measured values, which is accepted as the best representation of the real
measured value. This is the random component of error.
Bias - The systematic or persistent distortion of a measurement process which causes error in one
direction (i.e., the expected sample measurement is different from the sample's true value in one
direction). Bias will be determined by estimating the positive and negative deviation from the true or
accepted value as a percentage of the true or accepted value and then summing the deviations.
Representativeness - A measure of the degree which data accurately and precisely represent a
characteristic of a population, parameter variations at a sampling point, a process condition, or an
environmental condition.
Detectability - The determination of the low range critical value of a characteristic that a method-
specific procedure can reliably discern.
Detection Limits Exist in the Field Measurements and the Laboratory - A measure of the capability
of an analytical method to distinguish samples that do not contain a specific analyte from samples that
contain low concentrations of the analyte; the lowest concentration or amount of the target analyte that
can be determined to be different from zero by a single measurement at a stated level of probability.
Detection limits are analyte and matrix specific and may be laboratory dependent. This characteristic is
determined by the supporting contract laboratory and it should be explained fully in the laboratory's
QAPP.
In the field, it is described as the value at which imprecision becomes unacceptably high. This value is
typically identified by the manufacturer as the recommended range of measured parameter to which the
measurement device may be utilized. The device should also be certified as yielding comparable values
to a device that has been recently certified as traceable to the National Institute of Standards and
Technology (NIST) standard reference devices and procedures for making the same measurement.
Completeness - A measure of the amount of valid data obtained from a measurement system compared
to the amount that was expected to be obtained under correct, normal conditions. Data completeness
requirements are included in the reference methods (40 CFR Part 50).
Comparability - A measure of confidence with which one data set can be compared to another.
Agreement among methods measuring the same or similar parameter; that is, converging on the same
answer, provides a measure by which to judge bias of methods deployed in the network.
For PM2.5 speciation field activities, Section 14.0 of this QAPP presents the MQOs and the calculations
for accuracy, precision, bias, and completeness. Representativeness is determined through the original
Chemical Speciation Network (CSN) DQO and sampling design process and detectability, as discussed
in the Quality Assurance Project Plan: Chemical Speciation O/PM2.5 Filter Samples prepared by the
contracted support laboratory and available from the AMTIC website. Comparability is achieved in the
CSN by use of appropriate (performance-based) sampling instruments, adherence to standard operating
procedures and QAPPs, and the use of a single laboratory for analysis and data reporting.
The strategic plan for chemical speciation monitoring trends sites (U.S. EPA, 1999b) quotes MQOs for
the overall measurement process that must be achieved in order to meet the DQO for trend detection.
These MQOs, which should be interpreted as the total coefficient of variation (CV) attributable to
sampling and analysis, are summarized in Tables 7-1 and 7. la.
-------
PM2 5 CSN QAPP
Section No. 7
Revision No.: 1.2.0
Date: 03/12
Page 4 of 5
Table 7-1. MQOs for Total Measurement Error
Analysis
Ions (anions and cations) by 1C
Total Carbon by TOA
Elements by EDXRF
MQO for Total Measurement Error (Expressed as % CV)
10
15
20
Source: U.S. EPA, 1999b.
To calculate the total measurement error corresponding to the values given in Table 7-1, it is necessary
to quantify the individual components of random error using quality control data collected by the
monitoring program. This process is described in Section 24.0. The limits on total error given in Table
7-1 apply only to major ions and elements. Elements present at significantly lower concentrations will
have much larger CVs, both from natural variability and from measurement uncertainty. These values
were reevaluated in 2011 using the collocated sampler results from 6 sites in the CSN. The results of
the analysis are summarized in Table 7. la below.
Table 7.la MQOs for Total Measurement Error
(2007-2010 Data From Collocated Samplers
Upper Bound 90th Percentile CV From Values> MDL)
Table 5. 2007-2010 Collocated CSN Data Upper Bound 90th Percentile CV From Values > MDL
Site ID
MDL ug/m3
Bakersfield, CA
Rubidoux, CA
Boston, MA
New Brunswick, NJ
Cleveland., OH
Houston, TX
Site Average CV
Al
0.015
26.56
32.84
43.49
56.26
50.42
36.02
37
Ca
0.006
25.03
22.79
30.12
50.29
43.36
23.22
32.55
Fe
0.002
23.83
20.99
20.53
46.64
25.81
28.66
28.83
Ti
0.005
31.38
33.67
39.09
59.6
37.34
29.89
32.32
Si
0.013
24.34
19.22
27.36
49.76
33.61
33.97
30.41
NO3
0.011
12.15
11.73
12.68
25.15
3.24
25.43
16.61
SO4
0.013
9.74
10.78
8.51
21.5
8.24
22.93
13.09
OC
0.064
13.45
10.19
8.67
9.33
26.57
8.82
12.84
EC
0.064
9.67
10.78
13.13
10.9
24.62
13.55
13.13
It should be noted that the data for ions are less precise on the whole than what was projected in 1999.
This may or may not be a real random variability, because the results from two sites were significantly
different than the others and skewed the average. These results have led EPA to consider moving the
samplers from three of the collocation sites and rotating them for 1-year deployments over the entire
NCore network. On the other hand, the energy-dispersive X-ray fluorescence results were not quite as
bimodal so it is not as easy to draw the conclusion that it was site-specific issues.
-------
PM2 5 CSN QAPP
Section No. 7
Revision No.: 1.2.0
Date: 03/12
Page 5 of 5
7.4 References
U.S. EPA (Environmental Protection Agency). 2006. Guidance on Systematic Planning Using the
Data Quality Objectives Process: EPA QA/G-4, Report No. EPA/240/B-06/001, U.S. EPA,
Washington, DC.
U.S. EPA (Environmental Protection Agency). 1999. Paniculate Matter (PM2.s) Speciation Guidance
Document (ThirdDraft), U.S. EPA, Research Triangle Park, NC, January 5, 1999.
U.S. EPA (Environmental Protection Agency). 1999a. Data Quality Objectives for the Trends
Component of the PM2.s Speciation Network, U.S. EPA, Research Triangle Park, NC, 1999. Available
online on AMTIC at http://www.epa.gov/ttn/amtic/files/ambient/pm25/spec/dqo3.pdf.
U.S. EPA (Environmental Protection Agency). 1999b. Strategic Plan: Development of'the Paniculate
Matter (PM2.s) Quality System for the Chemical Speciation Monitoring Trend Sites, U.S. EPA,
Research Triangle Park, NC, April 16, 1999.
-------
PM2 5 CSN QAPP
Section No. 8
Revision No.: 1.2.0
Date: 11/11
Page 1 of 3
8.0 Special Training Requirements/Certification
8.1 Training
Personnel assigned to perform PM2.5 chemical speciation monitoring activities should meet the
educational, work experience, responsibilities, personal attributes, and training requirements for
their positions. Each state, local, and/or Tribal (SLT) monitoring agency is ultimately responsible
for adequately training the personnel performing supervisory, quality assurance/quality control
(QA/QC), data handling, and other duties related to the PM2.5 Chemical Speciation Network
(CSN). Monitoring Agency management must ensure that these personnel have access to the
relevant guidance documents, standard operating procedures, Quality Assurance Project Plans
(QAPPs), and sampler operations manuals. Each monitoring agency is also responsible for
assessing the adequacy of their personnel's training and performance and for ensuring training
information is documented.
Section 8.1.1 lists the minimum requirements and some additional suggestions and ideas for
training personnel who may be unfamiliar with the operation of PM2.5 speciation samplers and
sample handling requirements. Section 8.1.2 describes sources of targeted training and internet
addresses that are specifically applicable to operating the CSN. The Ambient Monitoring
Technical Information Center (AMTIC) PM2.s Chemical Speciation website should be consulted
for the most recent training materials and workshop presentations.
8.1.1 State, Local, and/or Tribal Agency PM2.s Training
Minimum requirements for training personnel in field and/or laboratory operations for the CSN
are as follows:
• Review of most recent QAPPs for field operations and/or laboratory operations, as
applicable;
• Site operators are to review the pertinent speciation sampler manual(s) for their
site(s);
• Review of two EPA documents, Paniculate Matter Speciation Guidance Document
(January 1999) and Strategic Plan: Development of the PM2.5 Quality System for the
Chemical Speciation Monitoring Trend Sites (April 1999).
Suggestions for SLT management personnel for training operators in CSN site operations are as
follows:
• Supply an overview of PM2.s sampling and analysis through a video presentation.
Videos are available on AMTIC at http://www.epa.gov/ttn/amtic/spectraining.html.
• Schedule meetings and/or teleconferences with personnel from the SLT agencies, the
Office of Air Quality Planning and Standards (OAQPS), the Regional Speciation
Coordinator, the contracted support laboratory, and the Delivery Order Project
Officer prior to beginning sampling operations.
-------
PM2 5 CSN QAPP
Section No. 8
Revision No.: 1.2.0
Date: 11/11
Page 2 of 3
• Offer operator training classes at a centralized location. For operators who cannot
attend, a video of the procedures for operating the sampler and submitting the
samples and data reports could be supplied for training.
• Field personnel who are unfamiliar with operating and quality assuring data from
meteorological monitoring systems should review the system's operating manual and
consult the EPA's Quality Assurance Handbook for Air Pollution Measurement
Systems, Volume IV: Meteorological Measurements (EPA/600/R-94/038d).
• Continue training through update memoranda, personal instruction during on-site
systems and performance reviews, and information distributed by the
OAQPS/AMTIC Technology Transfer Network and Public Forum under the
speciation topic area. This website is available at
http ://www. epa.gov/ttn/amtic/specgen.html.
Internet Resources—The following Internet resources are specifically applicable to the PM2.5
speciation monitoring program:
Ambient Monitoring Technical Information Center
http://www.epa.gov/ttn/amtic/specgen.html—PM2.5 monitoring information: Contains links to
other areas including chemical speciation documents.
http://www.epa.gov/ttn/amtic/pmspec.html—Current chemical speciation audit and assessment
documents.
EPA/OAQPS
http://www.epa.gov/oar/oaqps/pm25/—PM2.5 data analysis: Contains ongoing activities of the
data analysis virtual workgroup, including background information, the data analysis workbook,
analysis tools (software), data sets, contacts, and links to documents and to other PM2.5 sites.
National Park Service
http://vista.cira.colostate.edu/IMPROVE/—National Park Service's visibility monitoring
information, including the Interagency Monitoring of Protected Visual Environments
(IMPROVE) network.
http://vista.cira.colostate.edu/IMPROVE/publications/IMPROVE SOPs.htm—IMPROVE
standard operation procedures (SOPs).
http://vista.cira.colostate.edu/IMPROVE/publications/news letters.htm—IMPROVE
newsletters.
-------
PM2 5 CSN QAPP
Section No. 8
Revision No.: 1.2.0
Date: 11/11
Page 3 of 3
PM2.5 Speciation Sampler Manufacturers' Information and Contacts
www.urgcorp.com—URG Corp., manufacturer of the 3000N OC/EC quartz filter-based
sampler.
www.metone.com—MetOne Inc., manufacturer of the SASS sampler.
8.2 Certification
There are no special certification requirements applicable to operation of the chemical speciation
trends samplers.
-------
PM2 5 CSN QAPP
Section No. 9
Revision No.: 1.2.0
Date: 11/11
Page 1 of6
9.0 Documentation and Records
This section defines the records critical to the Chemical Speciation Network (CSN), the
information to be included in reports or to be available for inspection, the data reporting format,
and the document control
procedures to be used.
Each monitoring agency prepare a Quality Assurance Project Plan (QAPP) and have it
. , ~,,,,T , ,, approved prior to the commencement of'data gathering
participating in the CSN should operations. In the case ofthe chemical Speciation Network,
A fundamental requirement for the operation of any data
gathering project is that the sponsoring organizations should
the approvals would occur at the state, local, and/or Tribal
(SLT) agency level and at the EPA Regional Office level. The
EPA signatories would be the Regional Speciation
Coordinator and the Regional QA Officer. However, there are
some states, who have acquired full approval and signature
authority through Regional approval of their Monitoring
Program's Quality Management Plan.
structure its records management
system in a manner that facilitates
easy retrieval of information
during internal and external
systems audits and reviews. Table
9-1 identifies the documents and
records ofthe CSN that will be
filed and retained according to the
requirements discussed in Section 9.6. The field and laboratory standard operating procedures
(SOPs) will provide instructions on the proper distribution and filing of data collected during
specific procedures.
The following subsections describe the documents and records to be included in the quality
assurance (QA) reporting packages for management, site field information operations, laboratory
operations, data reporting and management, and quality assurance functions ofthe CSN. The
term "reporting package" is defined as all the information required to support the chemical
speciation concentration data (and ancillary data) reported to the EPA, which includes all data
required to be collected as well as other data deemed to be important to the CSN.
Implementation of new or modified field or laboratory procedures may require concurrent
comparison against the old method. The document control feature (generally the date of
issuance) ofthe new version will be clearly marked, and the field or laboratory user will be asked
to discard the old version.
9.1 Information in the Management and Organization Reporting Package
There are three distinct management organizations associated with the CSN: (1) the EPA Office
of Air Quality Planning and Standards and Regional EPA offices; (2) the state, local, and/or
Tribal (SLT) monitoring offices; and (3) the contracted support laboratory. The management
reporting package for these organizations will vary but would normally consist of an
organizational structure diagram; records of personnel qualifications and training (e.g., resumes);
a quality management plan; and files containing records of grants, contracts, and official
correspondence regarding the CSN.
-------
PM2 5 CSN QAPP
Section No. 9
Revision No.: 1.2.0
Date: 11/11
Page 2 of6
Table 9-1. CSN Reporting Package Information
Categories
Management and
Organization
Site Information
Field and
Laboratory
Environmental Data
Operations
Raw Data
Data Reporting
Data Management
Quality Assurance
Record/Document Types
Organizational structure
Personnel qualifications and training
Training Certification
Quality Management Plan
EPA directives
Support contracts
Site characterization file
Site maps
Site photographs
QA Project Plans
Standard operating procedures (SOPs)
Field and laboratory notebooks and communications
Sample handling/custody records
Inspection/Maintenance records
Any original data (routine and quality control (QC)
data) including data entry forms
Data/summary/progress reports
Journal articles/papers/presentations
Data algorithms
Data management plans/flowcharts
PM25 data
Data Management Systems
Good Laboratory Practice guidelines/requirements
Control charts or tables
Data quality assessments
QA reports
System audit reports
Response/Corrective Action reports
Site and laboratory audit reports
9.2 Information in the Field Operations Reporting Package
9.2.1 Site Information
A file containing site documentation will be maintained in the SLT agency's central or field
office for each CSN monitoring site. At a minimum, the site information file must contain the
following: site characterization information that documents how and why the site was selected,
including identification of the scale of the site and the locations of nearby sources of particulate
matter; site maps and sketches; and slides, prints, or digitized images of the site taken soon after
installation of the PM2.5 speciation sampler(s). The CSN coordination office or the Regional
Speciation Coordinator may ask for copies of this material for a central file.
-------
PM2 5 CSN QAPP
Section No. 9
Revision No.: 1.2.0
Date: 11/11
Page 3 of6
9.2.2 Field Operations
Operators of individual field sites must maintain records as well. The documents and records to
be maintained at the field office are listed under "Field Operations" in Table 9-1. The following
types of notebooks or binders are to be used by field personnel to keep documents in order and
readily accessible during a site systems review.
Field Notebooks—Each field site operator will obtain hard-bound notebooks. The notebooks will
be uniquely numbered and associated with the individual and the CSN program. Generally, all
data from all routine field operations will be entered on field data forms or downloaded
electronically from the sampler's memory. The field notebook is used to record additional
information about these operations, such as information regarding weather conditions and
activities in the area that may influence the PM2.5 sample content and concentration (wind or
electrical damage to equipment, construction or mowing activities in the area, welding, traffic).
Such information should also be included in the comments section of the field data form so the
laboratory is made aware a sample may be compromised. Maintenance needs for the sampler and
the platform on which it is positioned (appearance, upkeep, and safety concerns) should be
relayed to the site operator's supervisor for consideration and action. In addition, the field
operator may use this notebook to record important communications.
Some organizations may have the capability of substituting the field notebook for electronic
communications (i.e., electronic site notebooks). This is appropriate as long as it is used
consistently.
Field Binders—A three-ring binder is a convenient repository for the appropriate data forms for
routine operations, inspection and maintenance forms, systems audit and corrective action forms,
the field QAPP, SOPs, and updates or advisories received from EPA or from other management
sectors.
Sample Shipping/Receipt—One uniquely numbered notebook to record information about sample
receipt and shipment will be used by the field site operator. These notebooks are to be dedicated
to CSN work. It will include examples of standard shipping/receiving forms and areas for free-
form notes about shipment difficulties or concerns, such as equipment that arrives damaged or
has missing parts.
9.2.3 Electronic Data Collection at Field Sites
All electronic data stored in the PM2.5 speciation sampler will be downloaded to a laptop
computer or other electronic transfer device at each field site. A diskette or other removable
memory device containing information from each of the sampling events will be created. A copy
of the electronic download should be retained by the field operator for sampler troubleshooting
and for later use in data validation. It is recommended that data be downloaded after each run;
however, data from a number of runs may be accumulated in the sampler's memory, if bad
weather or scheduling difficulties prevent prompt downloads. NOTE: The sampler's memory is
finite. Do not accumulate more than two runs before downloading. Refer to the sampler
operating manual for details.
-------
PM2 5 CSN QAPP
Section No. 9
Revision No.: 1.2.0
Date: 11/11
Page 4 of6
9.2.4 Hand-Entered Data at Field Sites
A number of forms will require hand entry of data. These forms are primarily associated with the
field operations. An example of the combined custody and field data form and explanations of its
content and use can be found in the SOPs of Appendix A to this QAPP and in Section 12.0. The
forms should be in three-part carbonless paper format. A quality assurance/quality control report
form, shown in Section 14.0, also requires hand-entry of data from checks of the flow,
temperature, and pressure sensors of a sampler. Information should be entered clearly with a
black or blue indelible ballpoint pen. Any entry errors should be marked out with a single line
and the correct information entered above this line. The operator must sign or initial the form to
verify the accuracy and completeness of the entries.
Information recorded on these field forms is very important because it serves as a backup in case
the data downloaded from the sampler become corrupted or lost. Difficulties with, or suggestions
for improved operation of the samplers should be recorded here and in the field notebook.
Information about significant events near the site that may affect the representativeness of the
sample should also be entered into this form so the laboratory will be on alert for an unusually
high concentration sample.
A form for audit findings from the internal (or external) systems audits of the field site or the
contract laboratory is another example of a hand-entered data form. The response to the audit
finding, a Corrective Action Response (CAR) form, may also be hand-entered. Copies of all field
site reviews conducted by internal or external agencies and the site's responses to them should be
retained at the field site office.
9.3 Information in the Laboratory Operations Reporting Package
9.3.1 General Laboratory Information
A file or files of general laboratory information should be available for inspection. It will include
schematics of the laboratories showing locations of analytical stations, a record of equipment
purchases or leases, warranty information, and maintenance and service agreements with
instrument vendors and suppliers. The laboratory's safety manual should be a part of this
information.
9.3.2 Laboratory Operations
Information related to the laboratory operations and data management reporting packages is
given in the QAPP for the CSN contracted support laboratory.
9.4 Information in the QA Reporting Package
Four distinct QA organizations are associated with the PM2.5 CSN.
9.4.1 Network QA Manager
The Office of Air Quality Planning and Standards (OAQPS) management organization has a QA
Manager. This Manager and staff members will produce and update the quality system and will
-------
PM2 5 CSN QAPP
Section No. 9
Revision No.: 1.2.0
Date: 11/11
Page 5 of6
review QA documents from the CSN's field and laboratory functions. QA document updates and
directives will be issued from this office. The QA Manager for OAQPS has overall approval
authority for the CSN QAPP and any subsequent changes made to it.
9.4.2 State and Local Field Site QA Offices
The various SLT agencies whose personnel operate the CSN sites will keep a file of documents
and procedures that are part of their customary QA plan for site and sampling, equipment receipt
and acceptance testing, equipment maintenance, sampler operational checks, and operator
training. Documented results of and CAR responses to internal site inspections and external
reviews will also be kept on file for reference and inspection. Each SLT is expected to provide
independent assessments of their monitoring sites and samplers according to the schedule given
in Table 14-1 QA/QC activities.
9.4.3 Contracted Support Laboratory QA Manager's Office
The contracted support laboratory's QA Manager and staff will develop and periodically update
the laboratory QAPP and SOPs for routine servicing of the sites and analysis of filter samples.
Results of internal systems and performance reviews of laboratory operations will be kept on file
and available for inspection by the EPA.
9.4.4 EPA QA Laboratory and Regional Offices support
The EPA's Office of Radiation and Indoor Ai/National Air and Radiation Environmental
Laboratory in Montgomery, Alabama, will provide QA services for CSN laboratory activities.
The EPA OAQPS and Regional laboratories will provide some QA services to the field and
laboratory components of the CSN such as management system reviews every 3 years and an
occasional independent audit of one or more monitoring sites. Their procedures will be
documented and kept on file and updated as necessary using document control methods.
9.5 Reports to Management
The SLT field site operators will follow their customary procedures for reporting information to
their management. It is suggested that any verbal communications be documented in the field
notebook or that a copy of the notebook entry be sent to the manager and to the Regional
speciation coordinator as appropriate. Field site operators will also report information on PM2.5
sample and meteorological data capture rates and problems with sampling equipment and check
devices for the speciation samplers. The EPA has constructed electronic reporting forms for all
performance verifications, technical systems audits, and sampler audits. These are to be uploaded
to a web-based receptacle from which a non-public data base will be populated. Report functions
are under development for use by the monitoring agencies and the EPA.
They may also report on their review and levels 2 and 3 validation of the draft data sets supplied
by the contract laboratory, if they perform these validation activities.
-------
PM2 5 CSN QAPP
Section No. 9
Revision No.: 1.2.0
Date: 11/11
Page 6 of6
The contracted support laboratory is responsible for forwarding completed level 0 and level 1
data validation checklists (Section 22) and the results of filter analyses to the EPA delivery order
project officer (DOPO) and to the SLT agency contact, who submitted the delivery order, within
20 business days from receipt of the sample. Levels 0 and 1 data validation is described in the
laboratory QAPP. Upon approval, the final data set will be released from the laboratory in
hardcopy and electronic format to the EPA DOPO, the state or local agency contact, and the Air
Quality System (AQS) electronic data base. Draft and final composite semiannual data summary
reports will be issued from the laboratory to the SLT agency contact, the DOPO, and the EPA
Project Officer. Results of resolution of problems with data sets will also be prepared, sent to the
concerned parties, and filed for the record.
9.6 Archival and Retrieval of Data Reporting Packages
Each organization participating in the CSN is expected to keep updated versions of the
documents and data sets listed in Table 9-1 on file and accessible for the duration of the trends
network's operation or a specified lesser time. The files of the sites, the contracted support
laboratory, Regional and EPA QA offices, and validated electronic AQS data sets must be
retrievable for inspection and review during regular business hours by the OAQPS management
or other U.S. Government authorities.
Limits on the time of data (and sample filter) retention are decided by OAQPS program
management. Storage and archival of all field and laboratory data associated with each analysis
in electronic format for up to 3 years following sample analysis is required, with the exception of
particle photomicrographs and associated spectral data in electronic format which may be
discarded 12 months after receipt of the sample for examination. Data archivists should contact
OAQPS a month before the data discard date to ask for a decision on further archival or disposal.
-------
PM2 5 CSN QAPP
Section No. 10
Revision No.: 1.2.0
Date: 03/12
Page 1 of 5
10.0 Sampling Process Design
This section describes all components associated with on-site field operations of the nationwide
PM2.5 Chemical Speciation Network (CSN). The network will be operated by state, local and/or
Tribal (SLT) agencies. This section describes the field activities and the key parameters to be
estimated, the networks where primary speciation samplers and collocated quality assurance
(QA) speciation samplers are located, and the frequency of sampling using the primary and QA
samplers. The EPA Ambient Monitoring Technical Information Center (AMTIC) website
[http://www.epa.gov/ttn/amtic/specgen.html] contains a listing of all current CSN sites. The
network design components comply with the regulations specified in 40 Code of Federal
Regulations (CFR) Part 58, Section 58.13, Appendices A, D and E, which are further described
in the document Guidance for Network Design and Optimum Site Exposure for PM2.s andPMw
(U.S. EPA 1997). The NCore and other network components of the CSN have, for the most part,
already been designed based on collaborative input from SLT agencies and other federal
agencies (for example, the National Park Service and the National Forest Service). The final
NCore design has been solidified and the network began operation January 1, 2011. More
information is available at http://www.epa.gov/ttn/amtic/ncore/index.html.
10.1 Scheduled Project Activities, Including Management Activities
The CSN collects filter-based at all NCore locations and at selected state and local air monitoring
station (SLAMS) sites. Chemical components of PM2.s are determined by a contracted support
laboratory. The site locations are determined by the CFR Part 58, Section 58.13, Appendices A
and D, and by the current National Monitoring Strategy implemented by EPA and SLT agencies.
All NCore sites will have a collocated PM2.5 Federal Reference Method (FRM) sampler present
to collect samples for determination of particle mass concentration.
Attaining a complete understanding of the interactions and scheduling of field site and support
laboratory activities is a very important consideration of the sampling process design. The
sequence of activities for the contracted support laboratory for the NCore trends network
component of the CSN is presented in Table 10-1. The site supply and analytical aspects of the
NCore network component of the CSN will remain essentially the same as in prior years.
-------
PM2 5 CSN QAPP
Section No. 10
Revision No.: 1.2.0
Date: 03/12
Page 2 of 5
Table 10-1. Sequence of Activities for CSN Contracted Support Laboratory
Activity
Prepare and obtain approval of Quality Assurance Project
Plan (QAPP) and standard operating procedures (SOPs) for
laboratory activities and field site support.
Meet with OAQPS project scientists, regional speciation
coordinators, and delivery order project officers (DOPOs).
Receive and analyze routine speciation samples.
Assemble, validate, and report data. Review field and
laboratory interactions; review and resolve problems.
Complete analyses within 20 business days of sample receipt.
Prepare and submit monthly draft summary data reports to
DOPO and agencies for review and validation.
Submit final data packages following SLT and DOPO
approval of draft data.
QA auditing and inspections.
Systems audit and review of data archives.
Notes
Includes laboratory SOPs, tested and approved for
use.
Receive details on CSN site contacts and sampler
requirements. Finalize mechanism and schedule for
delivery order process for routine samples.
Continuing process.
Continuing process.
Conduct level 0 and level 1 validation of data.
Enter data into Air Quality System.
QA Manager for contracted support laboratory
conducted audits.
Coordinated by EPA.
10.2 Rationale for the Design
10.2.1 Network Design
The rationale for the design of the CSN originated in the monitoring regulations, promulgated at
Federal Register (62 FR 38763), as part of the PM2.5 National Ambient Air Quality Standards
review completed in 1997. The background to the planning and design of the initial CSN is
covered in Section 5 of this document. The overall network design strategy was to locate 25
trends-assessing PM2.5 samplers at photochemical air monitoring stations (PAMS) sites and the
remaining approximately 25 trends sites at "core" SLAMS. The SLT agencies were given the
responsibility to site supplemental speciation samplers for background, transport, and local
program needs. Essentially 54 trends sites were established and a supplemental network of up to
225 sites was deployed at the network's highest site-count in 2004. Revised monitoring
regulations in October 2006 called for an updated national strategy that would establish an
NCore network of up to 75 sites which would incorporate most of the original 54 trends sites.
Approximately 240 of the original 300 plus CSN sites have been retained to monitor for
speciated PM2.5. There are still approximately 60 Interagency Monitoring of Protected Visual
Environments (IMPROVE) protocol sites and about 180 CSN sites, which include 80
MetOne/URG speciation sampling combos at NCore sites.
Network design requirements stated in 40 CFR Part 58, Appendix D, Sections 4.7.4 and 4.7.5,
provide guidance on locating monitoring sites for PM 2.5 with respect to scale and monitoring
objectives. Site requirements with respect to roadways, sampler probe heights, and juxtaposition
with other samplers are specified at 40 CFR 58, Appendix E, and are summarized below in Table
10.2.
-------
PM2 5 CSN QAPP
Section No. 10
Revision No.: 1.2.0
Date: 03/12
Page 3 of 5
Table 10-2. Design Criteria for Collection Site Surroundings
1.
The height of the inlet to the sampler should be between 2 and 15 meters above ground surface.
2.
For samplers located on roofs or other structures, the minimum separation distance between the inlet
and any structure should be greater than 2 meters.
3.
The sampler should be located away from obstacles so that the sampler is at a distance at least twice
the height of the obstacle. For example, a tree is 10 meters tall and is east of the sampler. The
sampler would need to be placed at least 20 meters (preferably further) away from the drip line of a
tree.
4.
An unrestricted air flow of 270° must exist around the sampler inlet.
5.
If the sampler is located on the side of a building, a 180° air flow clearance is required.
6.
Sampler inlet should be located at least 10 meters from the drip line of any tree.
7.
Minimum distance to any roadway is 10 meters, but this value is determined by the average daily
number of vehicles (refer to table in 40 CFR Part 58 Appendix E for exact information).
The inlet height for collocated samplers should agree vertically within 1 meter.
The closest horizontal distance for placement of a collocated sampler to a Lo-Vol sampler is 1
meter; to a Hi-Vol sampler is 2 meters. The maximum horizontal distance a collocated sampler may
be from any sampler is 4 meters.
10.2.2 Speciation Sampler Design and Deployed Models
Sampler design requirements stated in 40 CFR Part 58, Appendix D, Sections 4.7.4 and 4.7.5,
require that speciation samplers incorporate particle inlets and size fractionators having particle
size discrimination curves comparable to the FRM for PM2.5, employ denuder technology to
remove acidic gases, have face velocities and sample volume capture similar to the FRM, and be
reliable and rugged in field use. However, speciation samplers are not required to attain reference
or equivalent method designation and should be selected based on performance in order to meet
the CSN data quality objectives. Desirable features of speciation samplers are discussed in the
PM2.5 speciation guidance document (U.S. EPA 1999). Operating information for speciation
sampler models—MetOne and URG—currently used in the CSN network is given in Appendix
A-l and A-2. of this QAPP.
.10.2.3 Sampling Frequency
The NCore sites of the CSN will sample consistently every 3r day. Routine samples will be
collected for a 24-hour period beginning and ending at midnight on the assigned day. Other sites
in the CSN network (such as SLAMS sites) will sample on either an every 3r day or an
abbreviated version of the lin 3 day frequency called the "alternate lin 3"1; or every 6* day
1 The national monitoring schedule to which all monitoring programs and networks adhere causes Friday and
Monday sampling events every third weekend. The Alternate 1 in 3 schedule was established due to the inability of
SLT's to authorize overtime work on weekends and holidays to recover samples and set-up the next sampling event.
It essentially requires the agency to sample on Friday and collect filters on Monday on one week-end; and on the
-------
PM2 5 CSN QAPP
Section No. 10
Revision No.: 1.2.0
Date: 03/12
Page 4 of 5
schedule. The sampling schedules are available on the AMTIC website. This schedule should be
printed and be distributed to all site supervisors and operators.
Field blanks shall be generated at a frequency of approximately 3 percent of the normal routine
sampling frequency of MetOne PTFE filters and 10 percent of URG 3000N quartz filters. The
procedure for MetOne field blanks will change in January 2012 to better simulate the field blank
procedure of the URG 3000N. That is, the field blank module will be attached to a channel for
which the air flow tube has been detached from the pump and capped. A very sharp cut cyclone
will be retained in its operating position. The module will remain on the non-sampling channel
for the entire period over which the routine PTFE module is attached; typically channel 1 (and 5
on a SuperSASS).
Trip blanks will not be collected for the MetOne unless they are requested by specific monitoring
agencies for specific reasons, which each requestor must supply. Trip blanks will be generated
for the URG 3000N at a frequency of 2 percent of the routine sample filter shipments. Again this
frequency may be adjusted based on issues that arise.
Sites in the CSN will not acquire samples on any other days unless specifically directed to do so
by network authorities. Should non-routine sampling be requested, such samples must be taken
on days other than those set by the every-3rd-day schedule ,or every-6th-day schedule, or taken on
different samplers on the same day, and must be handled separately when submitted to the
contracted support laboratory for analysis.
10.2.4 Collocation of samplers
Collocated samplers provide a set of PM2.5 speciation data, originating from a separate but
otherwise identical sampler, which can be used to estimate the precision of the total sample
collection, handling, and analysis/data reporting process. The locations of these sites and all of
the CSN sites are available on the AMTIC website. The original design included two sites for
every approved speciation sampler. As of 2009, only MetOne and URG 3000N samplers are
collocated at these sites. The collocated QA samplers are operated every 6* day to coincide with
the start and end run times of the site's primary sampler. Samples and data from the collocated
sampler will be handled in exactly the same way as those from the primary sampler. Section 14.0
discusses the calculation of precision in more detail.
The original design of the speciation trends network called for collocation at 10 percent of the
sites, which ultimately led to six collocation sites. The EPA has determined that better
representation of the entire network is needed; therefore, a design change is under way which
will leave collocated speciation samplers at two to four of the NCore sites. The remaining two to
four collocation samplers will be deployed on an annual rotating basis to other sites around the
country. This design strategy may change slightly based on results over the first 2 years. Other
options include:
• Decreasing the duration and increasing the number of sites that participate each year;
third week-end following, the Friday event is abandoned for sampling on Monday and recovering the sample on
Tuesday.
-------
PM2 5 CSN QAPP
Section No. 10
Revision No.: 1.2.0
Date: 03/12
Page 5 of 5
• Adding additional samplers to the collocation fleet to allow more sites to sample
collocated for 12 months
• Revise the ratio of permanent collocation to rotating collocation sites based on needs
assessed by the EPA as informed by SLTs, CASAC and organizations conducting health
studies.
10.2.5 Deviations From the Sampling Plan
Because one of the major goals of NCore sites of the CSN network is to determine trends in
PM2.5 chemical speciation concentrations overtime and within geographical areas, deviations
from the sampling plan are allowed only upon prior consultation with and approval of the EPA.
Should an occasional operational problem occur at a site, the site operator must note this in the
field site notebook and on the field data form that is returned to the contracted support laboratory
with the sample filters so the resulting analytical data can be flagged. Unapproved, continued
deviations from the sampling plan at one or more sites will give rise to a review of the site
agency's operating plan, the personnel involved, and a request for prompt corrective action.
Furthermore, the data sets acquired during periods when deviations from the sampling plan
occurred or operational problems were encountered may be sequestered and not used in trends
analysis studies.
10.3 References
U.S. Environmental Protection Agency. 1997. Guidance for Network Design and Optimum Site
Exposure for PM2.5 and PM10. Publication No. EPA-454/R-99-022. December 1997. Available
at: www.epa.gov/ttn/amtic/files/ambient/pm25/network/r-99-022.pdf
U.S. Environmental Protection Agency. 1999. Paniculate Matter (PM2.s) Speciation Guidance
Document. Final Draft. October 7, 1999. Available at:
www.epa.gov/ttn/amtic/files/ambient/pm25/spec/specfml.pdf
-------
PM2 5 CSN QAPP
Section No. 11
Revision No.: 1.2.0
Date: 03/12
Page 1 of7
11.0 Sampling Methods Requirements
The Chemical Speciation Network (CSN) provides for measurement of the mass and chemical
component concentrations of fine particulate matter with an aerodynamic diameter less than or
equal to a nominal 2.5 jim (PM^.s) in ambient air over a 24-hour period. At the inception of the
CSN, three different brands of PM2.5 speciation monitors were in use. These three samplers were
subjected to bench testing and to comprehensive, multi-city field test in which all the samplers
were positioned in a close array. They collected filter samples for several events. The results
from each sampler were compared with each of the others and against the Federal Reference
Method (FRM) and IMPROVE samplers. All of the samplers met the requirements stated in the
EPA requisition for speciation samplers. Now (as of 2011), the sampler fleet consists of the
MetOne SASS or Super SASS and the URG 3000N for EC/OC measurements. Use of the old
legacy URG MASS, Andersen RAAS, and Thermo-Fisher's RP Partisol Model 2300 has ended.
Standard operating procedures (SOPs) for setup, operation, and quality control of the most often
used samplers are given in Appendix A to this Quality Assurance Project Plan (QAPP); SOPs for
other legacy samplers are available on the EPA Ambient Monitoring Technical Information
Center (AMTIC) website.
11.1 Sample Collection and Preparation
11.1.1 Preparation
Before a site visit, the operator must gather sampling modules containing the filter, data forms,
and sampler verification equipment to check flow, temperature, and pressure (if a quality control
(QC) check is scheduled). The sampling modules must be transported to the sites in a protected
environment and not subjected to high temperatures.
Shipment of sampling modules to the laboratory will require the use of ice substitutes and
insulated containers. The operator must freeze the ice packs prior to use. During transport
to/from the sites, the ice substitutes may be placed in an electric transport cooler to maintain their
frozen state.
11.1.2 Field Sample Collection
The speciation sampler will be permanently installed within 1 to 4 m of the site's routine FRM
sampler. All NCore sites will have a PM2.5 FRM sampler present. The proper operation of the
speciation samplers and the collocated FRM sampler must be confirmed before the first run. A
testing and acceptance checklist appears in Section 15.0 of this QAPP. Sampling modules will be
installed. At NCore and the other remaining trends sites, the samplers will run every 3r day or on
the "alternate 1 in 3 schedule," from midnight-to-midnight on local standard time the whole
year. Refer to Appendix A for details on setup and operation of the sampler, handling of filter
sampling modules, hand-entry of data, downloading of electronic files, and sampler QC check
requirements.
-------
PM2 5 CSN QAPP
Section No. 11
Revision No.: 1.2.0
Date: 03/12
Page 2 of7
11.1.3 Sampler Recorded Measurements
Table 11-1 lists the information that is provided by any of the speciation samplers. This
information will be stored in the sampler's memory and can be downloaded to disks. Essential
information will also be transcribed from the display screen of the sampler and hand-entered on
to a custody and field data form. This form is described in Section 12.0 of this QAPP.
Table 11-1. Summary of Information Provided by Speciation Sampler
Information to be Provided
Flow rate, average, for the
sample period
Flow rate, CV, for the sample
period
Flow rate, 5-min average out of
specification (FLAG)
Sample volume, total
Temperature, ambient, min.,
max., average, for the sample
period
Barometric pressure, ambient,
min., max., average, for the
sample period
Filter temperature, differential,
30-min interval, out of
specification (FLAG)
Date and time
Sample start and stop time
settings
Sample period start time
Elapsed sample time
Elapsed sample time out of
specification (FLAG)
Power interruptions >1 min,
start time of first ten
interruptions
User-entered information, such
as sampler and site
identification
Units
L/min
L/min
m3
°C
mmHg
yr/mo/d/h/
min
yr/mo/d/h/
min
yr/mo/d/h/
min
hmin
hmin
Availability
Anytime
* (optional)
*
-------
PM2 5 CSN QAPP
Section No. 11
Revision No.: 1.2.0
Date: 03/12
Page 3 of7
fine particulate is predicated on a particular sampler channel's design flow, deviations of greater
than 10 percent will set a flag for that sample period. Further, if a sampling period is less than 23
hours or greater than 25 hours, the sample will be flagged. Other conditions may cause a flag to
be set. These include power losses and extreme difference in ambient and sampler interior
temperatures.
11.1.4 Sampling Module Transportation
The used sampling modules must be stored in a protective transport container and transported to
the contract laboratory as soon as possible. Sampling modules should be shipped out by
overnight express within 48 hours following the end of a sample run. Samples are shipped with
cold packs designed to maintain the sample temperature at 4°C. If an agency must deviate from
sending the sample out within 48 hours, please see the discussion in Section 6.4.3.
11.1.5 Field Maintenance and Calibration
A maintenance schedule must be developed for field sampling equipment and verification
devices. See Section 15.0 of this QAPP for more information. Consult the operator's manuals,
the SOPs in Appendix A, and Section 16.0 of this QAPP for requirements and procedures for
calibration of temperature and pressure sensors and the flow rates of sampling channels.
11.2 Sampling/Measurement System Corrective Action Process
11.2.1 Corrections to the SOPs
The state and local agency field site operators and their supervisors are responsible for
implementing this QAPP and the field SOPs and are, in part, responsible for the quality of the
data. If changes or corrections are suggested for the SOP methods or QAPP, state or regional
personnel will notify the Chemical Speciation QA Lead in the Office of Air Quality Planning
and Standards (OAQPS) Coordination Office. The QA Team members will review the
information and convey the issue to an ad hoc Chemical Speciation Workgroup consisting of
volunteer EPA and state, local and/or Tribal site operators and data managers, convened by the
OAQPS Speciation QA lead. The workgroup will review the proposed change and attempt to
classify the change according to the effect the change would have on the data. Class types
follow:
Class 1—The change improves the data and the new procedure replaces the current procedure. If
found to be acceptable by the workgroup, a new SOP will be issued that can be inserted into the
QAPP. The document control information in the heading will contain a new revision number and
date. A quality bulletin will be filled out describing the change and distributed to all affected
personnel.
Class 2—The change provides for an alternative that does not affect the quality of the data but
may provide for efficiencies in some circumstances or be cost-effective. If found to be
acceptable by the workgroup, the original SOP will not be altered, but an addendum to the
procedure will be issued that describes the modification and provides for the use of the
alternative method.
-------
PM2 5 CSN QAPP
Section No. 11
Revision No.: 1.2.0
Date: 03/12
Page 4 of7
Class 3—The change is grammatical in nature and does not reflect a change in the procedure.
The changes will be highlighted and modified during a Class 1 change (where appropriate) or
corrected during the development of a full revision to the document.
Upon agreement by workgroup to institute a change, hard copies of Class 1 and 2 changes will
be distributed using a quality bulletin such as that illustrated in Figure 11-1. The CSN laboratory,
each site, and the management of each site will be notified. New versions of SOPs will be
conveyed to CSN participants with instructions to discard the old version.
11.2.2 Data Operations
Corrective action measures in field operations of the CSN will be taken to ensure the data quality
objectives are attained. Potentially, there are many types of sampling and measurement system
corrective actions. Table 11-2 lists common problems and a few of the corrective actions needed
for a well-run chemical speciation network.
11.3 Avoiding Sample Contamination; Temperature and Holding Time
Requirements
This section details the requirements needed to prevent sample contamination, the temperature
preservation requirements, and the permissible holding times to limit degradation of the sample
catch.
11.3.1 Sample Contamination Prevention
The CSN must have rigid requirements for preventing sample contamination. Powder-free anti-
static gloves are worn while handling filter cassettes or sampling modules in the laboratory.
Once the filter cassette or sampling module leaves the laboratory, it must not be opened due to
the potential for filter damage or contamination. Sampling modules (used with the MetOne
SASS and Super SASS, and the URG 3000N) will be capped and protected in plastic sealable
bags during shipment to and from the site. When used cassettes or sampling modules are
removed from the sampler, they must be promptly capped or otherwise protected to prevent
contamination from dusts, gases, or abrasion. The site operator's hands must be clean when
handling sampling modules and it is suggested that they be cleaned immediately before the
sample handling step. It is recommended that clean disposable gloves be used.
-------
PM2 5 CSN QAPP
Section No. 11
Revision No.: 1.2.0
Date: 03/12
Page 5 of7
Quality Bulletin
Subject:
Number
Date
Page
of
Supersedes No.
Dated
Replace and Discard Original
Add Material to Document
Notes:
PM25 CSN QA Team Leader
Retain this bulletin until further notice
This bulletin will be invalid after (Date)
This bulletin will be incorporated into quality.
Procedure No. by (Date)
Figure 11-1. Quality bulletin.
-------
PM2 5 CSN QAPP
Section No. 11
Revision No.: 1.2.0
Date: 03/12
Page 6 of7
11.3.2 Temperature Preservation and Holding Time Requirements
During shipment from the laboratory to the sample location, there are no specific requirements
for temperature control; however, the filters or sampling modules should remain in their
protective containers and inside the transport container. Excessive heat must be avoided (e.g., do
not leave in direct sunlight or a closed-up car during summer). During the sampling (24-hour
period), the filters will be subject to ambient temperatures and should not exceed the ambient
temperature by more than 5°C for more than 30 minutes continuously. If this occurs, a flag
should be set automatically by the sampler.
Sampling modules should be removed from the sampler within 48 hours after the sampling
period ends. If this time limit is missed, the sample will be processed but a flag will be applied to
the data. An exception will also be when the sampler is run sequentially and filters are retrieved
after the second event occurs (within 110 hours of the first two sequentially programmed events).
Refer to Section 6.4 for information on how to handle shipment time exceptions. The
temperature of sampling modules must be brought to 4°C as soon as possible and the shipment
package, cooled to 4°C, should be ready for pickup by the courier service as soon as possible,
unless the samples will be transported back to the field office and stored in a refrigerator (i.e.,
Friday sample collection).
Table 11-2. CSN Field Operations Corrective Procedures
Item
Sampler failure
Sample flow rate
verifications)
Leak test
Problem(s)
Unacceptable performance.
Repair/replacement
needed.
Out of specification (+5 %
of reference standard) or
outside ± 5% of design
flow rate.
Leak outside acceptable
tolerance
Action
Repair or replace sampler or
component at factory. Obtain
spare sampler.
1. Completely remove flow rate
measurement adapter;
reconnect and perform flow rate
check again
2. Perform leak test
3 . Recalibrate flow rate
1. Completely remove flow rate
measurement adapter;
reconnect and perform leak test
again
2. Inspect all seals and O-rings;
replace as necessary and
perform leak test again
3 . Check sampler with different
leak test device if applicable
Notification
Notify contracted support
lab at both trade-out
and trade-in.
1. Document on QA/QC
field data form
2. Document on QA/QC
field data form
3 . Document on field
calibration data form
and notify supervisor
1 . Document in field
notebook and on
QA/QC field data form
2. Document in field
notebook and on
QA/QC field data form;
flag data since last
successful leak test.
3 . Document in field
notebook and on
QA/QC field data form
-------
PM2 5 CSN QAPP
Section No. 11
Revision No.: 1.2.0
Date: 03/12
Page 7 of7
Item
Sample flow rate
Ambient temperature
sensor verification
and filter
temperature sensor
verification
Filter temperature
should approximate
ambient
temperature
Ambient pressure
verification
Elapsed sample time
Elapsed sample time
Power
Power
Data downloading
Problem(s)
Consistently low flows
documented during
sample run
Out of specification (+2°C
of standard)
Filter Temperature 5°C >
ambient temperature for
more than 30 minutes
Out of specification (+10
mmHg of standard)
Out of specification
(+ 5 min)
Sample did not run, or
elapsed time was outside
limits of 24 ± 1 hours
Power interruptions
Liquid-crystal display
(LCD) panel on, but
sampler not working
Data will not transfer to
laptop computer or other
storage device
Action
1 . Check programming of sampler
flow rate
2. Check flow with a flow rate
verification filter in place and
determine if actual flow is
within acceptance limits
1. Recalibrate sensor; replace
sensor; Consult with
manufacturer for
troubleshooting, repair or
replacement.
For MetOne, Check filter temp
sensor with independent device;
Check for excessive heat source
near sampling head.
1 . Make certain pressure sensors
are exposed to the ambient air
and are not in direct sunlight
2. Call local airport or other
source of ambient pressure data
and compare that pressure to
pressure data from monitors
sensor; pressure correction may
be required
3 . Connect new pressure sensor
1 . Check programming; verify
power outages
2. Reset
1. Check programming
2. Try programming sample run to
start while operator is at site;
ensure the transport filter is in
the unit
1 . Check line voltage
1 . Check circuit breaker; some
samples have battery backup
for data but will not work
without AC power
1 . Document key information on
sample data sheet and
additional information in site
notebook. Make certain
problems are resolved before
data are written over in sampler
microprocessor
Notification
1 . Document in field
notebook and on
QA/QC field data sheet
2. Document in field
notebook and on
QA/QC field data sheet
1 . Document in field
notebook and on
QA/QC field data form
1 . Document in Field
notebook, and on
CAFDF. Consult with
mfgr. If occurs more
than once.
1 . Document in field
notebook and on
QA/QC field data form
2. Document in field
notebook and on
QA/QC field data form
3 . Document as above
1 . Document on data sheet
2. Document in field
notebook
1 . Document in field
notebook
-------
PM2 5 CSN QAPP
Section No. 12
Revision No.: 1.2.0
Date: 03/12
Page 1 of 8
12.0 Sample Handling and Custody Requirements
12.1 Introduction
This section describes sample handling and custody procedures that are necessary to ensure that:
• Chemical Speciation Network (CSN) site operators properly handle the sampling
components from the time of receipt at the field office until they are released to the
shipping agency for return to the CSN laboratory.
• Field sites use Chain of Custody (COC) documentation; subsequent laboratory COC
is maintained for each sample, beginning with placement of the filters in the sampler
collection modules and extending through all analytical steps to final sample archival.
12. 2 Filter Handling and Custody Procedures Prior to Sampling Event
Care must be taken when handling, storing, and transporting filters at all stages in their use due
to the small mass of particles collected on exposed filters, the potential for sample losses due to
rough handling or sample volatilization, and the potential for weight gain due to contamination
or uptake of reactive gases on the filter and particulate matter surfaces. Care must also be
exercised in handling denuders to ensure acidic gases are quantitatively removed from the
sample air stream and that the denuder's coating does not dislodge and fall onto the sample filter.
Sample handling procedures must be consistently followed in order to provide data meeting the
data quality objectives. These procedures are discussed below following the order of activity
from the point where it is ready to be loaded up to the sampling event.
12.2.1 Procedures in the CSN Laboratory
The contracted service laboratory loads the filter cassettes. The 46 mm filters go into the MetOne
canisters and the 25 mm filters go into cassettes that are fixed in the URG cartridges. This avoids
any need for the field operators to physically handle or touch the filter.
Details on how the contract laboratory handles the denuders and filters, loads the filters into
sampling modules, and packages the components for shipment to the field office are given in the
CSN contracted support laboratory's standard operating procedures (SOPs) and Quality
Assurance Project Plan (QAPP), available on the EPA Ambient Monitoring Technical
Information Center (AMTIC) website.
Sample custody procedures are required to avoid misplacement of samples or confusion of one
sample with another, and to provide documentation to assist in detection and resolution of COC
problems. A sample is considered to be in custody if it is in one's actual physical possession or
stored in a secured area restricted to authorized personnel.
Each set of sampling modules and other equipment supplied by the laboratory (such as cyclones
and denuders) will be accompanied by a carbonless three-part, PM2.5 chain of custody and field
data form. This form will contain the filter identification number, filter type, container (module
or cassette) identification number, and date by which the sampling media must be used. An
example form is illustrated in Figure 12-1 and its contents are explained in Table 12-1. The
-------
PM2 5 CSN QAPP
Section No. 12
Revision No.: 1.2.0
Date: 03/12
Page 2 of 8
laboratory fills in much of the information required in parts A, B, and C of the form and retains
the third copy. The information on the custody form is ultimately entered into a sample tracking
system, where an electronic record is kept.
Note that the use of quartz fiber filters (MetOne channels 3 or 7 on Figure 12-1) has been phased
out in all but approximately six monitoring sites. The sites that continue to use channel 3 on the
MetOne SASS or SuperSASS are collecting data for historical comparisons of the two methods.
The URG 3000N is now used to collect carbon and carbon-containing PM2.s particulate on
quartz filters. Refer to Appendix A-3 for details on the custody and field data form for the
URG3000N sampler.
12.2.2 Procedures at the Field Office
Upon receipt at the field office of a set of sampling components for a particular speciation
sampler, the CSN site operator must carry out the following documentation and handling steps:
• Enter receipt of the shipment in the operator's field notebook, noting the date and time
of receipt and any air bill or other identifying numbers associated with the shipment.
• See Section 12.2.1 for information about a second sampler (the URG 3000N) that has
replaced the quartz filter channel on the 3-channel speciation sampler. The free
channel may be used for special studies.
• Inspect the exterior of the shipping container, note any evident damage, and record
observations in the operator's field notebook.
• Open the shipping container and ensure that a COC and field data form is present for
each set of sampler components sent in the shipment. Also check to be sure shipping
items such as ice substitute gel packs and a minimum/maximum thermometer (if
required) are present. Ensure each identifying number printed on the COC form
corresponds to an enclosed sampling channel component. Do not use any sampling
component whose identifying bar code number is not listed on the COC form. Notify
the CSN support laboratory about any discrepancies. Remove the gel packs and
freeze them at -18°C (0°F).
It has been noted by support laboratory, that the interior temperature of some of the
containers received in the laboratory has been above 4°C. This problem may be due
to the gel packs not being frozen long enough or at a cold enough temperature. It is
recommended that gel packs be frozen for at least 3 days at a temperature of-18°C.
This will ensure that the filters do arrive at the support laboratory at temperatures at
or below 4°C.
• Sign and date the custody record portion of the COC form.
-------
PM2 5 CSN QAPP
Section No. 12
Revision No.: 1.2.0
Date: 03/12
Page 3 of 8
• Store all components for a sampler run together in a container in an air-conditioned
secure area for later transport to the site. Adopt a first-in, first-out use schedule.
Sampling components should be stored and tracked so that the correct set of sampling
components reaches the designated field collection site for use on the designated
sampling day.
Do not interchange sampler channel components intended for use with a particular speciation
sampler at a particular site with components for any other sampler or site. The CSN contracted
support laboratory has labeled each sampler channel component for use at a particular site.
Should an interchange occur, the CSN site operator must fully document the variance and inform
the support laboratory so the analytical results can be associated with the correct sampler and
site.
12.3 Sample Handling and Custody Procedures for Collection of Samples
The following procedures are brief descriptions of standard operating procedures. For more
detailed information about the process of removing sampling modules and denuders from a
sampler, filling in field data forms, downloading data electronically, and packaging samples in a
cooled container for shipment, refer to the SOPs in Appendix A to this QAPP, to SOPs
concerning sample packaging and shipment in the contracted support laboratory's QAPP, and to
the sampler's operating manual.
12.3.1 Installation of Filters
Sampling components/modules must be used at the field collection site on the sampling date
specified on the COC and data form. Unused sampling modules and denuders should remain
sealed or capped and kept from exposure to ambient air, temperature extremes, or vibrations.
Upon arrival at the site to set up a sampling event, the CSN site operator should follow the SOP
written for the sampler and to the sampler's operation manual. Refer to Appendix A SOPs for
instructions for the installation of sampling modules and denuders and programming of the
sampler. There are quality control (QC) procedures that may take place at prescribed frequencies
which may coincide with the installation of the sampling filters. Examples would be flow rate
checks, temperature probe and pressure sensor checks. These will be completed with appropriate
reporting forms. If a problem is discovered, the operator will take whatever steps are necessary
to initiate the agency's corrective response plan. Collecting filter samples with a sampler that is
not performing as designed, generates data of little value.
Once the sampling modules are installed at the site and the sampler is programmed to begin
operation, the operator should complete the appropriate sections of the COC data form.
12.3.2 Post-sampling Procedures at the Field Collection Site
Within 48 hours after the end of a sampling period, the CSN site operator should remove the
sampling modules from the sampler. On an alternate schedule is it possible that retrieval may
only be possible after 56-64 hours following a Friday or Monday holiday.
At the site, the operator must complete the following:
-------
PM2 5 CSN QAPP
Section No. 12
Revision No.: 1.2.0
Date: 03/12
Page 4 of 8
Read selected data from the sampler's display screen and enter them in Section E of
the custody and field data form. Double-check all entries against the sampler display.
Print clearly. Be certain the entries are clear on the second and third pages of the
carbonless form. Refer to the custody and field data form example in Figure 12-1 and
to Table 12-1 for details. A site may have more than one PM2.5 speciation sampler
and, thus, the operator must complete additional field data forms.
Remove the filter cassettes or sampling modules from the sampler. Briefly examine
the cassette or module for damage and ensure it is, in fact, the correct module for the
sampling channel from which it was removed.
Place the sampling modules in protective container(s); cap the denuders if they are to
be returned. Place all sampling materials in the shipping/transport container
containing ice substitutes, but do not seal.
Download data from the sampler to a laptop computer and then transfer it to a labeled
portable flash storage device. Alternatively, MetOne manufacturers a data transfer
device to which data can be downloaded for later transfer to a computer and then a
portable flash storage device.
Return to the field office to complete packing and shipping arrangements. There are
some sites that accommodate all the facilities needed to prepare the sample(s) for
shipping immediately upon retrieval, in which case, the operator would proceed to the
next set of procedures immediately.
-------
PM2 5 CSN QAPP
Section No. 12
Revision No.: 1.2.0
Date: 03/12
Page 5 of 8
Figure 12-1 CSN Custody and Field Data Form for MetOne SASS Sampler
Q144162P
PM 2.5 STN CUSTODY AND
FIELD DATA FORM
c. White (return to lab)
c. Yellow (site retains)
c. Pink (lab)
A. CUSTODY RECORD (Name. Date)
Bin ID: B15137
Set: 3
1. Laboratory, Out
2. Site, In
3. Site, Out
4. Lab, In
B, SITE AND SAMPLER INFORMATION
1. Site AIRS Code 450190049
2. Sampler S/N
3. Sampler Type SASS
4. Sampler POC 5
5. Site Name CPW
6. Intended date of use Monday. September 03, 2007
7. Date of Sampler set-up
8. Operator's name
C, SAMPLER CHANNEL COMPONENTS
Channel No. Component ID No.
Component Description
1
Kept at Site
SASS cyclone
I2925Q
SASS cassette (Teflon filter) (GREEN)
Kept at Site
SASS cyclone
I2926R
SASS cassette (MgO denuder, nylon filter) (RED)
Kept at Site
SASS cyclone
I2927S
SASS cassette (quartz filter) (ORANGE)
D. START, END, AND RETRIEVAL TIMES
Channel No.
Start date
Start time
End date
End time
Retrieval date
Retrieval time
1
E. SAMPLER CHANNEL INFORMATION (Post-Sampling)
Channel
No.
Run
Time
Run
Time,
Flag
Sample
Volume
(m3)
Avg.
flow
(L/min)
Avg.
flow CV
Avg.
ambient
T(°C)
Max.
ambient
T(°C)
Min.
ambient
T(°C)
Channel
No.
AT
Flag
Avg. Filter
T(°C)
Max. Filter
T("C)
Min. Filter
T(°C)
Avg. BP
(mm Hg)
Max. BP
(mm Hg)
Min. BP
(mm Hg)
F. Comments
-------
PM2 5 CSN QAPP
Section No. 12
Revision No.: 1.2.0
Date: 03/12
Page 6 of 8
Figure 12-2 CSN Custody and Field Data Form for URG 3000N Sampler
Q135885G
TRAINING DE
PM 2.5 CSN CUSTODY AND
FIELD DATA ¥ORM
FOR TRAIMIMGySE ONLY
o. White (return to lab)
c. Yellow (ste retains)
o. Pink (lab)
A. CUSTODY RECORO (Name, Pate)
Bin ID; B23102
Set; 6a
1. ^aboratory, Out
2- Site, In
3. Site, Out
4, Lao. In
B, SITE AND SAMPLER INFORMATION
1. Site AIRS Code 4Sfl11DQ04
2. Sampler S>'N
3. Sampler Type URG 3QDQN
4. Sampler POC 5
5, Site Narva Bountiful
8, Intended date of use Friday. April 06. 2007
7. Date of Sampler set-up
8. Operator's name
C, SAMPLER CHANNEL COMPONENTS
Position
Conoonent ID No.
Component Description
1801 SO
Quartz Cartridge ID
1
13018'P
Memory Card ID
D. START, END, AND RETRIEVAL TIMES
Position
Start date
Start time
End date
End time
Retrieval dale
Retrieval tit™"*
E. SAMPLER CHANNEL INFORMATION (Post-Sampling)
Run
T-rne
Run
Ti me,
Fiag
Sample
Volume
Avg.
flow
(L''min)
Avg,
'tow CV
.Avg.
ambient
T i* C)
Max,
ambient
T i* Cj
Mia
ambient
T f C}
Avg, BP
(mm Hg)
Max. BP
(mm Hg)
Mm, BP
(mm Hg)
F. Comments
-------
PM2 5 CSN QAPP
Section No. 12
Revision No.: 1.2.0
Date: 03/12
Page 7 of 8
Table 12-1. Explanation of CSN Custody and Field Data Form
CAFDF Section
Explanation of Section Contents
Top of form
The custody/data form number will be unique to each sample set and assigned in
advance by the laboratory. The 3-part carbonless form will be distributed as follows:
• Top copy (white original) ~ returned to the CSN support laboratory
• Second copy (yellow) - retained by the field site office
• Third copy (pink) ~ retained by the originator
A. Custody Record
Acknowledge relinquishing and receiving custody in this section. Persons should
sign their name (legibly) and record the date.
B. Site and Sampler
Information
Information about the site and the date the sampler modules are to be used. Most of
this information will be pre-entered by the laboratory.
C. Sampler Channel
Components
The sampling components needed for a particular sampler and its multi-channel
sampling arrangement are listed here. They are identified by bar code tracking
number and by a free-form description. The brand of the sampler is identified. This
information will be entered by the laboratory, preprinted on the form. A separate
custody/data form will be used for each set of sampling modules intended for routine
sampling, field blank studies, trip blank studies, and special studies.
D. Start, End, and
Retrieval Times
These entries are made by the site operator. The start and end times correspond to
those programmed into the sampler during the setup phase. The operator must
enter these data clearly and must double-check the values against the sampler
display screen. The retrieval date and time indicate when the sampling modules
were removed from the sampler.
E. Sampler Channel
Information
Postsampling information can be transcribed by hand directly from the display
screen of the sampler. The operator is responsible for making these entries at the site.
The sample volumes will be used by the support laboratory to compute analyte
concentrations. The SLT agencies will use these data in levels 2 and 3 validations to
identify problems with the sampler. Again, the operator must enter these data
clearly and must double-check the values against the sampler display screen.
F. Comments
This section offers a place to record further notes on any part of the form as well as
observations of abnormally high emissions in the vicinity. The person recording
information here should refer to the sections of the form. Detailed information
should also be recorded in the field or laboratory notebook and referenced to the
unique custody/data form number, location, sampler, and sampling date.
12.3.3 Post-sampling Shipping Procedures
Within 48 hours following the end of the sampling period (or within 24 hours after the second
sampling period of a sequential set), the CSN site operator will: return the modules to the field
office (which may be the monitoring site); complete all paperwork; seal the COC data form in a
plastic bag and tape the bag on the inside of the transport container, and package the modules
and denuders in the container for pickup by the shipping agency. Illustrated packing instructions
for the modules can be found in Appendix A. Please see Appendix A for details on packing the
modules for shipment to the contracted support laboratory. At the field office, the operator must
complete the following:
-------
PM2 5 CSN QAPP
Section No. 12
Revision No.: 1.2.0
Date: 03/12
Page 8 of 8
• Retain the second page of the three-part COC data form and package the top copy in
the shipping container. Package the sampling modules, insert the ice packs in the
shipping container, and take the container to a drop point or arrange for pickup by the
contracted overnight air shipping company. Chain of custody seals on the shipping
coolers or containers are not required.
• Complete the shipping air bill, attach it to the shipping container, and present the
package to the shipping agent.
• It is important to avoid shipping on Fridays. If a sample must be recovered on a
Friday, place the sample in cold storage—between 0-4°C, and ship it on Monday.
Special circumstances should be discussed with the SHAL of the lab services
contractor.
12.3.4 Procedures in the CSN Laboratory
The CSN contracted support laboratory's procedures for receiving the sampling components and
field data, disassembling the sampling modules, and handling the filters and denuders after their
distribution to the various laboratories are covered in the CSN laboratory's QAPP, which is
posted on the EPA AMTIC website.
12.4 Filter and Sample Archival in the CSN Support Laboratory
The contracted support laboratory's data base will assign a tracking number to all sample filters.
Extracts and remnants of filters will be archived in cold storage. Custody procedures for
inventorying and archiving these materials are given in the support laboratory's document
Quality Assurance Project Plan: Chemical Speciation ofPM2.s Filter Samples.
-------
PM2 5 CSN QAPP
Section No. 13
Revision No.: 1.2.0
Date: 10/11
Page 1 of 3
13.0 Analytical Methods Requirements
Analytical methods requirements have been extracted from the Quality Assurance Project Plan
(QAPP) for the contracted support laboratory which serves the Chemical Speciation Network
(CSN) and other state, local and/or Tribal (SLT) agency speciation sampling programs. (See
http://www.epa.gov/ttn/amtic/specguid.html.) They are listed below:
B.4 Analytical Methods Requirements
B.4.1 Gravimetric Mass Determination
Standard Operating Procedure for PM2.5 Gravimetric Analysis describes the procedure to be
used for gravimetric mass determination in RTFs laboratory.
B.4.2 EDXRF Analysis for Elements
Standard operating procedures used by RTI and CHESTER LabNet for EDXRF analysis are
listed below:
RTI
Standard Operating Procedures for X-Ray Fluorescence Analysis of PM2.5 Deposits on Teflon
Filters
CHESTER LabNet
• Standard Operating Procedures for the Sample Receipt and Log In
• Standard Operating Procedures for the Analysis of Elements in Air P articulates by XRF
(Kevex 771)
• Standard Operating Procedures for the Analysis of Elements in Air P articulates by XRF
(Kevex 770 and 772)
• Standard Operating Procedures for the Kevex Spectrometer Data Generation,
Interpretation, and Reporting
• Standard Operating Procedure for the Kevex XRF Spectrometer Calibration
B.4.3 Extraction and Analysis ofAnions and Cations
For an overview of RTFs laboratory facility and procedures for extraction and analysis of anions
and cations, including nitrate and sulfate, see:
• Standard Operating Procedure for Cleaning Nylon Filters Used for Collection ofPM2.s
Material
• Standard Operating Procedures for PM2.5 Anion Analysis
• Standard Operating Procedures for PM2. s Cation Analysis
BAA Carbon Analysis
RTFs laboratory facility and procedures for organic, elemental, carbonate, and total carbon
analysis by the CSN TOT method are detailed in the Standard Operating Procedure for the
Determination of Organic, Elemental, and Total Carbon in P articulate Matter Using a
Thermal/Optical-Transmittal Carbon Analyzer.
-------
PM2 5 CSN QAPP
Section No. 13
Revision No.: 1.2.0
Date: 10/11
Page 2 of 3
DRI's laboratory facility and procedures for OC/EC analysis by the IMPROVE_A/TOR-TOT
method are detailed in the Standard Operating Procedure for Thermal/Optical Reflectance
Carbon Analysis of Aerosol Filter Samples.
RTFs laboratory facility and procedures for OC/EC analysis by the IMPROVE_A/TOR-TOT
method are detailed in three SOPs:
• Standard Operating Procedures for Temperature Calibration of the Sample Thermocouple in
a Sunset Laboratory or a DRIModel 2001 Carbon Aerosol Analyzer
* Standard Operating Procedure for the Determination of Carbon Fractions in Particulate
Matter Using the IMPROVE A Heating Protocol on a Sunset Laboratory Dual-Mode Analyzer
* Standard Operating Procedure for the Determination of Carbon Fractions in P articulate
Matter Using the IMPROVE A Heating Protocol on a DRI Model 2001 Analyzer
While the CSN/TOT analysis and the IMPROVE/TOR-TOT analysis are similar in technical
approach, and both require heating ramps under non-oxidizing conditions followed by heating
ramps under oxidizing conditions, the two methods are fundamentally different in the way they
define carbon fractions. The CSN/TOT analysis is a timed analysis with fixed times at each
temperature in the heating profile, and the IMPROVE/TOR-TOT analysis is an event-driven
analysis with the sample remaining at a given temperature until evolution of carbon from the
filter drops to near zero. A CSN/TOT analysis runs for a total of 12 minutes; an
EVIPROVE_A/TOR-TOT analysis may take anywhere from 15 minutes to 70 minutes. The
maximum temperatures for the various fractions are also different as are the way that "Peaks" are
defined. For the CSN/TOT analysis, the five OC peaks are defined as their contributions to OC.
For CSN/TOT, the sum of the five OC peaks is always equal to OC. For the FMPROVE A/TOR-
TOT method, the four OC peaks and the three EC peaks are independently calculated without
regard to the OC/EC split and pyrolyzed carbon has a negative value if the OC/EC split comes
before the addition of oxygen. Table B.4.1 below describes and compares the carbon fractions
measured for both methods and the conditions (e.g., non-oxidizing or oxidizing atmosphere and
maximum temperature) under which each fraction is measured.
B.4.5 Semi-volatile Organic Compounds
SVOCs will be analyzed by RTFs subcontractor DRF The analysis is detailed in DRFs Standard
Operating Procedures for Analysis ofSVOC by GC/MS.
B.4.6 Characterization of Particles by Electron and Optical Microscopy
RTI will provide SEM and optical microscopy for characterization of particulate samples. RTI
has extensive experience in the analysis of airborne PM by both optical and electron microscopy
techniques, having analyzed a large number of PMio filters by optical microscopy for several
state air quality agencies and a significant number of air filters for commercial firms.
B.4.6.1 Scanning Electron Microscopy
SEM can be employed to characterize individual particles collected on a filter. Particles may be
sized and the morphology described on an individual basis. The composition of a particle may be
-------
PM2 5 CSN QAPP
Section No. 13
Revision No.: 1.2.0
Date: 10/11
Page 3 of 3
determined by EDXRF. Characterization of a large number of particles provides information as
to the particle size, distribution, and chemistry of the PM. Any of several filter media can be used
to collect particulate material, but smooth-surface filters such as polycarbonate filters are far
superior for the purposes of analysis by SEM. The procedures for analysis by particulate material
by SEM are described in detail in Standard Operating Procedure for Sample Preparation and
Analysis ofPMjo and PM2.5 Samples by Scanning Electron Microscopy.
6.4.6.2 Optical Microscopy
The RTI optical microscopy laboratory is fully equipped with both stereo binocular and
polarizing light microscopes (PLM) capable of both reflected and transmitted light analysis.
Photomicrography capabilities allow for documentation of particle characteristics. No RTI SOP
currently exists for optical examination of filter media. Procedures will be carried out at the
direction of the DOPOs, and reporting criteria and formats will be established at the time of the
initial requests. Analysis by optical microscopy allows for examination of particles having
apparent diameters less than 0.25 um. Optical characteristics such as color, refractive indices,
bi-refringence, and morphology (size and shape) can be determined, which may aid in the
identification of particles. The Teflon filter commonly employed in PM2.5 sampling is not a
suitable substrate for analysis by optical microscopy because the thickness and translucent nature
of the filter severely limits the transmittal of light. Additionally, the surface of the filter is highly
irregular, making it very difficult to observe individual particles. Other filter media such as
mixed cellulose ester (MCE) or polycarbonate provide substrates that are more suitable for
analysis by optical microscopy. Refer to that QAPP, Quality Assurance Plan: Chemical
Speciation ofPM2.s Filter Samples, for more information. The contracted laboratory's QAPP is
available on the EPA's AMTIC website: http://www.epa.gov/ttn/amtic/specguid.html. "
The CSN laboratory's technical management and staff organization are presented in Section 4.0
of this QAPP and the contract support lab's QAPP. Refer to Section 4.3 and Figure 4-2 for more
information.
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 1 of 20
14.0 Quality Control Requirements
Quality control (QC) is the overall system of technical activities that measures the performance of an
ongoing process against established standards to verify that such performance meets the stated
requirements established by the data user or stakeholder. In the case of the Chemical Speciation Network
(CSN), QC activities are used to ensure and document that the measurement uncertainties, as discussed in
Section 7.0, are maintained within specified limits so that the data quality objective for trend detection can
ultimately be met.
There are three distinct but interrelated functions that the PM2.5 CSN QC program requires of state, local
and/or Tribal (SLT) field site participants, as follows:
1. The site operator and his or her supervisor must make every effort to keep the sampler
maintained, cleaned, and operating properly; to retrieve samples according to the network
schedule; and to ship (cold as prescribed) the samples, field blanks, and trip blanks with
supporting hand-entered and computer-downloaded data to the CSN contracted support
laboratory on schedule.
2. The site operator must control the sampler's collection process through proper handling of
cassettes and sampling modules, by properly calibrating samplers, and by conducting both
scheduled and as-needed performance checks for leaks, flow rate, temperature, and pressure.
Any out-of-tolerance findings must be followed by corrective actions.
3. The third function is carrying out validation of data sets sent from the CSN support laboratory
to the SLT contacts and then reporting the data quality statistics that describe how well the
accuracy, bias, precision, and completeness goals have been met.
Table 14-1 contains a complete listing of field QC parameters, their frequency of assessment or period of
performance, and the acceptance criteria or advisory limits, i.e., measurement quality objectives (MQOs).
The Contracted Support Laboratory QC procedures are included in its Quality Assurance Plan Chemical
Speciation ofPM2.s Filters which is available on the EPA Ambient Monitoring technical Information
Cemter (AMTIC) website at http://www.epa.gov/ttn/amtic/specguid.html.
The MQOs have been established using two sets of criteria. Some of the MQOs were derived to be
consistent with the data quality objective (DQO) for identifying a trend of 5 percent (in either direction)
can be made after 5 years of sampling. Other MQOs are based on requirements for PM2.5 monitoring, set
forth in the Code of Federal Regulations, or by standards of good practice described in Section 2.12 of the
U.S. Environmental Protection Agency Quality Assurance Handbook for Air Pollution Measurement
Systems (U.S. EPA, 1998) available through the AMTIC website. The strategic plan for the Speciation
Trends Network (STN) (U.S. EPA, 1999) set out the initial QC requirements for field and laboratory
activities. Those requirements are herein applied to the chemical speciation sites in the NCore network.
Some acceptability limits have been made more rigorous and a few additional requirements have been
added due to the historical performance of the network. For consistency, the EPA recommends that all
chemical speciation sampling sites adopt these MQOs with the exception of monitoring frequency Table
14-1 also lists the action to be taken if a QC check shows a parameter to be outside of acceptance limits or
outside of advisory limits, and how the corrective action is demonstrated and documented. Specific
procedures for implementing field QC activities are in the standard operation procedures (SOPs) in
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 2 of 20
Appendix A to this Quality Assurance Project Plan (QAPP). The remainder of this section describes the
types of QC checks called for in the PM2.5 CSN and the procedures used to calculate values for the
principal QC indicators. The QC results should be recorded on a form or in a field notebook dedicated to
this purpose. Some will be included on monthly performance check forms or audit reports and uploaded to
a website from which data can be extracted, formatted and then posted to the EPA's Air Quality System. It
is recommended that results of the QC checks be entered into a control chart or graph to help visualize
changes or drifts in sensor responses and to alert the site operator to the need for preventive maintenance,
repair, or replacement of a speciation sampler. Detailed information on multipoint calibrations and repairs
should be recorded in the field notebook.
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 3 of 20
Table 14-1. MQOs and Associated QC Activities for the PM2.5 CSN
Measurement
Filter visual checks
by Shipping and
Handling laboratory
Collocation with
another chemical
speciation sampler:
The result is used to
calculate precision,
reported as CV in
terms of a variation
in the relative
percent differences
in concentrations
measured by the
two collocated
samplers. (This is
different from the
CV of flow rates
explained below.)
Evaluation of
precision results
Frequency
before and after
each exposure
As of 2007, about
4% of CSN (6
sites) were
conducting
collocated
sampling on a 1-
in-6-day basis(a)
Single event
Monthly (5
sampling events)
Annual
Acceptance Criteria
(MQO) or Advisory
Limit
free of visible defects
see Tables 7.1 and
7. la for all maximum
precision limits of
PM25 concentration
and various chemical
species concentrations
expressed as % CV
(b)
Total Carbon 15%
Sulfate ion 10%
Nitrate ions 10%
Trace elements 20%
If Avg. CV(ofa
sampling event) > 2o
of the previous 3
year's average of CVs
for the sampling site
If CV> 1.05 oof the
previous 3 year's
average CV
If CV>0.5oofthe
previous 3 year's
average CV or the
DQO whichever is
greater
Corrective Action if Out
of Specification
record in laboratory
database; contaminated
filter usually discarded
prior to use; exposed filters
are flagged or invalidated
Present data to ad hoc
workgroup to determine if
the acceptance limits are
appropriate. Review trends
analysis data to determine
if a 5% trend over 5 years
is discernable. If yes,
readjust % CV
requirements.
Review analytical history
with contract support lab.
Review flow checks and
flow CVs for individual
sampling events. Field
blanks for evidence of
contamination. Review
Shipping procedures and
delivery records. Initiate
TSA of site and lab retain
data or invalidate based on
conclusions
Samples or
Channels
all filter types
all sampling
media
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 4 of 20
Measurement
Frequency
Acceptance Criteria
(MQO) or Advisory
Limit
Corrective Action if Out
of Specification
Samples or
Channels
Temperature:
Check
Audit
Monthly
Quarterly
±2°C of a certified
transfer standard
±2°CofaNIST-
traceable standard that
is independent of the
monthly check
standard
note on QC data form;
troubleshoot; recalibrate or
replace sensor and conduct
recheck; Consult with
manufacturer if outside of
limits in 2 of any 3
contiguous months.
note on data form; review
monthly performance
checks; recalibrate or
replace sensor; re-audit
after calibration or
replacement. Consult with
manufacturer if outside of
limits in 2 of any 3
contiguous months.
all temperature
sensors
all temperature
sensors
Pressure:
Check
Audit
Monthly
Quarterly
± 10 mmHgvs.
certified NIST-
traceable transfer
standard
± 10 mmHgvs. a
NIST-traceable
standard that is
independent of the
monthly check
standard
note on data form;
recalibrate or and replace
sensor if not responsive;
and conduct recheck.
Consult with manufacturer
if outside of limits in 2 of
any 3 contiguous months.
note on data form; review
monthly performance
checks; advise
recalibration or
replacement of sensor; re-
audit after calibration or
replacement. Consult with
manufacturer if outside of
limits in 2 of any 3
contiguous months.
barometric
pressure sensor
barometric
pressure sensor
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 5 of 20
Measurement
Flow Rate:
Average CV
Operator Check (d)
& Review
Single event
flow rate check
Operator Check(e)
Flow rate
one-point audit (e)
Frequency
Review after every
event
Month
Semi-Annual, but
no intervening
periods greater
than 7 months;
none less than 5
months unless
there are more than
2 audits per year
Acceptance Criteria
(MQO) or Advisory
Limit
MetOne (c)
IfCV>2%ofset
flow , or The avg diff
between sequential
pairs of flow readings
is > 0.1 3 liters/min)
URG 3000N
Tf PV > 1 % nf spt
flow calculated from
data polled after first
five minutes
±5%ofNIST
Traceable standard
± 5% vs. a standard
that is independent of
the monthly check
standard; an
independent transfer
standard
Corrective Action if Out
of Specification
Conduct flow check.
Note deviation on data
form; review last 6
monthly flow checks and
CV data from previous 2
months monitoring results.
See Figure 14.1 for
decision tree
Note on audit form; review
monthly flow rate checks
performance checks. Use
decision tree for evaluating
performance history. See
Figure 14.1. Advise
recalibration, repair or
replacement of sampler.
Re-audit after calibration
or replacement of sampler
or replacement of flow
related components.
Samples or
Channels
all flow
channels
all flow
channels
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 6 of 20
Measurement
Blanks(f):
trip blanks
MetOne
Mass, elements, &
ions
Carbon (URG3000)
field blanks
MetOne
Mass,
elements,
& ions
Carbon (URG3000)
SHAL Blanks
Frequency
Eliminated in lieu
of SHAL blanks
0% (subject to
change if
significant
contamination is
detected)
2%
3% or one set
every 10 sampling
days (subject to
change)
10%
10%
backup
filters
MetOne 2% of all
PTFE modules;
(nylon filter
modules assumed
= 0.0 ug/filter)
URG 3000N Same
as Trip Blanks
Acceptance Criteria
(MQO) or Advisory
Limit
refer to support
laboratory QAPP.
If collected, 15 ug/
filter
5.64 ug/filter (g)
30 ug mass per filter;
other limits apply to
chemical species
concentrations (refer
to lab QAPP)
6.0 ug/filter (g)
MetOne — to be
established as avg of
up to 3 years
preceding current year
+ 2 Std Dev.
URG 3000N—
1.8 ug/filter(h)
Corrective Action if Out
of Specification
Review SHAL blanks and
SHAL loading and
unloading procedures N/A
retroactive troubleshooting
and/or data validation after
notification by support
laboratory
Review SHAL blanks and
SHAL loading and
unloading procedures
Samples or
Channels
one per channel
one per
cartridge
one per channel
Carbon sampler
uses only one
channel per
event
PTFE Module
Randomly
select one of
loaded cassettes
(a) The number and location of sites that have produced collocated measurements in the CSN have been constant
through the life of the program at least through 2011. It is anticipated that some of the sites may be relieved of this
duty and a number of samplers redeployed to other sites for 1-year terms.
(b) See 40CFR part 58 appendix A section 4.2.1.
(c) This is a new QC check instituted as a warning mechanism for abnormal or deteriorating sampler performance.
The Average Flow CV is reported by the sampler at the end of every sampling event along with max/min
temperatures and pressure. The MetOne and the URG calculates the CV from an average of all polled data from
the onset of the sampling event. The average CV has always been computed by the MetOne and recorded on the
even COC and field data form. It will be included in the batch event result files upon implementation of the new
MQOs specified in this QAPP revision, approximately January 1, 2012.
(d) The maximum expected CV for the MetOne was reported in a conversation between Dennis Grumpier, EPA,
OAQPS and Michael Putnam, Technical Service Manager for MetOne, September 26, 2011. The expected CV for
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 7 of 20
the URG flow rate should be less than 1 % based on a couple of years of data captured from the URG memory
card.
(e) The one-point flow rate check should be taken when sampler is supposed to be operating at its design flowrate
(6.7 L/min in the case of the MetOne SASS sampler, and 22.0 L/min for the URG3000N); monthly flow check or
audit result should be within ± 5% of design flow.
(f) Over the first 10 years of the program, the frequency of blank generation has been decreased based on the
consistency of the data that has been generated from them and the routinely low contamination values. An SLT
and EPA may call for additional trip and or field blanks if results suggest evidence of contamination at one or both
steps in the process.
(g) Based on the highest mean+2o value (1 std dev added to the mean) in the first 3 years of operation. See Annual
Data Summary Report for the Chemical Speciation of PM25 Filter Samples Project at
http://www.epa.gov/ttn/amtic/files/ambient/pm25/spec/
(h) Based on the first year 2010 of SHAL data provided by RTI and DRI.
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 8 of 20
Exceeds ± 5 percent
of working standard
Note deviation on data form; review last 6 monthly flow
checks and CV data from downloads of previous 2 months
monitoring results;
The first occurrence
out of specification
The 2nd occurrence out of
specification in 3 months
Recalibrate and
conduct recheck;
If Satisfactory,
observe
performance in
following month
If fails to calibrate,
Troubleshoot and
consult with
Manufacturer
Full Refurbishment or
replacement. (Ask
manufacturer for a
leaner for immediate
deployment)
Service or Repair
Figure 14.1 Monthly Flow Check Decision Tree for Corrective Actions
14.1 Quality Control Checks
14.1.1 Checks
A parameter check, as used in this QAPP, is described as a verification, through the of a NIST traceable
reference standard, that the sampling instrument is operating within acceptance limits. Usually, if the
check shows a failure to meet specification, troubleshooting and instrument recalibration and final
verification should follow. Calibration is discussed in Section 16. Figure 14-1 is the current report form
that can be used to record and report the QC checks that should occur on a monthly basis. This form may
be revised occasionally to reflect improvements in procedures either in the field or in data handling. The
actual form is an Excel Workbook that can be downloaded from a website that is managed by the lab
support contractor for the CSN. (This website has restricted access to only those agencies or their
contractors that perform QC checks and audits.)
14.1.2 Audits and Independent Checks
The chemical speciation sampler(s) will be audited with an independent, National Institute of Standards
and Technology (NIST)-traceable transfer standard on at least a semi-annual basis. NIST-traceable
transfer standards are commercially available. Flow rate, temperature, and barometric pressure will be
checked. For samplers with multiple flow channels, each channel and the associated sensors will be
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 9 of 20
audited. The transfer standards may be verified against an agency's primary reference standard annually
but transfer standards should be certified by an independent metrology lab or the manufacturer's facility
every third year. Primary standards against which transfer standards are verified must be recertified or
recalibrated annually. Recertification or recalibration can occur at the manufacturer's facility or at a
metrology laboratory that uses current NIST-traceable equipment of a higher precision and accuracy.
The CSN laboratory must also assess the accuracy of its analytical measurements. This will include
assessment of field blanks and other samples supplied by the EPA quality assurance (QA) laboratory.
Refer to the CSN laboratory QAPP for more information.
14.2 QC Samples
Several types of standard QC samples are defined. The field program will mainly be concerned with
blanks sent from time to time by the CSN support laboratory to assess the effects of field operations and
shipping and handling.
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 10 of 20
Figure 14-2. CSN QA/QC Report Form
Chemical Speciation Network
Flow Check Worksheet
MetOne SASS - Primary Sampler
version
25
Note - Cyan fields are entered from URG worksheet or calculated - yellow fields are to be filled in here
Location
AQS Site ID
AQS Sampler
i
DOC 5
Site Information
Sampler Type (Model)
Operator(s)
Check Type
Date]
Missing Date, Site Name, Site Code, or POC
Select From Dropdown List
Flow Verification by Site Operator
Last Calibration Date
Affiliation
Sampler S/N
Head S/N
Pump S/N
Reference Standards
Flow Reference Std Model Select From Dropdown List
Specify if "Other"
Temperature
Ref Std Model | Select From Dropdown List
Specify if "Other"
BP Std Model | Select From Dropdown List
Specify if "Other"
Significant
Findings:
General
Findings:
Standard S/N
Calibration Date
Standard S/N
Calibration Date
Standard S/N
Calibration Date
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 11 of 20
MetOne SASS - Primary Sampler
Clock Check
If Local Time is under daylight savings, convert Ref Std to Local Standard Time. Daylight Saving Time begins
for most of the United States at 2:00 a.m. on the first Sunday of April. Time reverts to standard time at 2:00 a.m.
on the last Sunday of Octobe
Check Time
Date
Recalib Time
Date
Time (hh:mm)
Ref Std SASS
1/0/1900
1/0/1900
Difference
Minutes
Missing Sampler Date!
5 minut
Pass
es or less?
Fail
Leak Test
Channel 1
Channel 2
Channel 3
Channel 4
Channel 5
Channel 6
Channel 7
Channel 8
Initial Check
A L/min
Channel 1
Channel 2
Channel 3
Channel 4
Channel 5
Channel 6
Channel 7
Channel 8
After Correction
B L/min
Greater than 0.10 L/min fails
Pass
Fail A
Fail B
Flow Check
For the referer
Channel 1
Channel 2
Channel 3
Channel 4
Channel 5
Channel 6
Channel 7
Channel 8
Retest after Cs
Channel 1
Channel 2
Channel 3
Channel 4
Channel 5
Channel 6
Channel 7
Channel 8
ce standard, enter "UR" for under range and "OR" for over range flow readme
L/min
Lower Limit
NA
NA
NA
NA
NA
NA
NA
NA
Ref Std Upper Limit
NA
NA
NA
NA
NA
NA
NA
NA
SASS
% Difference
SASS - Ref
s.
Within ±
Pass
10%?
Fail
libration
L/min
Lower Limit
NA
NA
NA
NA
NA
NA
NA
NA
Ref Std Upper Limit
NA
NA
NA
NA
NA
NA
NA
NA
SASS
% Difference
SASS - Ref
Within ±
Pass
10%?
Fail
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 12 of 20
MetOne SASS - Primary Sampler
Reference Standard vs Design Flow
Channel 1
Channel 2
Channel 3
Channel 4
Channel 5
Channel 6
Channel 7
Channel 8
Retest after Cc
Channel 1
Channel 2
Channel 3
Channel 4
Channel 5
Channel 6
Channel 7
Channel 8
L/min
Lower Limit
6.03
6.03
6.03
6.03
6.03
6.03
6.03
6.03
SASS
6.7
6.7
6.7
6.7
6.7
6.7
6.7
6.7
Upper Limit
7.37
7.37
7.37
7.37
7.37
7.37
7.37
7.37
Ref Std
% Difference
Ref- Design
Within ±
Pass
10%?
Fail
ilibration
L/min
Lower Limit
6.03
6.03
6.03
6.03
6.03
6.03
6.03
6.03
SASS
6.7
6.7
6.7
6.7
6.7
6.7
6.7
6.7
Upper Limit
7.37
7.37
7.37
7.37
7.37
7.37
7.37
7.37
Ref Std
% Difference
Ref- Design
Within ±
Pass
10%?
Fail
Ambient Temperature Check
Retest After R(
Degrees C
Lower Limit
NA
Ref Std Upper Limit
NA
SASS
Difference
Within ± 2
Pass
I degrees?
Fail
^calibration
NA
NA
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 13 of 20
Metone SASS - Primary sampler
Filter Temperature Check
Channel 1
Channel 2
Channel 3
Channel 4
Channel 5
Channel 6
Channel 7
Channel 8
Retest After R<
Channel 1
Channel 2
Channel 3
Channel 4
Channel 5
Channel 6
Channel 7
Channel 8
Degrees C
Lower Limit
NA
NA
NA
NA
NA
NA
NA
NA
RefStd Upper Limit
NA
NA
NA
NA
NA
NA
NA
NA
SASS
Difference
Within ± ;
Pass
I degrees?
Fail
^calibration
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Pressure Check
Retest after re
mm Hg
Lower Limit
NA
RefStd Upper Limit
NA
SASS
Difference
Within d
Pass
: 10 mm?
Fail
calibration
NA
NA
14.2.1 Blanks
Field Blanks—These provide an estimate of total measurement system contamination. By comparing
information from laboratory and SHAL blanks against the field blanks, the amount of contamination due
to field activities can be estimated. In addition, if trip blanks are utilized, one can further evaluate
contamination occurring during field operations. Field blanks, loaded in sampling modules, for each type
of filter will be sent from the laboratory. The field operator is to handle the field blank sampling module
just as he/she would a module to be exposed but without drawing a sample through it. Corrective actions
will be taken when excessive contamination is found on field blanks.
The CSN support laboratory will also determine blank concentrations of analytes for each lot of Teflon™,
nylon, or quartz fiber filters received. The CSN support laboratory's QAPP and its SOPs discuss the
procedures.
Trip Blanks—These provide an estimate of measurement system contamination during transport to and
from the field sites. Trip blanks are usually instituted when field blank contamination is a problem or to
understand the measurement uncertainty introduced by contamination during transport. Trip blanks are
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 14 of 20
sent to the field as a normal sample but remain unopened. They are processed as a normal field sample
and sent back to the support laboratory and treated as a routine sample from the point of sample receipt
and beyond. As a practical matter, trip blanks are a measure of contamination introduced during the
loading of filters into cassettes and cassettes into the sampling modules or cartridges. The number of trip
blanks as percentage of the total number of filters (of each type) has been reduced over the life of the CSN
due to the very small values that have been observed. The MQO plan, as of 2011, is to eliminate the trip
blanks for the MetOne filters and reduce the URG trip blanks to 2 percent of the routine filters that are
shipped for sampling. If episodes of sustained high loads of contamination occur trip blanks may be
reinstituted to assist in determining the source(s) of contaminants.
SHAL Blanks—These blanks are generated by the lab service contractor to assess how much
contamination might be introduced to the filters during the loading into the cassettes and canisters or
cartridges. These are more informative than trip blanks as a rule. They will replace the MetOne trip blanks
that were collected as QC controls during the first 12 years of the STN. The URG 3000N deployment
instituted SHAL blanks at its inception in 2008.
14.3 Collocated Samplers
Collocated sampling occurs at approximately 4 percent of sites in the CSN (6 sites listed in Section 7) and
4 of which are in the NCore network. Data sets from collocated samplers are intended to assess the
precision of the total sampling, analysis, and data handling process, but they can also be very useful in
troubleshooting sampler siting and operational problems. Thus, such data are useful in detecting quality
problems that may not be evident from the results of periodic QC checks of flow rates, temperature, and
pressure. Refer to Section 14.4.2 through 14.4.4 for discussion and procedures for estimating bias and
precision from collocated sampler data sets.
14.4 Calculations of Accuracy, Bias, Precision, and Completeness
Accuracy is defined as the degree of agreement between an observed value and a true value or an accepted
reference value. It includes a combination of random error (precision) and systematic error (bias). The
following four accuracy/bias-related checks and audits are conducted in the chemical speciation program:
Flow Rate, Temperature, Barometric Pressure, and Other Checks Against a NIST-Traceable
Standard
The ability to separate PM2.5 from larger size particles is due to the inherent design of the separation
device (whether a cyclone or impactor) in a given sampler. The performance of the separation device is a
function of the flow rate, i.e., velocity of the air stream that is conducted through it. Consequently, the
ability of the sampler to maintain the designed flow rate is a critical performance parameter. The periodic
measurement of the sampler's actual flow rate by a NIST-traceable reference standard constitutes the most
accepted surrogate for measuring the accuracy of the sampler's results. In this case, the deviation might be
called a bias since the cut point of the PM would be direct!onally influenced by whether the velocity is
less than or greater than the design value for the given separator. A single flow check or audit, however,
only indicates the potential bias by the sampler since the last flow check or audit in which the flow was
within advisory limits or acceptance criteria. Systematic bias by the sampler can only be established by a
series of flow checks and calibrations. If the flow checks reveal that the flow rate varies plus and minus of
the calibration set point then there is no systematic bias. If the flow checks reveal the flow rate always
drifts in one direction from the calibration set point, then there is a systematic bias. However, it should
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 15 of 20
never fall outside of acceptance limits without a remedial follow-up. Consequently, long-term bias should
not occur.
< Collocation with a Federal Reference Method (FRM) sampler (gravimetric only)
< Collocation with another PM2.5 speciation sampler (all analytes)
< Gravimetric balance checks and other laboratory performance audits.
Balances are challenged with class 1 metallic reference standards. Analytical procedures are challenged
with known reference and internal standards and by interlaboratory performance evaluations of single
blind samples prepared by a facilitating laboratory, EPA's Office of Radiation and Indoor Air/National
Air and Radiation Environmental Laboratory in Montgomery, Alabama.
14.4.1 Percent Difference of a Single Measurement
Specifically for: Temperature;
Barometric pressure;
Other checks against a known standard
The error inherent in any single measurement is a function of both the underlying bias and the imprecision
(noise) in the measurement. Only after repeated measurements over a period of time is it possible to
separate bias and precision.
The percent difference, dt, is calculated from a standard of known value or an accepted reference value.
Calculate the percentage difference (4) for a single calibration check, /', using the following equation:
Equation 1
meas-audtt^
audit
Where:
di = percentage difference for a single parameter check
audit = parametric value as measured by the audit standard used in the QC check
meas = parametric value indicated by the monitoring organization's instrument
With respect to the CSN, Equation 1 applies to concentrations or flow rates, and not to Temperature (T)
and Barometric Pressure (P). For T and P, agreement is usually expressed in terms of absolute differences
in measurements (i.e., within +/- 2°C or within 10 mm Hg).
Flow rate:
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 16 of 20
Equation 1 could be used to characterize a test of the sampler's calibrated flow rate. This determination is
adequate if the sampler's "meas" flow is at or near the design value for achieving the proper separation
and collection of PM2.5 for the separation device used with that sampler. However, if the "meas" flow is
substantially different than the design flow rate it may actually pass the acceptance criteria according to
the equation in 40 CFR part 58 Appendix A for flow check, but it could generate a significantly inaccurate
sample of PM2.5. For example, if a MetOne speciation sampler says its flow rate is 9.8 1pm and the flow
standard indicates it is 10.0, the value is a perfectly acceptable 2.0 percent low but the cut point is 10
percent or more lower than it should be, because the actual flow rate is 30 percent too high. In other
words, the sampler is performing perfectly wrong. However, the audit value needs to be compared to the
design flow rate of the sampler to have any real relevance to accuracy. Consequently, a second equation
should also be used for single point flow rate checks and audits:
Equation la
= audit - design .
Cl -r 1UU
design
Where:
difr = percentage difference for a single flow rate check
audit = actual flow rate of sampler based on the measurement by a NIST-Traceable
standard
design = the design flow rate of the sampler as specified by the manufacturer
14.4.2 Percent Difference for Collocation with a Federal Reference Method (FRM) Sampler
(Gravimetric Only) and Collocation With Another PMi.5 Speciation Sampler (All
Analytes)
If one takes the position that any given FRM sampler is a representation of the true PM2.s mass sampled at
its location, the comparison of the total mass derived from the FRM sampler could arguably be a basis for
a relative accuracy or bias of a speciation sampler that is collocated with it. If a number of collocated
measurements indicate that the total masses measured by both are essentially equal, then we can postulate
that at least the speciation samplers are relatively accurate with respect to the FRM total mass
measurements. If they are different, then we must determine if the difference is systematic, a true bias, or
if it is random. It has been established that the masses determined by FRM vs. the CS samplers are
statistically different.1 The dominant sampler in the speciation network, the MetOne SASS or SuperSASS,
does not use the same flow rate as the PM2.s speciation samplers based on the FRM design; it is
considerably lower, i.e., 6.7 liters per minute. Therefore, it uses a differently designed separator than a
PM2.5 FRM sampler. The speciation samplers that were designed to pull 16.67 liters per minute through a
Teflon filter were replaced by MetOne SASS samplers in 2008. The comparison of the total gravimetric
mass derived from a speciation sampler to that obtained by a collocated FRM sampler may be more
accurately described as a measurement of bias. It must be noted that the recovery, shipping and storage
1 See Evaluation of PM2.5 Chemical Speciation Samplers for Use in the EPA National PM2.5 Chemical Speciation Network.
EPA-454/R-01-005
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 17 of 20
procedures of speciation filters is also different than FRM PM2.5 filters, which could also create more
randomness due to the character of the aerosols that are collected in different areas of the country. There
are no other independent constituent analyses so that a similarly calculated bias for the ions and carbon are
not possible.
Precision is estimated for manual instrumentation via duplicate measurements from ostensibly identical
collocated samplers providing filter samples containing analytes above minimum detectable
concentrations.
Since there are only six collocation sites across the CSN, precision can only be aggregated at the network
level on a quarterly, annual, and the multi-year basis. For each collocated data pair, the relative percent
difference, dt, is calculated by Equation 2.
Equation 2
-100
Where:
dt = the percent difference of concentration of the selected analyte (|ig/m3) determined for the
samplers, divided by the average of the two concentrations,
Xt = the concentration of the primary sampler, and
Yi = the concentration value from the, i.e., the collocated sampler
14.4.3 Calculation of Bias
Bias is a systematic deviation from the true value. Data must be averaged or aggregated over a period of
time or over a set of measurements in order to assess bias. Bias can be used to assess systematic errors of a
single sampler, a reporting organization, or an entire network. For example flow rate checks over a year
can provide some sense of whether the network sampler(s) flow rate drifts positive or negative from the
calibration value over a period of time. When the chemical speciation network was first deployed there
were several sites to which all three of the legacy samplers were deployed to determine if there was a
relative bias introduced by one, two or all of them. A report suggested that bias was present but was site-
and constituent-specific.
The following equation is used to calculate bias:
Equation 3
AS
where n is the number of single point checks being aggregated; to.95,n-i is the 95th quantile of a t-
distribution with n-1 degrees of freedom; the quantity AB is the mean of the absolute values of the dt 's
2 Evaluation of PM2.5 Chemical Speciation Samplers for Use in the EPA National PM2.5 Chemical Speciation Network, Paul
A. Solomon, et.al EPA-454/R-01-005, May 2001. http://www.epa.gov/ttn/amtic/files/ambient/pm25/spec/fourctv.pdf.
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 18 of 20
(calculated by Equation 2) and is expressed in Equation 4 as follows:
Equation 4
i=\
and the quantity^ is the standard deviation of the absolute value of the dj's and is calculated using
Equation 5 as follows:
Equation 5
AS=
Since the bias statistic as calculated in Equation 3 above uses absolute values, it does not have a tendency
(negative or positive bias) associated with it. A sign will be designated by rank ordering the percent
differences (4's) of the QC check samples from a given site for a particular assessment interval. Calculate
the 25* and 75* percentiles of the percent differences for each site. The absolute bias upper bound should
be flagged as positive if both percentiles are positive and negative if both percentiles are negative. The
absolute bias upf
straddling zero).
absolute bias upper bound would not be flagged if the 25th and 75th percentiles are of different signs (i.e.,
Bias can also be expressed as a percentage of the standard or expected value. This definition is applicable
to flow rate, temperature, barometric pressure, and derived quantities such as air volume through a filter.
Laboratory biases can be calculated in a similar way.
14.4.4 Calculation of Precision
Precision defines the random variability of a set of measurements and excludes systematic bias. The
precision estimate is used to assess the one-point QC checks for repeated parametric measurements or
laboratory measurements of target analytes (soil elements, nitrate, sulfate, ammonium, OC, and EC). The
precision estimator is the coefficient of variation upper bound and is calculated using Equation 6 as
follows:
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 19 of 20
Equation 6
where j^ o.i,n-i is the 10th percentile of a chi-squared distribution with n-1 degrees of freedom.
Precision can be used to assess the random errors inherent in a single sampler, a reporting organization, or
an entire network. Random errors, characterized by the precision calculations, can be used as input to the
DQO model for the assessment of data trends. Excessive variability or noise in data can also indicate
equipment malfunction and the need for corrective action.
There are several other ways of expressing precision of a measurement, including the CV and the
confidence interval. The precision of a measurement is often a function of several different sources of
random variation. For example, the precision of a calculated particulate concentration may be a function
of the variability of measurement in calculating the flow rate, sampling temperature, and pressure or the
variability introduced in shipping/handling and by the laboratory. Assessment of these errors is beyond the
scope of this QAPP but should be considered when performing advanced statistical analysis of the data.
14.4.4 Calculation of Completeness
The goal for completeness is 75 percent. Meeting the DQO for identification of a systematic trend in the
data requires that the data meet minimum completeness criteria. Data must span the time range of interest,
and there must be a sufficient number of measurements to reduce the statistical uncertainties. The DQO
analysis found that l-in-3 day sampling was sufficient to meet the DQO for assessing a data trend after
5 years provided other uncertainties described above are controlled. Significant data loss, however, could
compromise the ability to assess the DQO.
Completeness should be assessed quarterly and annually as shown in the following equation:
N - N
%C=— - ^^xlOO
Nt
where
%C = percentage completeness
Nt = the total number of possible or potential measurements in the data set
= the number of missing or invalidated measurements
Assessment of completeness is applicable to derived measurements such as elemental concentration in the
air (for assessment of the DQO). Maintaining the level of data completeness ensures that the minimum
number of valid concentration data points are available to meet the DQO for speciation trends
identification after 5 years of monitoring. To be counted for completeness, a sample must have passed all
of the various screens in the data validation process. Thus, a particular sample can be invalidated in
-------
PM2 5 CSN QAPP
Section No. 14
Revision No.: 1.2.0
Date: 03/12
Page 20 of 20
several different ways. Completeness assessment is an important element in the oversight of operational
and laboratory activities. Poor or marginal completeness figures should prompt the reporting organization
to re-evaluate and improve operating procedures.
14.5 References
U. S. EPA (Environmental Protection Agency). 1998. Quality Assurance Guidance Document 2.12.
Monitoring PM2.5 in Ambient Air Using Designated Reference or Class I Equivalent Methods. April 1998.
U.S. EPA (Environmental Protection Agency). 1999. Strategic Plan: Development of the Paniculate
Matter (PM2.s) Quality System for the Chemical Speciation Monitoring Trend Sites. April 16, 1999.
U.S. EPA (Environmental Protection Agency). 2006. Code of Federal Regulations, Part 58, Appendix A,
Section 4, "Calculations for Data Quality Assessment. "
-------
PM2 5 CSN QAPP
Section No. 15
Revision No.: 1.2.0
Date: 10/11
Page 1 of4
15.0 Sampler/Equipment, Testing, Inspection and Maintenance
Requirements
This section discusses procedures to test, inspect, and maintain PM2.s speciation samplers and
support equipment to be used at Chemical Speciation Network (CSN) field sites.
15.1 Testing and Acceptance Criteria
The CSN employs a set of testing and acceptance criteria that were used to accept Federal
Reference Method PM2.5 gravimetric samplers at 40CFR Part 53, Section 53.34. Table 15-1 lists
the 13-step procedure by which final acceptance of each of the three originally deployed
speciation sampler occurred. URG 3000N samplers added to the CSN program in 2007-2009
were not available for participation in the original intercomparison studies. However, they were
subjected to a similar initial field tests and the first 59 were run side-by-side with MetOne SASS
or SuperSASS samplers (on which quartz filters were loaded at channel 3) for a period of at least
10 sampling events. The collocated events were reduced in deployments 2 & 3. At one point, 13
CSN sites also have run IMPROVE samplers for network intercomparison data. These sites
provided yet another set of sampling results to compare with the performance of the URG 3000N
samplers. The standard operating procedure for the URG 3000N and the Operators Manual both
provide start-up testing and acceptance criteria. See
http ://www. epa.gov/ttn/amtic/spectraining.html.
Sampler calibration and verification devices also will require testing and acceptance upon
receipt. Each of the following devices must arrive with a certificate of National Institute of
Standards and Technology (NIST) traceability, an instruction booklet, and a warranty statement.
Simple testing and acceptance methods are listed below:
• Flow rate transfer standard (check against another flow rate standard).
• Temperature transfer standard (check against a NIST-traceable thermometer,
(preferably a thermistor or thermocouple-based sensor that does not use mercury), ice
bath, or other temperature standard).
• Pressure transfer standard (check against another barometer or against a corrected
barometric pressure reading from a nearby national weather station).
Testing and acceptance criteria for the CSN contracted support laboratory are discussed in the
Quality Assurance Project Plan (QAPP) prepared by the CSN support laboratory.
15.2 Maintenance
Many items require maintenance in the CSN. Preventive maintenance should be practiced in
accordance with the "manufacturer's instructions" given in the operators' manuals. This section
describes maintenance items for CSN field equipment. The CSN contracted support laboratory's
QAPP discusses laboratory equipment maintenance. Maintenance of speciation samplers is
becoming increasingly important as the equipment ages. CSN site operators and managers
-------
PM2 5 CSN QAPP
Section No. 15
Revision No.: 1.2.0
Date: 10/11
Page 2 of4
should read and follow maintenance instructions given in the sampler operator's manual and
updates and any maintenance bulletins that may be issued. Table 15-2 lists field items known to
require preventive maintenance and periodic recertification. Others may be added to this list as
experience is gained. A 3-year factory service program for maintenance of samplers is planned
for implementation.
Table 15-1. Testing and Acceptance Criteria Checklist for
Speciation Samplers
Check or Criteria
1 . Check the enclosed packing list. Were all parts included in the delivery of the
sampler? (Including filter packs, denuders, and an operator's manual)
2. Were any of the enclosed parts broken during shipment of the sampler?
3 . Check the enclosed assembly instructions. Did all the parts fit together during
assembly of the sampler?
4. Does the sampler's pump turn on when supplied with electrical power?
5. Using an independent timer, check to ensure the timer (clock) operates properly.
Check to see if the timer will automatically turn the sampler on and off during a
fixed time by setting the timer to turn the sampler on and off for a short period
while the operator observes.
6. Does the sampler's computer (display screen) boot up and operate properly? Check
to see if the computer (microprocessor) has working software (firmware) by
performing manual input of information to the sampler through the keyboard and
observe the results on the display screen.
7. Does the computer download information properly? Check this process by
attempting to download stored information through the RS232 port to a computer
and then to other data transfer devices; or .
8. Does the internal cooling fan operate properly? Check this function by supplying
electrical power to the unit and listening for and observing the fan turn on and off.
9. Do the temperature sensors operate properly? Check this function by comparing
sensor readings to an independent thermometer (or by warming the sensor tip
between your fingers to see if the temperature rises).
10. Does the filter holder (sampling module, cassette, or filter pack) hold the filter
correctly and does it connect easily and snugly to other components of the sampling
stream? Check this function by installing a filter in the sampling module and by
connecting the module to the sampling manifold, denuder, and so on.
11. Does the sampler's casing or other enclosure protect internal units from the
weather? Check by visually inspecting the unit's latches, locks, seals, and gaskets
for gaps, holes, and leaks. Do not disassemble the unit.
12. Does the support structure (base, tripod, and so on) hold the sampling unit secure,
upright, and level?
13. When all sampler parts are assembled and operated as a unit, does the sampler
function properly? Check this function by assembling the unit per instructions,
installing a filter or sampling module, setting the timer, activating the sampler, and
running it for 24 hours as would be done for routine sampling.
Certifying Official's Signature Date Circle one —>—>
Yes?
Accept?
No?
Reject?
-------
PM2 5 CSN QAPP
Section No. 15
Revision No.: 1.2.0
Date: 10/11
Page 3 of4
Table 15-2. Preventive Maintenance and Recertification of CSN Field Equipment
Maintenance Item
Recommended Frequency
PM2.5 Speciation Samplers
2. 3. Check sampling inlets, and URG downtubes for bugs and obstructions, and
water intrusion
1. Clean sampler inlet surfaces.
2. Examine O-rings.
3 . Clean interior of sampler case (if applicable) .
4. Inspect denuder for breakage (RASS and MASS only). Replace denuders with
freshly coated ones and return used denuder to laboratory for refurbishment.
5. Inspect and service cooling air intake filter and fans.
1. Inspect O-rings of inlet assembly (if applicable). Apply very light coat of
vacuum grease if required.
2. Clean sampler downtube (unless it contains a denuder).
3. Inspect and service O-rings of inlet and water seal gasket at downtube entry to
case if applicable.
4. Clean cyclones and manifolds upstream of sampler module.
5. Inspect and service O-rings in sampler head or platform assembly of URG
3000N.
6. Inspect and service vacuum tubing, tube fittings, and other connections to
pump and electrical components.
7. Overhaul or replace sampling pump and solenoids.
Each visit to site
Every 30 sampling events
or more often as needed or
as specified by the network
Quarterly (every 3 months)
Per vendor guidance
Calibration and Check Devices
NIST-traceable flow rate transfer standard
1. Recertify vs. NIST standards.
2. Replace batteries (if applicable).
3 . Visually check orifices for dust or breakage.
NIST-traceable temperature transfer standards (digital thermometer)
1. Recertify vs. NIST standards.
2. Replace batteries.
3 . Inspect probe tip and connecting cord.
NIST-traceable pressure transfer standard
1. Recertify vs. NIST standards.
2. Replace batteries.
Annually
As needed
Each use
Annually
As needed
Each use
Annually
As needed
15.3 Critical Spare Parts
Maintain an inventory of critical spare parts at the field office to prevent sampler downtime or
interruption of the required QC checks. The CSN contracted support laboratory will also
maintain a spare parts inventory to service the sampler modules and denuders. The field site
should send the laboratory any spare parts associated with the components the support laboratory
is supplying to the site (for example, O-rings that are part of the sampling module that is sent
back and forth between the contracted support laboratory and field site) so that these components
may be serviced at the contracted support laboratory.
-------
PM2 5 CSN QAPP
Section No. 15
Revision No.: 1.2.0
Date: 10/11
Page 4 of4
Speciation sampler
O-rings for downtube connections (if applicable)
Vacuum/pressure tubing and connecting compression or other types of fittings
Filter packs or cassettes (to be forwarded to CSN support laboratory)
Denuders (to be forwarded to CSN support laboratory)
Sampling lines/tubing (as applicable)
Fuses
Digital thermometer, pressure device and flow transfer standard
Spare batteries
Spare temperature probe
Spare tubing
-------
PM2 5 CSN QAPP
Section No. 16
Revision No.: 1.2.0
Date: 10/11
Page 1 of6
16.0 Instrument Calibration and Frequency
16.1 Overview
Calibration is the comparison of a measurement standard or instrument with another reference
standard, a reference material, or reference instrument to quantify and report, any difference or
variation (deviation) from the reference value. The purpose of calibration is to minimize bias;
therefore, calibration may include adjustment of the performing measurement device or
instrument so that it measures and reports values that closely replicate those same values
represented by standard reference materials or reference devices. For PM2.5 chemical speciation
instruments, calibration activities follow a two-step process:
1. Certifying the calibration standard and/or transfer standard against an authoritative
standard (usually those produced by the National Institute of Standards and
Technology (NIST) or reference standards certified to be a certain accuracy based on
procedures developed by the NIST).
2. Comparing the routine sampling or analytical instrument against a calibration or
transfer standard.
Parameters of the chemical speciation samplers that are subject to routine calibration checks in
the field include the following:
• Flow rate (all filter channels).
• Ambient temperature (one per instrument).
• Filter or manifold temperature (one per channel or instrument).
• Barometric pressure (one per instrument).
Calibrations that involve permanent changes due to instrument adjustments should only be
initiated when it is obvious that a measurement parameter does not meet its acceptance criteria.
Therefore, the Chemical Speciation Network (CSN) uses a two-tiered approach to calibration
that involves the following:
• Checks (see Section 14) to ensure that calibration is within acceptance criteria.
• Multipoint calibration when there is a failure during a one-point check or audit.
Instrument adjustments occur during multipoint calibrations and are followed by a
one-point check to ensure that the transfer check standard is also operating properly.
16.2 Calibration and Verification of Field Instrumentation
Calibrations and checks for the chemical speciation samplers should generally follow the
schedule set for the PM2.5 Federal Reference Method (FRM) total mass monitoring program. The
speciation samplers are similar to the PM2.5 FRM filter-based samplers in that they draw air
through a filter. They also measure and record ambient temperature, barometric pressure,
date/time and elapsed time of the sampling event. The samplers also poll the flow rate on a
-------
PM2 5 CSN QAPP
Section No. 16
Revision No.: 1.2.0
Date: 10/11
Page 2 of6
prescribed schedule and calculate the coefficient of variation in flow over the entire sampling
event. They may differ from the PM2.5 FRM samplers; however, with respect to the flow rates,
the number of sampling channels that need to be calibrated, and the number and location of
internal temperature sensors, and the filter media that is used to collect the filtrate in each
different channel. The URG 3000N is patterned after the IMPROVE network's C module that
collects carbon.
After receipt and acceptance of a new chemical speciation sampler, single point calibrations of
temperature, barometric pressure, and multipoint calibrations of the flow rate sensors will be
performed. After installation, regular checks and maintenance are carried out by the CSN site
operator at the specified intervals. During each quarter, an internal audit is performed using an
independent set of standards. These checks are detailed along with the acceptance criteria in
Table 16-1.
The following calibrations are performed in the field:
• Verification/calibration of sampler's temperature probes and against the working
temperature standard,
• Verification/calibration of the sampler's barometric pressure sensor against the
working pressure standard,
• Verification/calibration of volumetric flow rate meter in the sampler against the
working flow rate standard, and
• Verification of the sampler's internal clock against an accurate timepiece.
Temperature Probes—Each CSN chemical speciation sampler has both ambient and internal
temperature probes. The CSN site operators will perform one-point field verifications of both
sensors as needed or at least every month using a digital NIST-traceable temperature probe. A
quarterly temperature audit will be performed using an independent temperature standard.
Barometric Pressure—A NIST-traceable digital handheld pressure indicator will be used in the
field for one-point check of the sampler's pressure sensor as needed or at least every month. A
quarterly pressure audit will be performed using an independent pressure standard. A NIST-
traceable digital manometer will be used in the field office or field as a primary standard.
Time Sensor—The time sensor should be within + 5 min of a watch that has been recently
checked and set. Time (operator's watch) can be set against an atomic clock that can be found on
the Internet (available online at http://www.time.gov) or through a phone number (303-499-
7111).
Flow Rate—As needed or at least monthly, a one-point flow rate verification will be performed
for each sampling stream using a NIST-traceable flow rate transfer standard. A quarterly flow
rate audit will be performed using an independent flow rate standard. The NIST-traceable
calibration standard will be used in the field office or field as a primary standard to perform
multipoint calibrations once a year or after a one-point verification failure.
-------
PM2 5 CSN QAPP
Section No. 16
Revision No.: 1.2.0
Date: 10/11
Page 3 of6
Calibration Standards—Calibration standards for the temperature, barometric pressure, and flow
rate verifications and calibrations are given in Table 16-2. All calibration standards must be
recertified annually. The recertifications must be traceable to NIST. These recertifications may
be done by the standards' manufacturer or by a third-party metrology laboratory and must be
performed in accordance with American National Standard for Calibration - Calibration
Laboratories and Measuring and Test Equipment - General Requirements (ANSI/NCS1
Publication No. Z540-1-1994). Records of all certifications must be maintained, including the
identity of the NIST reference, the procedure used to establish traceability, and a certificate of
traceability.
Temperature, barometric pressure, and flow rate transfer standards that are used to perform
routine verifications of CSN chemical speciation samplers should be recertified at the same
intervals as the transfer standards for the PM2.5 gravimetric samplers. When both gravimetric
samplers and chemical speciation samplers are situated at the same monitoring location, the same
equipment and transfer standards may be used for both sets of samplers (if they are compatible
with all samplers).
Calibration Procedures—Procedures for temperature, barometric pressure, and clock/timer
calibrations are typically identical to those used for the PM2.5 gravimetric samplers from the
same manufacturer, and it may be possible to follow standard operating procedures (SOPs)
developed for the gravimetric samplers. Procedures for performing flow rate calibrations and
leak checks for the chemical speciation samplers may differ from the gravimetric sampler by the
same manufacturer. Chemical speciation samplers typically have several different air sampling
streams, each of which is used to sample a different group of chemical species. Whenever
possible, the individual flow rates should be calibrated independently of each other. In some
models, however, this may not be possible due to the use of vacuum manifolds and passive flow
rate controls (e.g., flow control orifices). The manufacturer's procedures for flow rate calibration
should be followed.
Calibration Frequency—See Table 16-1 for a summary of calibration frequencies.
Documentation—All verifications and calibrations, as well as sampler and calibration equipment
maintenance, will be documented in field data records and notebooks and annotated with the
flags required in Appendix L of 40 Code of Federal Regulations (CFR) Part 50, the
manufacturer's operating instruction manual, and any others indicated in the field and laboratory
SOPs (refer to Appendices of this Quality Assurance Project Plan for SOPs). The records will
normally be controlled by the CSN site operators, and they will be located in the field offices or
field collection sites when in use. Eventually, all calibration records will be appropriately filed
(see Section 9.0).
16. 3 Calibration and Verification of Laboratory Instrumentation
Calibration and verification of laboratory analytical equipment will follow the procedures given
in Quality Assurance Project Plan: Chemical Speciation ofPM2.s Filter Samples, prepared by
the CSN contracted support laboratory.
-------
PM2 5 CSN QAPP
Section No. 16
Revision No.: 1.2.0
Date: 10/11
Page 4 of6
Table 16-1. Acceptance Criteria and Calibration and Maintenance Frequencies for
Chemical Speciation Samplers
Criteria
Acceptance
Criterion or
advisory limits
Frequency
SOP
Comments
Field Calibrations and Routine Checks (by Operator or Site Supervisor)
One-point flow rate
check at design flow
rate
External leak
check(a)
Internal leak check
One-point
temperature check
Pressure verification
Clock/timer
verification
Other calibrations as
specified by
manufacturer
±5% of transfer
standard; and ±5%
of design flow rate
MetOne: <0.1
L/min
URG<0.55L/min
Both signal
pass/fail;
MetOne: <0.1
L/min
URG < 0.55 L/min
Both signal
pass/fail;
±2 °C of standard
ilOmmHg
1 min/mo
per manufacturer's
SOP
Monthly
Conducted with
monthly flow check
If external leak
check fails
Monthly
Monthly
Monthly
Per manufacturer's
SOP
Refer to
MFGR's
operating
manual
Same as for gravimetric
samplers. Applies to all flow
channels.
Same as for gravimetric
samplers. Applies to all flow
channels.
Same as for gravimetric
samplers. Performed as a
troubleshooting procedure
only. May not be applicable
to all sampler designs.
Same as for gravimetric
samplers. Applies to all
temperature sensors.
Same as for gravimetric
samplers.
Same as for gravimetric
samplers.
Quarterly Checks and Audits (by Auditor or person 2 mgmt. tiers from operator, using different equipment)
External leak
check(a)
Internal leak check
Temperature audit
MetOne: <0.1
L/min
URG < 0.55 L/min
Both signal
pass/fail;
MetOne: <0.1
L/min
URG < 0.55 L/min
Both signal
pass/fail;
±2°C
Semi-annual unless
failed audit then at
least quarterly until
passes for 2 quarters
If external leak
check fails
Semi-annual unless
failed audit then at
least quarterly until
passes for 2 quarters
Refer to
MFGR's
operating
manual
Same as for gravimetric
samplers. Applies to all flow
channels.
Same as for gravimetric
samplers. Performed as a
troubleshooting procedure
only. May not be applicable
to all sampler designs.
Same as for gravimetric
samplers.
-------
PM2 5 CSN QAPP
Section No. 16
Revision No.: 1.2.0
Date: 10/11
Page 5 of6
Criteria
Pressure audit
Flow rate audit
Acceptance
Criterion or
advisory limits
ilOmmHg
±5% of audit
standard
±5% of design flow
rate
Frequency
Semi-annual unless
failed audit then at
least quarterly until
passes for 2 quarters
Semi-annual unless
failed audit then at
least quarterly until
passes for 2 quarters
SOP
Comments
Same as for gravimetric
samplers.
Same as for gravimetric
samplers. Applies to all flow
channels.
Initial Installation Calibration and recalibrations thereafter
Temperature
calibration
Pressure calibration
Multipoint flow rate
calibration; URG
only; MetOne
utilizes single point
Design flow rate
adjustment
±2°C of standard
ilOmmHg
±2% of transfer
standard at each
flow rate
±2% of design flow
rate
On installation,
annually, or if
verification/audit
indicates drift or
failure
On installation, then
annually, or if
verification/audit
indicates drift or
failure
On installation,
annual , or if
verification/audit
indicates drift or
failure
As needed
Applies to ambient
temperature and all internal
filter temperature sensors.
Same as for gravimetric
samplers.
Applies to all flow channels
individually.
Applies to all flow channels
individually.
Sampler Maintenance
Inlet/do wntube
cleaning
Filter chamber
cleaning
Cyclone and
manifold cleaning
Pump box air supply
fan filter cleaning
Manufacturer-
recommended
maintenance
cleaned
cleaned
cleaned
cleaned/changed if
present
per manufacturer's
SOP
Lesser of every 15
sampling events or
every quarter
Monthly
Approximately
every 30 use days
Monthly
Per manufacturer's
SOP
Same as for gravimetric
samplers.
If applicable to sampler
design.
Consult operator's manual
Recertification of Standards (audit and calibration) *
Flow rate transfer
standard
±2%ofNIST-
traceable standard
Annually
Same as for gravimetric
samplers.
-------
PM2 5 CSN QAPP
Section No. 16
Revision No.: 1.2.0
Date: 10/11
Page 6 of6
Criteria
Field thermometer
Field barometer
Acceptance
Criterion or
advisory limits
±0.1 °C resolution,
±0.5 °C accuracy
±1 mmHg
resolution,
±5 mmHg accuracy
Frequency
Annually
Annually
SOP
Comments
Same as for gravimetric
samplers.
Same as for gravimetric
samplers.
Table 16-2. Calibration Standards for PM2.s Chemical Speciation Samplers
Description of
Calibration Standard
Digital thermometer
Digital pressure gauge
Flow meters
QA Objective
Acceptance
Criterion11
±0.5°C
±0.7%c
±2%
Listed Uncertainty
for Calibration
Standard
Manufacturer
of Calibration
Standard
Model Number
of Calibration
Standard
The MetOne provides a failure message based on a leak rate of > 0.1 liters per minute. The URG 3000N has
programmed a test based on loss of vacuum of the sealed sample train which roughly equals an infiltration rate of
0.08 liters per minute.
Acceptance criteria taken from Table 16-1 of the Quality Assurance Project Plan for the Federal PM25
Performance Evaluation Program (U.S. EPA, OAQPS Revision 1, December 2007.
The pressure criterion (±5 mmHg) is equivalent to ±0.7% of atmospheric pressure (760 mmHg).
-------
PM2 5 CSN QAPP
Section No. 17
Revision No.: 1.2.0
Date: 10/11
Page 1 of 3
17.0 Inspection/Acceptance for Supplies and Consumables
17.1 Purpose
The purpose of this section is to establish and document a system for inspecting and accepting all
supplies and consumables that may directly or indirectly affect the quality of operations at a
PM2.5 site of the Chemical Speciation Network (CSN). The monitoring network relies on various
supplies and consumables that are critical to its operation. This section gives details on the
supplies and consumables, their acceptance criteria, and the required documentation for tracking
the process. Note that the inspection and acceptance procedures for the filters, of which high
quality and consistency of performance is inextricably tied to data quality, is covered in the
Quality Assurance Project Plan (QAPP) for the Contract Support lab's operations. See
http://www.epa.gov/ttn/amtic/specguid.html.
17.2 Critical Supplies and Consumables for Field Site Operations
This section describes the supplies needed by the field sites. The choice of field supplies and
consumables is, in part, dictated by the choice of speciation sampler. Table 17-1 lists the major
items of equipment needed for the CSN. The state, local and/or Tribal (SLT) office must keep an
inventory of all supplies and the warranty period and certifications of equipment and can use
Table 17-1 for this activity. Since participants in the CSN are also operating routine samplers for
the mass PM2.5 program, the calibration and check standards may be used for the CSN since they
should meet the performance standards.
The contracted support laboratory will care for and track the cassettes and filter modules used in
servicing the sampling and analysis needs of the sites but will not carry them on its own
inventory list. For more information, refer to the support laboratory's QAPP and standard
operating procedures that are available on the EPA's AMTIC website.
As consumables run low or replacement purchases are required, the site operator will be
responsible for assisting in the procurement of these items by following the policies and
procedures of the SLT agency. The operator should purchase the same model equipment and
spare parts and the same consumables as were initially acquired, unless told to do otherwise.
17.3 Acceptance Criteria
Selection of major pieces of capital equipment is based on the item's advertised specifications
and performance in analysis of particulate matter and particulate matter extracts. Newly received
field equipment will be inspected to ensure all parts are present and undamaged. If damage has
occurred in shipping, the shipping agent will be notified. All new equipment for field or
laboratory use should carry a warranty for a 6-month to 1-year period. Refurbished equipment
should also be inspected carefully and subjected to operational tests since the warranty on such
equipment may not be as comprehensive.
-------
PM2 5 CSN QAPP
Section No. 17
Revision No.: 1.2.0
Date: 10/11
Page 2 of 3
Table 17-1. Inventory List for CSN Field Equipment and Supplies
Quantity
per Site
Operator
Equipment and Supply Description
Vendor/
Catalog No.
Make/Model No.
Speciation Sampling Equipment and Supplies
1
1
6 sets
Dependent
on sampler
type
Speciation sampler (a capital equipment item)
Speciation sampler operating manual
Filter cassettes or sampling modules for speciation
sampler
Denuders for acid gases. Magnesium oxide coated
aluminum honeycomb denuders in the case of the SASS
sampler for which approximately 6 will be required.
Custody and field data form for each sample run
(supplied by contracted support laboratory)
Sample shipping containers/ice packs (supplied by
contracted support laboratory)
Max./min. thermometer (if required)
Mounting Equipment and Tools
1
1
Tool kit
Jig for separating and rejoining URG Cassettes for Flow
Calibration/leak check cartridge
Supplied by EPA
OAQPS
Calibration/Check Standards and Related Equipment
1
1 or more
1
1
1 per site
and 1 for
Audits
Flow rate adapter (size depends on sampler brand)
Flow transfer standard (wet or dry type)
Pressure transfer standard (portable barometer)
Digital thermometer (or thermocouple calibrator)
URG Flow Calibration/leak check cartridge
See "Operation
Manual" Parts list
Spare Parts
O-rings, tubing, fuses, impactor oil (if applicable),
impactor filters (if applicable), compression fittings
Filters for Audit and Calibration cassettes and cartridges
Request from EPA,
OAQPS
Cleaning and Maintenance Supplies and Equipment
Distilled water in spray bottle
Lint-free cloths and lint-free laboratory wipes
Additional warranty periods may be purchased; check equipment stated to be NIST-traceable
that is subject to wear and tear during use (for example, temperature, pressure, and flow rate
-------
PM2 5 CSN QAPP
Section No. 17
Revision No.: 1.2.0
Date: 10/11
Page 3 of 3
check devices). Such equipment should be returned annually to the vendor or an appropriate
metrology laboratory for cleaning, servicing, and recertification vs. NIST standards.
SLT agency personnel should use procurement logs for purchases of new equipment and
consumables. These logs should also indicate whether the items were accepted or rejected. In
addition, the laboratory and field personnel must keep an equipment inventory form that lists
each piece of equipment and its warranty dates.
-------
PM2 5 CSN QAPP
Section No. 18
Revision No.: 1.2.0
Date: 11/11
Page 1 of2
18.0 Data Acquisition Requirements (Nondirect Measurements)
This section identifies the types of indirectly acquired data needed for implementation and
continuation of the Chemical Speciation Network (CSN). These data are obtained from
nonmeasurement sources and historic or concurrently acquired databases not under the direct
control of the CSN.
18.1 Acquisition of Nondirect Measurement Data
The CSN will produce almost all required data through its own field and laboratory operations.
Some data, however, will come from outside the network. This section lists several such data
sources, considers the quality of the data, and gives cautionary notes.
18.1.1 Chemical and Physical Properties Data
Chemical and physical properties data and values of fundamental constants are often needed in
ambient air studies. Examples of acceptable sources for fundamental units and constants and the
relationships between metric and U.S. or British units are:
• National Institute of Standards and Technology (available online at www.nist.gov).
• International Organization for Standardization (ISO) (available online at
www.iso.ch), International Union of Pure and Applied Chemistry (IUPAC) (available
online at www.iupac.org), American National Standards Institute (ANSI)
(www.ansi.org), and other national and international standards organizations.
• U.S. Environmental Protection Agency (available online at www.epa.gov) sources.
• Current editions of handbooks on chemistry and physics such as the Handbook of
Chemistry and Physics (CRC Press).
18.1.2 Sampler Operation and Manufacturers' Literature
Important information is found in the manufacturers' literature and operating manuals. Manuals
for the speciation samplers, the devices used to verify a sampler's proper operation (temperature
sensors, pressure gauges, and flow meters) and to calibrate it, data acquisition devices (laptop
computers and the programs they contain), and all analytical instrumentation used in the
laboratory will be available.
18.1.3 Site Location Information
The highest priority objective of the PM2.5 CSN data is the development of common spatial and
seasonal/annual compilations and displays of the concentrations of fine particle constituents
across the major urban areas of the country that can be used to determine if trends exist.
Characterization of PM2.5 constituents in rural or regional environments (especially when data
sets are combined with IMPROVE network data) is also possible if some sites are in transport
and/or background locations. To select the best locations for chemical speciation sites (especially
any new sites), the network designers rely on several external sources of information to minimize
the collection of samples with components that are uncharacteristic of urban areas. Information
-------
PM2 5 CSN QAPP
Section No. 18
Revision No.: 1.2.0
Date: 11/11
Page 2 of2
about local emissions sources will be needed, for example, to avoid locating a sampler too close
to a particle source such as a chimney, an industrial vent, a major highway, or a dusty unpaved
road. Information in state, local and/or Tribal agency data bases could be appropriate, but this
information needs to be spot-checked for accuracy by visiting the proposed site and surveying
the immediate (within a 300-m radius) area for hot spots of particulate emissions and other
sources or air flow impediments that are not recorded on older maps or in data bases. The
meteorological characteristics near a site should be checked by reviewing several years of data
from the nearest NOAA or NWS site in order to assess seasonal variations. In cases where the
site is quite far for the nearest government weather station, it may be necessary to set up and
operate meteorological sensors for wind speed and direction to characterize the
micrometeorology of the location.
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 1 of 17
19.0 Data Management
19.1 Overview
This section presents information on how field and analytical data for the Chemical Speciation
Network (CSN) will be managed. It does not address any data sets obtained outside the network
nor does it include management of data that may be given in summary and interpretive reports of
special studies. This section addresses these common data management topics: recording,
transformation, transmittal, reduction, validation, analysis, management, storage, and retrieval.
The CSN sampling data will ultimately be hosted by the Air Quality System (AQS). Flow audits
must be reported to AQS as well, and monthly flow verifications can voluntarily be submitted.
Detailed chain of custody (COC) and quality assurance/quality control (QA/QC) data collected
in association with the sampling data and network operations are hosted by a nonpublic data base
accessible to only state, local and/or Tribal (SLT) monitoring agencies. The CSN Contracted
Analytical Support Laboratory is the depot for incoming field data associated with ambient
sampling data as well as the QA/QC data collected by the monitoring agencies for validation of
the results. The CSN Contracted Analytical Support Laboratory's Quality Assurance Project Plan
(QAPP) for Chemical Speciation of PM2.5 Filter Samples provides additional information about
how the data is generated and managed.
Sample and QA/QC data for the CSN come from several sources at both the CSN Contracted
Support Laboratory and its subcontractors, and the organizations that perform the field
operations. The general flow of data between organizations and data systems is illustrated in
Figure 19-1. This figure emphasizes the organizational responsibilities, data base systems, and
major operations. The CSN Support Laboratory, shown at the top of Figure 19-1, is responsible
for integrating data from the various analytical laboratories (mass, ions, carbon, and elements)
with the shipping/receiving and chain-of-custody information for the sample module sets
provided to the field sampling personnel in (Monitoring Agencies) and SLT agencies. After
level 0 and level 1 validation, the data are posted monthly to a password-protected data base that
is accessible to the Delivery Order Project Officers (DOPOs) and their respective SLT
Monitoring Agencies for completion of data validation. The contract support lab notifies the
DOPOs who, in turn, notify each SLT Monitoring Agency of the availability of the data for
review. The SLT agencies conduct levels 2 and 3 validations and notify the contract laboratory
that data sets are acceptable or they identify invalid data, or other issues, which must be
addressed before the data is ready for posting on the AQS. A non-response indicates acceptance
by estoppel. The support laboratory uploads the data to the AQS data base with appropriate flags
as required.
The following list describes the data management activities in more detail. In many cases, double
entry methods are used and comparisons are made to reduce typographical errors and resolve
near-illegible or faint handwritten information. This list is presented in chronological order:
1. Presampling Laboratory Activities. This stage is conducted by the CSN support
laboratory and first includes inputting information to the laboratory data base
management system (DBMS) for weight values for the Teflon™ filters, filter lot
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 2 of 17
2.
acceptance test results, quality control (QC) information such as the background
concentrations of analytes on blank Teflon filters, washed nylon filters, and oven-
fired quartz fiber filters, as well as records of laboratory temperature and humidity
conditions during weighing sessions. Secondly, information about express shipping
and from the COC forms is entered so that shipments of filter-containing modules to
and from the various SLT agencies can be tracked.
Field Activities. These data are initially handwritten on the COC (CAFDS) form and
upon receipt with the filter samples, are input, using double entry, by the CSN
support laboratory to the laboratory DBMS. Items include data on operation of the
speciation sampler (sample identification numbers, volume of air sampled, ambient
STN Laboratory
Filter Lot
Acceptance
Records
SHAL Shipping
Records
Filter Pre
exposure
Weights
CSN Contract
Laboratory
DBMS
Chain of Custody
Data Records
Data Validation
Level 0 & 1
Monitoring Organization
Internal QA/QC Data
COCD Forms
Logbooks
Shipping Records
Audit Results
CA. Data, etc.
DOPO
Monthly Data
Reports in
hardcopy and
electronic format
Analytical Results
Distribution to the
MO'sto Final
Validation
Final Data
Validation
All levels
Sampler Verification
and Audit Reports
Fully validated
data to
Contract Lab
AQS Upload
Figure 19.1 The general flow of data between organizations
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 3 of 17
air temperature, sampling flags indicating nonstandard operation, and so on) as well
as operator's notes that can be used for validating the data. All data necessary for the
laboratory to calculate concentration values and to validate data will be entered on the
form, which is illustrated in the sampler operation SOPs contained in appendices to
this QAPP.
3. Postsampling Laboratory Activities. The CSN laboratory will enter the analytical
chemistry data for particle mass, elements, ions, and carbon species into the DBMS in
this stage. Data from QC operations such as results of multipoint calibrations and
linearity or span checks and analysis of duplicate samples, split samples, blanks, and
spiked samples are also entered.
4. Data Verification and Validation at the CSN Laboratory. Data verification and
validation will be carried out at both the CSN support laboratory and at the
Monitoring Agencies. The CSN support laboratory will base its validation on the
original COC forms, shipping records, and analytical QA/QC information. These
activities are explained in Section 22.0 of this QAPP.
5. Data Reporting to the DOPO. The CSN laboratory will post its validated data
monthly on a password-protected secure website which is accessible to the DOPO
and designated personnel in each SLT agency. Data from the laboratory and field are
combined at this stage of operations and are expressed in terms of concentration units
such as micrograms per cubic meter (|ig/m3) of air. The electronic data set will be in
the AQS format and is accompanied by a QC report that further describes the flagging
of the data.
6. Data Validation Activities at the SLT Monitoring Agencies. The Monitoring Agencies
will further validate the data sets at levels 2 and 3, based on its internal records,
analytical results of collocated sampling, and additional screening tests. These
activities are described in Section 22.0 of this QAPP.
19.2 Data Management Activities at the CSN Contracted Support
Laboratory
Data management activities in the CSN support laboratory are described in the QAPP for
laboratory operations and support. The current support laboratory QAPP is available on the EPA
AMTIC website.
19.3 Data Management Activities at the Monitoring Agency
This section describes in more detail the data management activities and responsibilities of the
SLT Monitoring Agencies. These data management activities should be conducted in accordance
with the recommended practices for data management described in Section 19.4.
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 4 of 17
19.3.1 Shipping and Receiving Records
The first stage of data management activities at the Monitoring Agency begins with receipt of a
sampling module set containing a partially completed COC and field data sheet form(s), which
could be combined into one. The Monitoring Agency records whenever a set of sampling
modules and URG cartridges is received or shipped. Recording the following information in a
notebook kept at the Monitoring Agency's shipping/receiving area is recommended:
Before Sampling:
• Date that each unexposed set of sampling modules is received.
• Site name or number and sampling date that the module set is to be used.
• Condition of the shipping container (if the container was damaged in shipment, the
CSN support laboratory should be notified through the DOPO).
• Warehousing area, if the set of sampling modules is not to be picked up immediately.
• Person to whom the set of sampling modules is released for installation at the
monitoring site (the site operator or someone who will take the modules to the site).
After Sampling:
• Person bringing the exposed set of sampling modules back to the shipping area.
• Site name or number and sampling date that the sampling module set was exposed.
• Condition of the modules and the external shipping container.
• Whether the ice substitute was pre-cooled as required.
• Overnight express air bill number and destination (the CSN support laboratory's
address should be pre-printed on air bills supplied in each module set).
• Date and time that the cooled container package was released to the overnight express
representative.
• One copy of the multi-part COC form should be retained by the Monitoring Agency
and checked to ensure that all information is complete and legible.
These shipping and receiving records may be necessary references in case there is a discrepancy
later in data attribution. Each Monitoring Agency should develop its own system for maintaining
shipping and receiving records.
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 5 of 17
19.3.2 Custody and Field Data Form
The COC and Field Data Sheet are shown in the sampler-specific SOPs in Appendix A and in
Section 12.0 of this QAPP. The field operator is responsible for filling the form(s) completely
and accurately. The handwritten notations on the form are the only source that the CSN
laboratory will have for critical information, including the following:
• Site/date that sampling actually occurred (usually preprinted on the custody and field
data form, but the operator may indicate changes).
• Total volume(s) sampled for each of the sampling channels.
• Average temperature(s), flow(s), barometric pressure and other information provided
by the PM2.5 sampler during the sampling period and the maximum and minimum
values.
• The CV of the sampler's flow rate for sampling period.
• Actual sampling time.
• Data validation flags issued by the sampler.
• Information about any field and/or trip blanks included with the set of sampling
modules.
• Operator's name and free-form notes.
The COC and Field Data Form are formatted so that there is a specific space for each item noted
above and is customized for the different speciation samplers in the CSN program. The operator
will repack the exposed modules and return the top copy of the COC in the cooler package. The
operator is responsible for keeping one copy of the COC form and returning it to the designated
person at the Monitoring Agency for filing and later use in data validation.
19.3.3 Data Management Activities During Data Validation by the Monitoring Agency
Every month the DOPOs will inform their respective SLT Monitoring Agencies that the next
"Batch" of data sets are available for review on the password-protected website. (The Contractor
has 45 calendar days to receive, analyze, and post data for states review on an interim password-
protected data base.) The website is divided into a set of subdirectories, one for each CSN
monitoring site. In addition to a detailed spreadsheet of the sampling data acquired for the period
covering that batch, each subdirectory will contain the site's data summary and QC report
prepared by the CSN support laboratory. The data summary will include an electronic file of
partially validated, AQS-formatted data that will require further validation by the Monitoring
Agency. The CSN laboratory will have performed the following level 0 and level 1 validation on
the delivered data:
• Data attribution and COC verification.
• Validation of laboratory analytical data.
• Screening for data input/output problems.
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 6 of 17
Refer to Section 22 of this QAPP or the CSN support laboratory QAPP and its accompanying
data management SOPs for a full description of the data validation procedures that are applied
and the checklist that is used.
The QC report delivered with the data file includes the following information (if any) that the
Monitoring Agency will need in order to complete the data validation:
• Data records with missing or uncertain attribution or COC information (principally
the site/date of exposure or defective/incomplete COC).
• Data records marked questionable or invalid due to sampler data flags.
• Data records marked questionable or invalid due to laboratory problems.
• Data records marked as possible outliers as a result of data screens applied by the
CSN support laboratory.
The Monitoring Agency will use its in-house records such as monthly performance verifications
and audits to examine data that have been flagged for attribution, sampler flags, and outliers. It is
possible that other data, such as operator's notes, shipping/receiving records, and QA/QC
records, can be used to validate the questioned data. There is usually little that can be done if a
measurement is assigned an invalidation flag due to a laboratory problem. Questions about flags
applied by the CSN support laboratory should be sent through the DOPO.
After reviewing the CSN support laboratory's QC information, the SLT Monitoring Agency must
perform further validation on the data set.
The specific procedures used for data validation are described in Section 22.0 of this QAPP.
Data management methods and means for ensuring correct inputs/outputs during data handling
are described in Section 19.4.
19.3.4 Reporting Data to AQS
After the data set has undergone levels 2 and 3 validation, each Monitoring Agency is
responsible for re-reporting appropriate revisions to the CSN support laboratory before the data
are loaded into AQS. The Monitoring Agency will have 45 days to review the data and respond
back to the support laboratory. The DOPO should be notified when the approval of the data takes
place or changes are reported to the CSN support laboratory. The support laboratory will have 15
days to upload data into AQS. This process is delineated in the lab service contract "reports of
work" provisions.
19.4 Recommended Data Management Practices
This section describes recommended data handling practices that are applicable to the CSN
support laboratory and to the Monitoring Agencies. It is expected that all organizations
participating in the CSN program will follow guidelines similar to those specified by EPA as
good automated laboratory practices (GALPs).
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 7 of 17
19.4.1 Manual Data Entry
Manually entered data will include entries from the COC forms and chemical analyst's notes and
calculated concentrations for certain analyses. Hand-recorded data must necessarily also be
manually entered. The following techniques apply to manual data entry:
* 100 Percent Data Verification on Input—Manually entered data will be reviewed,
preferably by a different operator, before it is committed to the data base. Analytical
data, however, may be entered and proofed by the analyst who is responsible for
generating the data. Large amounts of data entered in bulk, such as a group of COC
and field data forms entered by a data clerk, will be verified by duplicate entry in
which two clerks enter the same data. The two sets of inputs are then compared and
discrepancies are resolved.
• Range Checking—Many parameters lend themselves to checking for reasonable
range during or after data input. Properties of data sets, such as minimum and
maximum values, median and average values are easily calculated by spreadsheet
programs. This type of checking is frequently very useful for manual entry into screen
forms to catch misplaced decimal points, incorrect units, and omitted or extra digits.
19.4.2 Electronic Data Entry
Current procedures include manual recording of meta data and parametric sampling data
associated with each sampling event by the operator. The lab service contractor enters these data
into the data base which relates it by monitoring site ID and filter number. The data entry
personnel are trained to perform the levels 0 and 1 validation steps as the data are entered. It is a
convenient time and at a point where anomalies can be easily recognized. Queries can be made
prior to further filter handling and analyses. These data are then made available to the
Monitoring Agency for its use in performing levels 2 and 3 validations.
Migration to full electronic data transfer is possible. A full set of parametric sampling event
information is retained in the memory of the speciation sampler. It can be retrieved by use of a
download program and a laptop computer or portable data transfer device connected to the
sampler's RS232 port. (Sampler upgrades in the future will likely include use of flash drives or
other portable storage devices; USB ports have replaced serial ports on all current day
computers. They can usually be connected to serial ports with a transition cable and some
software.) The data may be stored on a portable data storage device which can be shipped with
the filter canisters and cartridges to the lab service contractor, but the ability to integrate the
more thorough level 0 and level 1 validation is lost. The lab service contractor will extract
parametric data associated with the sampling event and post it with the sampling results (see
below). A copy of the data will be retained by the site operator. No further checking of data
integrity will be done for electronic file transfers because these are internally validated by the
transmission medium.
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 8 of 17
Although migration to full electronic data transfer is possible, the ability to integrate the more
thorough level 0 and level 1 validation by a human is lost. Furthermore, any information entered
by the site operator (AQS code, filter numbers, etc.) still has to be added to the data base.
The laboratory services contractor posts data sets of analytical results and sampling field and
meta data to a restricted website data base which is password-accessible to the SLT Monitoring
Agencies. Each data set will be subjected to a level 2 and level 3 validation by its Monitoring
Agency before the data set is approved for uploading to AQS.
19.4.3 Sample and Data Tracking
Within the CSN support laboratory, a laboratory information management system (LEVIS) tracks
the cycle of activity for any filter for any sampling date and provide the history and present
status of each sample. Analytical samples will be internally tracked using a batch-oriented
internal COC form. Module sets sent from the CSN support laboratory to the Monitoring
Agencies are tracked in the laboratory's data base system. These processes are described further
in the CSN support laboratory's QAPP and the data management SOP.
Each Monitoring Agency must also develop methods for tracking, shipping, and receiving data
and for filing custody and field data forms, audit reports, and other CSN-related paperwork.
These records may be hardcopy or electronic, at the organization's discretion. Adoption of
procedures similar to those already in use for other programs such as the National PM2.5
Monitoring Network is encouraged. File, data records, and document retention is covered in
Section 9.0 of this QAPP.
19.4.4 Data Recording, Security, and Archiving
SLT Monitoring Agencies and the CSN contracted support laboratory should address data
security in their data management processes for CSN and other reportable environmental data.
The processes sand practices should be recorded in the organization's QAPP for participating in
the CSN. It will include, but not be limited to, practices such as the following (or equivalent):
• Organizations must perform regular backups of the CSN data base pertaining to their
monitoring site(s).
• As insurance against fires and other problems affecting an entire building, data
backup sets should be kept off-site.
• Weekly full backups of the data base, with nightly incremental backups of changed
information, are recommended (at a minimum).
• Use of passwords should be required and enforced. Passwords should be issued to
each individual user of the data system that contains the CSN data.
• Access to computers or networks containing unreleased CSN data should be
password-restricted, if possible, to authorized users.
• Access to unreleased CSN data should be restricted to personnel working on the CSN
program.
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 9 of 17
Archival of raw data and other program information is important because it allows the processing
and validation of questioned data items to be reconstructed in the event of challenge or for
research purposes. Table 19-1 summarizes the data and records that must be retained by the CSN
contracted support laboratory or Monitoring Agency for the life of the contract or for the term
required by contractual agreement, (often 5 years). The Location or Responsibility column of
Table 19-1 describes a typical location where such information is often stored; however, it may
be preferable to store some information in a more centralized location. Laboratory data records
should be stored in a secure location with limited personnel access and with protection against
fire and natural disasters or with suitable backup copies stored separately.
19.5 Data Validation
19.5.1 Validation Checks and Procedures
Data validation is a combination of checking that data processing operations have been carried
out properly and of monitoring the quality of field and laboratory operations. If a problem is
identified, the data can be corrected or invalidated, and corrective actions can be taken to prevent
its recurrence. The following considerations relative to data management practices during data
validation will apply:
• Flags denoting error conditions or QA status will be associated with each observation
down to the level of individual analyses, but flags must never overwrite the data
values so that recovery and review of the original data will be possible. However, if a
value is deemed invalid, based upon associated flags and other evaluations, a null
data code will be substituted for the original value during entry to AQS.
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 10 of 17
Table 19-1. Suggested Support Laboratory Data Record Archival Summary
Type of Record
Completed chain of custody /field data forms
Shipping records
Internal custody forms
Analytical and weighing raw data and instrument
traces
Analytical and weighing control charts
Weighing room environmental records
Certificates for all equipment and materials
standards (e.g., NIST or manufacturer's certificate)
Instrument calibration and QC records
Instrument maintenance and service records
Audit trails generated during data validation
QA records, including audit checklists and audit
reports
Copies or files of all partially validated data sets
sent from the CSN support laboratory to the DOPO
Copies or files of all fully validated data sets sent
from the Monitoring Agencies to the support
laboratory for transfer to AQS
Purchasing records related to the CSN program
Correspondence with the DOPO, including
consolidated sample requests
Correspondence and business records with
subcontractors involved with the CSN
Training records
Laboratory
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
MONI-
TORING
AGENCY
T
T
T
T
T
T
T
T
T
T
Data Archival Location or Responsibility
Lab: sample handling and archival
laboratory; MONITORING AGENCY: CSN
Manager's files
Shipping department files
Lab: sample handling and archival laboratory
Analytical laboratories and gravimetric
laboratory files
Laboratory files
Weighing laboratory files
Laboratory or QA Officer's files
Laboratory or QA Officer's files
Laboratory or field organization's files
Data processing department's and/or QA
Manager's files
QA files
CSN support laboratory Program Manager's
files
MONITORING AGENCY'S CSN Manager's
files
CSN support laboratory Program Manager's
files
MONITORING AGENCY: CSN Manager's
files
Support Lab: Program Manager's files
Support laboratory Program Manager's files
Program Manger's files
Levels 2 & 3 Data validation must include 100 percent manual review of both flagged and
unflagged data. A summary of various data to be validated and suggested methods are given in
Table 19-2. A qualified reviewer, such as a qualified analyst or the QA Officer, should also
examine randomly selected data transfer operations, according to Table 19-3 for reasonableness
before it is reported out of the organization. Completeness checks of the data set must be
included in the validation system. These checks are often automated screening programs,
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 11 of 17
particularly for within-record checks. The Monitoring Agency should verify the completeness of
the data set created by the CSN support laboratory by checking that all exposures have been
accounted for by comparison with its COC and field data forms and other records. An audit trail
is strongly recommended to document all changes made to the data set during validation
operations. Audit trails are described later in this section.
Table 19-2. Validation Check Summaries
Type of Data Check
Data parity and transmission protocol checks
Data review
Date and time consistency
Completeness of required fields
Range checking
Statistical outlier checking
Manual inspection of charts and reports
Sample batch data validation
Electronic
Transmission and
Storage
•/
Manual Checks
•/
•/
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 12 of 17
Description of Data
Transfer
• Field site sampler
• Laboratory
instruments
AQS data summaries
Finalized AQS data
summaries
Originator
Field site operator or
Field auditor
Laboratory auditor;
Laboratory analysts
Laboratory data clerk and
data supervisor
CSN laboratory
Recipient
Results that are outside of
advisory or acceptance
limits are reported to
personnel responsible for
remediation and
maintenance.
Reports are transferred to
support contractor who
posts data to AQS and in
QA data facility.
Values are reported to the
SLT data reviewers who
have the final decision on
accepting or invalidating
monitoring data
Laboratory data clerk
Results that are outside of
advisory or acceptance
limits are reported to
personnel responsible for
remediation and
maintenance.
QC Check values are
reported to the SLT data
reviewers who have the
final decision on
accepting or invalidating
monitoring data
AQS (EPA) (via the
DOPO)
AQS
QA Measures Applied
Critical Field data entries
are checked by field
operator. SLT QA person
reviews performance
verification and audit
results prior to reporting
to the support contractor
for loading into AQS.
Laboratory data checked
by laboratory supervisor
and spot-checked by QA
staff
Entries checked by
laboratory data clerk and
data supervisor; OAQPS
QA Officer
Checked by the
laboratory's Program
Manager or QA Officer
19.5.2 Data Flagging
As a result of data validation, individual items will be marked by a variety of validation flags that
generally indicate that the item is suspicious or invalid. Invalid data should not be reported to
AQS. The CSN support laboratory data base may contain a superset of the AQS flags for internal
use to facilitate QC reporting; however, these flags will be mapped onto the set of approved AQS
flags before they are released from the support laboratory. A summary of AQS data flags is
given in Table 19-4.
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 13 of 17
Table 19-4. Summary of AQS Data Flags for PM2.5
FLAG
1
2
3
4
5
6
AAR
AB
AC
AD
AE
AF
AG
AH
AI
AJ
AK
AL
AM
AN
AO
AP
AQ
AR
AS
RTI_DESCRIPTION
Critical Criteria Not Met
Operational criteria not met
Possible field contamination
Possible lab contamination
Outlier - cause unknown
Data prior to QAPP approval
Above Analytical Range
Technician Unavailable
Construction/Repairs In Area
Shelter Storm Damage
Shelter Temperature Outside Limits
Scheduled But Not Collected
Sample Time Out Of Limits
Sample Flow Rate Out Of Limits
Insufficient Data (Can't Calculate)
Filter Damage
Filter Leak
Voided By Operator
Miscellaneous Void
Machine Malfunction
Bad Weather
Vandalism
Collection Error
Lab Error
Poor Quality Assurance Results
AQS VALIDITY
CODE
1
2
3
4
5
6
AQS NULL V
ALUE CODE
AM
AB
AC
AD
AE
AF
AG
AH
AI
AJ
AK
AL
AM
AN
AO
AP
AQ
AR
AS
AQS_DESCRIPTION
CFR/Critical Criteria Deviation
Operational Deviation
Field Issue
Lab Issue
Outlier
QAPP Issue
Miscellaneous Void
Technician Unavailable
Construction/Repairs In Area
Shelter Storm Damage
Shelter Temperature Outside Limits
Scheduled But Not Collected
Sample Time Out Of Limits
Sample Flow Rate Out Of Limits
Insufficient Data (Can't Calculate)
Filter Damage
Filter Leak
Voided By Operator
Miscellaneous Void
Machine Malfunction
Bad Weather
Vandalism
Collection Error
Lab Error
Poor Quality Assurance Results
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 14 of 17
FLAG
AU
AV
AW
AZ
BA
BB
BE
BI
BJ
DCI
DMA
DSI
FBS
FMC
FNA
FSB
FSL
IA
IB
1C
ID
IE
IF
IG
ffl
II
IJ
RTI_DESCRIPTION
Monitoring Waived
Power Failure (Powr)
Wildlife Damage
Q C Audit (Audt)
Maintenance/Routine Repairs
Unable To Reach Site
Building/Site Repair
Lost Or Damaged In Transit
Operator Error
Channel Invalid
Module assembled in correctly
Shipment Invalid
Field or Trip Blank appears to be actual sa
Moisture contamination
Field operator designates no analysis
Sample is blank
Sample lost
African Dust
Asian Dust
Chem. Spills and Industrial Accidents
Cleanup After a Major Disaster
Demolition
Fire - Canadian
Fire - Mexico/Central America
Fireworks
High Pollen Count
High Winds
AQS VALIDITY
CODE
mple
3
IA
IB
1C
ID
IE
IF
IG
IH
II
IJ
AQS NULL V
ALUE CODE
AU
AV
AW
AZ
BA
BB
BE
BI
BJ
AQ
AR
BI
AQ
AL
AQ
BI
AQS_DESCRIPTION
Monitoring Waived
Power Failure (Powr)
Wildlife Damage
Q C Audit (Audt)
Maintenance/Routine Repairs
Unable To Reach Site
Building/Site Repair
Lost Or Damaged In Transit
Operator Error
Collection Error
Lab Error
Lost Or Damaged In Transit
Collection Error
Field Issue
Voided By Operator
Collection Error
Lost Or Damaged In Transit
African Dust
Asian Dust
Chem. Spills and Industrial Accidents
Cleanup After a Major Disaster
Demolition
Fire - Canadian
Fire - Mexico/Central America
Fireworks
High Pollen Count
High Winds
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 15 of 17
FLAG
IK
IL
DVI
IN
IO
IP
iQ
IR
IS
IT
IU
LFH
LFT
LHT
QAC
QCR
QL1
QMB
SA
T
V
W
X
Y
RTI_DESCRIPTION
Infrequent Large Gatherings
Other
Prescribed Fire
Seismic Activity
Stratospheric Ozone Intrusion
Structural Fire
Terrorist Act
Unique Traffic Disruption
Volcanic Eruptions
Wildfire-U. S.
Wildland Fire Use Fire-U. S.
Filter inspection flags* - Holes in filter
Filter inspection flags* - Tear
Lab holding times exceeded
Cation/ Anion total charge ratio out of limi
Between-analyte correlations
Outlier detected by QAO based on Level :
Total mass balance outside limits
Storm Approaching
Multiple Flags; Misc.
Validated Value
Flow Rate Average Out Of Spec.
Filter Temperature Difference Out Of Spe
Elapsed Sample Time Out Of Spec.
AQS VALIDITY
CODE
IK
IL
DVI
IN
IO
IP
iQ
IR
IS
IT
IU
2
ts 5
5
check 5
5
T
V
W
;. X
Y
AQS NULL V
ALUE CODE
AJ
AJ
SA
AQS_DESCRIPTION
Infrequent Large Gatherings
Other
Prescribed Fire
Seismic Activity
Stratospheric Ozone Intrusion
Structural Fire
Terrorist Act
Unique Traffic Disruption
Volcanic Eruptions
Wildfire-U. S.
Wildland Fire Use Fire-U. S.
Filter Damage
Filter Damage
Operational Deviation
Outlier
Outlier
Outlier
Outlier
Storm Approaching
Multiple Flags; Misc.
Validated Value
Flow Rate Average Out Of Spec.
Filter Temperature Difference Out Of Spec.
Elapsed Sample Time Out Of Spec.
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 16 of 17
19.5.3 Audit Trails
The audit trail is an important means for documenting changes to a data set made during
validation. The audit trail is important for establishing the reason for data changes, the authority
under which the change was made, and the data values before and after the change was applied.
Organizations are strongly urged to implement audit trails for the CSN program. Typical reasons
for making audit trail entries include the following:
• Corrections of data input due to human error.
• Application of revised calibration factors to sample results from an analytical run
queue.
• Addition of new or supplementary data.
• Flagging of data that are invalid or suspect based on manual examination or
automated validation of the data.
• Logging of the date and time when automated data validation programs are run.
Audit trail records usually include the following fields:
• Operator's identify (identification code).
• Date and time of the change.
• Table and field names for the changed datum.
• Complete identifying information for the item changed (date, time, and so on).
• Value of the item before and after the change (or image of the entire record before
and after the change).
19.6 Data Transformations
Calculations for transforming analytical data in units of mass per filter or mass per volume of
extraction solution to concentration units are relatively straightforward. Table 19-5 summarizes
transformations applied to analytical data to produce volume, mass, and concentration data.
Information about measurement uncertainties and method detection limits (MDLs) often
accompany summaries of such data.
Table 19-5. Raw Data Calculations
Parameter
Volume of air
sampled through
filter
Total mass on filter
(PM25)
Units
m3
ug
Conversion type
Calculated from average flow rate (Qavg) in L/min and
total elapsed sampling time, 10 minutes, multiplied by
the unit conversion (m3/L)
Calculated from filter postsampling weight (Mf) and
filter presampling weight (M;) in mg, multiplied by the
unit conversion (1000 ug/mg)
Equation
Va = QavgxtxlO-3
PM25 = (Mf - M,) x 103
-------
PM2 5 CSN QAPP
Section No. 19
Revision No.: 1.2.0
Date: 11/11
Page 17 of 17
PM25 concentration
(CPM25)
ug/m3
Calculated from laboratory data and sampled air
volume
CPM25 = PM25/Va
NOTE: Table 19-5 is applicable to the general categories ofanalytes to be produced by the
CSN program. Calculations within these categories are similar. Standard calculations such as
unit conversions and equations for calculating standard statistics are not provided. Calculations
involved in instrument calibrations are described in the respective SOPs and operating manuals
for the PM2.5 speciation sampler.
19.7 Data Reduction
Data reduction is the process of aggregating and summarizing results so they can be understood
and interpreted. For the activities covered, if data reduction is performed primarily to assess the
quality or uncertainty of the ambient concentration data.
• Average PM2.5 mass or species concentration for comparison against averages of the
same parameters over previous time periods to determine the reasonableness of
results.
• Bias and precision statistics based on accumulated FRM and speciated PM2.5 data and
sampler flow rate statistics.
• Data completeness reports based on the number of valid samples collected and
analyzed during a defined period of time versus the expected number of samples.
19.8 Data Analysis
For the activities covered by this QAPP, data analysis techniques and products will be limited to
those designed to assess the quality or uncertainty of the ambient concentration data. Tools to
ascertain the air quality trends or make air quality assessments are beyond the scope of the CSN
QAPP. Feedback from data analysts is necessary, however, to determine if the data quality
objectives have been met in allowing an interpretation of the data at the confidence level that was
originally projected. This will be covered in Section 24.
19.9 Data Storage and Retrieval
The CSN's contracted support laboratory's data storage and retrieval techniques with respect to
the recorded sampling and analytical data are described in the laboratory QAPP and SOPs.
Data storage and retrieval techniques for the Monitoring Agencies should be described in their
system documentation or through in-house SOPs developed for the CSN program.
Documentation of data storage and retrieval should include a summary of the type of data, the
media on which they are stored, security measures for safeguarding the data against destruction
and access by unauthorized persons, and the retention time for the data.
-------
PM2 5 CSN QAPP
Section No. 20
Revision No.: 1.2.0
Date: 11/11
Page 1 of 9
20.0 Assessment and Response Actions
20.1 Types of Assessments
The following types of assessments will be performed within the Chemical Speciation Network
(CSN):
• Management systems reviews (MSRs).
• Network reviews.
• Technical systems audits (TSAs).
• Performance evaluations (PEs).
• Audits of data quality (ADQs).
• Data quality assessments (DQAs).
MSRs are described in Guidance for Preparing, Conducting, and Reporting the Results of
Management Systems Reviews (U.S. Environmental Protection Agency Publication No. EPA
QA/G-3). TSAs, PEs, and ADQs are described in Guidance on Technical Audits and Related
Assessments for Environmental Data Operations (EPA Publication No. EPA QA/G-7). DQAs
are described in Guidance for Data Quality Assessment (EPA Publication No. EPA QA/G-9).
Information in these documents follows the specifications and guidance given in the American
Society for Quality Publication ANSI/ASQC E4-1994. The EPA's Office of Environmental
Information (OEI) periodically reviews and revises these documents. The current versions or
their equivalent documents are accessible through OEI's webpages at
http ://www. epa.gov/quality/index.html.
20.2 Assessment Frequency
Assessments will be performed at the frequency described in Table 20-1.
-------
PM2 5 CSN QAPP
Section No. 20
Revision No.: 1.2.0
Date: 11/11
Page 2 of 9
Table 20-1. Assessment Summary
Assessing Agency
Type of Assessment
Entity Assessed
Frequency
A. Laboratory and Network Operations
OAQPS/NAREL
OAQPS
(w/contractor
support)
Regional Offices
Regional Offices w
Monitoring
Agencies
TSA, MSRs and
PEs of laboratory operations
DQAs
MSRs
Network Reviews
Contracted support laboratories and
Laboratories of participating
Monitoring agencies
All data and QA Data
State and Local agencies,
State and local agencies
PE's Annually
TSA/MSRs Every
3 years
DQA's Annually
Once every 3
years
Once every 5
years
B. Field Operations
State, Local, or
Tribal Monitoring
Agenciesf
State, Local, or
Tribal Agencies
TSAs of field operations
Performance Audits of field
operations
Monitoring sites
Every Sampler
Annually*
Semi-annual*
fEPA Regional Offices during their 3-year MSR, or OAQPS on an ad hoc basis as needed, will conduct federal
level TSAs
* If a sampler fails an audit, or 2 monthly performance verifications, the audit frequency should be increased to
quarterly until the sampler performs satisfactorily in 2 consecutive audits.
20.3 Acceptance Criteria
The acceptance criteria for assessments were originally based on the data quality objectives
(DQOs) and measurement quality objectives (MQOs) established for the network (Section 7.0).
Historical performance has provided a basis to make a few adjustments. Further adjustments
could result from performance assessments and data results compiled over specified time
periods, e.g., annual, 3 years or 5 years. Table 11-2 in Section 11.0 and Table 14-2 summarize
the acceptance limits (and some advisory limits) for field operation of the speciation samplers. If
monitoring organizations recognize significant performance issues or changes in data
characteristics that in turn changes their disposition toward an MQO or DQO, they should
present the issue to their EPA Regional CSN contact and the Office of Air Quality Planning and
Standards (OAQPS).
Methods for and results of assessments of CSN field operations can be found on the Ambient
Monitoring Technical Information Center (AMTIC) website
(http://www.epa.gov/ttn/amtic/specguid.html). Documents at this site include "IMPROVE
Technical Systems and Field Audit Procedures," "MetOne SASS and URG Technical Systems
and Field Audit Procedures," as well as Technical Systems Audit Questionnaires.
-------
PM2 5 CSN QAPP
Section No. 20
Revision No.: 1.2.0
Date: 11/11
Page 3 of 9
The assessment of acceptability of CSN and IMPROVE laboratory operations is conducted by
the EPA personnel of the National Air and Radiation Environmental Laboratory (NAREL),
Montgomery, Alabama. Examples of laboratory assessments (both systems and performance
evaluations) are available at the EPA AMTIC website under the title "PM2.5 Speciation Lab
Audit Reports and Assessments" (http://www.epa.gov/ttn/amtic/pmspec.html).
20.4 Assessment Personnel
Assessors should have a minimum of 4 years of full-time appropriate and practical experience
(not including training) in air quality monitoring, including at least 2 years in quality assurance
activities.
Lead assessors should have technical experience with the samplers as well as assessment and
quality system experience. Other assessment team members also may have such experience, or
they may have only technical experience and currently be receiving assessment and QA training.
Lead assessors should have knowledge and understanding of the applicable environmental
statutes and regulations. They should be familiar with the EPA management systems and with
the organizational and operating procedures for environmental data collection. Lead assessors
should have a working knowledge of the technical assessment techniques for examining,
questioning, evaluating, and reporting environmental data operations and for following up on
response actions. They need to understand the assessment planning process. They also need
technical understanding of the PM2.s CSN. In general, they need to be able to evaluate the PM2.s
CSN's scope of work, its management system structure, and its operating procedures and to
judge the PM2.5 CSN's adequacy compared to this Quality Assurance Project Plan (QAPP).
Assessment team members should be familiar with technical assessment concepts and techniques
and with the structure and operating procedures for environmental data collection. They should
have technical knowledge of the PM2.5 CSN. Depending on the scope of the technical
assessment, assessors may need to meet additional qualifications, including health and safety
requirements.
Technical specialists, who have specialized knowledge of PM2.s CSN and basic knowledge of
assessment techniques and procedures, may participate in assessments. They may need basic
training in assessment techniques and procedures. Under the direct supervision of the lead
assessor, they may help prepare the technical portions of assessment checklists and may conduct
the technical portions of an assessment. They can verify findings and observations that are made
by other assessment team members concerning any specialized technical aspects of the PM2.5
CSN.
-------
PM2 5 CSN QAPP
Section No. 20
Revision No.: 1.2.0
Date: 11/11
Page 4 of 9
Three general standards for assessors are as follows:
• The assessors assigned to conduct a specific assessment should collectively possess
adequate professional proficiency for the tasks required. This standard places
responsibility on the assessors' organization to ensure that the assessment is
conducted by assessors who collectively have the technical knowledge and
assessment skills necessary for the assessment. This standard applies to the assessors
as a group, not necessarily to every individual assessor. Assessors should have
specific training regarding the procedures and reporting requirements for field and
laboratory assessments. The EPA OAQPS provides a comprehensive training course
for CSN and IMPROVE network auditors (assessors). Monitoring Agencies should
enroll their assessors in this training course at least once and conduct periodic
refreshers or evaluations of the assessors. IMPROVE Auditors will be required to
attend an initial certification course and participate in an annual reevaluation
conducted by the EPA.
• The assessors should be organizationally independent from the operations being
assessed, and able to maintain an independent attitude and appearance. This standard
places responsibility on the assessors' organization and on individual assessors to
maintain independence so that assessment findings will be both objective and viewed
as objective by knowledgeable third parties. See 40 CFR part 58, appendix A, section
2.2.
• The assessors should use due professional care in conducting the assessment and in
preparing related reports. This standard places responsibility on the assessors'
organization and on individual assessors to follow all applicable standards in
conducting assessments. Assessors should use sound professional judgment in
determining and interpreting the standards that are to be applied to the assessment.
The authority and independence of assessors, and the limits on their authority, must be clearly
defined in the organization's quality documents. Assessment personnel should have sufficient
authority, access to programs and managers, and organizational freedom to:
• Identify and document problems that affect quality;
• Identify and cite noteworthy practices that may be shared with others to improve the
quality of their operations and products;
• Propose recommendations (if requested) for resolving problems that affect quality;
• Independently confirm implementation and effectiveness of solutions; and
• When problems are identified, provide documented assurance (if requested) to line
management that further work performed will be monitored carefully until the
deficiencies are suitably resolved.
Prior to an assessment, it is important to establish whether the assessors have the authority to
stop or suspend work if they observe conditions that present a clear danger to personnel health or
-------
PM2 5 CSN QAPP
Section No. 20
Revision No.: 1.2.0
Date: 11/11
Page 5 of 9
safety or that adversely affect data quality. If not, assessors need to know what communication
they may be required to have with the authorized official who can stop work. Safety is
paramount; no assessments will be made in any unsafe conditions.
20.5 Assessment Reports
The product of an assessment is a written report. The objective of the report is to communicate
assessment findings to the proper levels of management in the EPA and the assessed
organization. The report must include:
• Assessment/review title and number and any other identifying information;
• The lead assessor, assessment team members, and the management and key personnel
of the assessed organization;
• Background information about the PM2.5 CSN activity being assessed, the purpose
and date(s) of the assessment, the particular parameter evaluated, and a brief
description of the assessment process;
• Summary and conclusions of the assessment and proposed response actions; and
• Attachments and appendices that include all evaluation and finding information.
Typically, two reports are produced: the draft findings report and the quality assurance final
report. The lead assessor is responsible for producing the draft findings report and should
organize the work to get the report written during the audit debriefing that should occur at the
end of the audit. In this case, the draft findings report might simply be a listing of the positive
findings as well as the findings that call for some corrective action. The EPA has developed an
Excel workbook template that documents a full TSA, a sampler audit (and a set of performance
verification checks). It is designed to expedite the reporting process and ultimately enable the
results and findings to be posted in a password-protected data base. These Excel Workbooks are
available to Monitoring Agency and EPA assessors. A full draft findings report for a should be
written within 15 working days following the assessment. Each Monitoring Agency will have
specific procedures for reporting results to the appropriate lines of management and remediation
contacts as appropriate. Examples of PM2.5 speciation network assessment questionnaires and
reports for field and laboratory reviews are available on the EPA AMTIC website at
http://www.epa.gov/ttn/amtic/specguid.html and http://www.epa.gov/ttn/amtic/pmspec.html,
respectively.
The assessed organization should be given the maximum opportunity to respond to the draft
findings report. This response should address the findings and discuss how any response actions
will be resolved. If the assessed organization disagrees with the findings, the response can
contain a rebuttal. Upon receipt of this response, the lead assessor should determine if the
response adequately addresses the findings or, if a follow-up assessment is required, when it is
appropriate to close out the assessment.
After the assessed organization's comments have been addressed, the final assessment report
should be prepared. The final assessment report should be similar in format to the draft findings
-------
PM2 5 CSN QAPP
Section No. 20
Revision No.: 1.2.0
Date: 11/11
Page 6 of 9
report and should be based on the draft findings report. Typically, the assessed organization's
response will be integrated into the summary of findings and response actions sections. The lead
assessor is responsible for correcting any findings that are demonstrated to be incorrect by
objective evidence to the contrary supplied by the assessed organization. Opinions of the
assessed organization that differ from those of the assessors are not valid reasons to alter the
report. The final assessment report should be uploaded to the Contract Support Lab's quality
assurance (QA) website. Flow audit results are extracted by the Contract Support Lab and posted
to Air Quality System (AQS). The EPA Regional CSN contact will be notified of the availability
of the assessment and if adverse findings have been unresolved. Documentation for TSAs and
network reviews will be archived at the respective EPA Regional Offices or Monitoring
Agencies for at least 10 years. The results of MSRs will be on file in assessor organizations' QA
filing system in accordance with their filing procedures. The MSR reports should be retained at
least 5 years.
20.7 Implementation of Response Actions
After an assessment, any necessary response actions should be timely and effective. In certain
cases, it may be necessary to perform response actions as quickly as possible. Such cases may
include adverse impacts on data quality or preventing data acquisition and/or risks to personnel
health and safety. Verbal approval for remediation from responsible parties suffices under these
conditions. Remediation is primarily the responsibility of the state, local, and/or Tribal (SLT)
monitoring organization. If issues exist that impede the successful remediation of the
nonconforming sampler or operational procedures, the EPA Regional CSN contact should be
notified and a resolution coordinated.
Response actions encompass immediate actions to eliminate problems such as errors in
calibrations, weighing, and other internal procedural problems and long-range response actions
instituted to improve overall data quality. Management of the assessed organization responsible
for the assessed activities is responsible for ensuring that effective and timely response actions
occur. The response actions should address the following:
• Measures to correct each nonconformance,
• Identification of all root causes for significant deficiencies,
• Determination of the existence of similar deficiencies,
• Response actions to preclude recurrence of like or similar deficiencies,
• Assignment of response action responsibility, and
• Completion dates for each response action.
Management of the assessed organization should implement the response actions and provide
objective evidence to the EPA of the effectiveness of the correction. Once such objective
evidence is received, the assessment will be closed unless a reassessment is planned. In some
cases, the assessment team may be needed to confirm the successful implementation of response
actions.
-------
PM2 5 CSN QAPP
Section No. 20
Revision No.: 1.2.0
Date: 11/11
Page 7 of 9
When an assessment or series of assessments leads to a finding or better procedure that, if
implemented, would simplify the work and/or improve the entire network's data, the finding will
be submitted to an ad hoc workgroup composed of SLT and EPA personnel and facilitated by the
OAQPS CSN program lead. This workgroup will develop specific procedures which will be
implemented by a Quality Bulletin indicating a forthcoming change to a standard operating
procedure and to a section of the QAPP. Refer to Section 11.2 of this QAPP for discussion and
an example of a Quality Bulletin.
-------
PM2 5 CSN QAPP
Section No. 20
Revision No.: 1.2.0
Date: 11/11
Page 8 of 9
Assessment Finding Response Form
Assessed Site or Laboratory :
Assessment Title: Assessment #: Finding #:_
Finding:
Cause of the problem:
Actions taken or planned for correction:
Responsibilities and timetable for the above actions:
Prepared by:
Signed by:
Speciation QA Manager
Reviewed by:
Date:.
. Date:
Date:
Response actions and remarks:
Is this assessment finding closed?.
Signed by:
. When (date).
File with official assessment records. Send copy to assessed organization.
Figure 20-1 Assessment Finding Response Action Form
-------
PM2 5 CSN QAPP
Section No. 20
Revision No.: 1.2.0
Date: 11/11
Page 9 of 9
20.8 References
American Society for Quality. 1994. Specifications and Guidelines for Quality Systems for
Environmental Data Collection and Environmental Technology Programs. ANSI/ASQC E4-
1994. Milwaukee, WI.
American Society for Quality. 2004. Quality Systems for Environmental Data and Technology
Programs. ANSI/ASQ E4-2004. Milwaukee, WI.
U.S. Environmental Protection Agency. 2003. Guidance on Assessing Quality Systems. EPA
Publication No. EPA QA/G-3, EPA/240/R-03/002. Washington, DC.
U.S. Environmental Protection Agency. 1998a et seq. Quality Assurance Handbook for Air
Pollution Measurement Systems, Volume II: Part 1, Ambient Air Quality Monitoring Program
Quality System Development. EPA Publication No. EPA-454/R-98-004. Washington, DC.
U.S. Environmental Protection Agency. 1998b. SLAMS/NAMS/PAMS Network Review
Guidance. EPA Publication No. EPA-454/R-98-003. Washington, DC.
U.S. Environmental Protection Agency. 2006. Data Quality Assessment: A Reviewer's Guide.
EPA Publication No. EPA QA/G-9R, EPA/200/B-06/002. Washington, DC.
U.S. Environmental Protection Agency. 2006. Data Quality Assessment: Statistical Tools for
Practitioners. EPA Publication No. EPA QA/G-9S, EPA 240/B-06/003. Washington, DC.
U.S. Environmental Protection Agency. 2002, reissued 2006. Guidance on Technical Audits and
Related Assessments for Environmental Data Operations. EPA Publication No. EPA QA/G-7,
EPA/600/R-99/080. Washington, DC.
U.S. Government Accountability Office. 2007. Government Auditing Standards. Washington,
DC.
-------
PM2 5 CSN QAPP
Section No. 21
Revision No.. 1.2.0
Date: 11/11
Page 1 of 5
21.0 Reports to Management
This section describes the quality-related reports and communications to management necessary
to support PM2.5 Chemical Speciation Network (CSN) operations and the associated data
acquisition, validation, assessment, and reporting. Effective communication among all personnel
is an integral part of a quality system. Planned reports provide a structure for apprising
management of the project schedule, the deviations from approved quality assurance (QA) and
test plans, the impact of these deviations on data quality, and the potential uncertainties in
decisions based on the data. These reporting documents should spell out the frequency of such
checks, the personnel responsible for generating the report, and recipient of the information.
Although the CSN has not been a regulatory network, the data produced by it might be used for
regulatory programs in the future. Furthermore, the October 17, 2006, rule revisions to the
monitoring regulations specifically name the CSN as monitoring sites that are subject to the
reporting requirements at 40 CFR part 58.16. Participating agencies should, therefore, follow the
quality control documentation guidelines applicable to other monitoring operations such as PMio
and PM2.5 federal reference method (FRM) network. This is particularly convenient when the
speciation samplers are located atNCore and compliance network sites (e.g., state and local air
monitoring station (SLAMS)). When speciation samplers are not located at such sites, and are
not part of the NCore network, Monitoring Agencies are urged to consult with their Regional
EPA QA management regarding the appropriate level of documentation.
Table 21-1 shows this information for the key reports to management documentation for the
CSN. These documents should be considered mandatory for samplers in the NCore network and
strongly recommended for other CSN sites. Columns in the table are as follows:
• Type of Report - name or description of report.
• Contents - brief summary of contents.
• Author - organization responsible for producing the report. Each organization listed as
Author or Recipient in Table 21-1 should develop procedures and schedules for
preparation of the referenced reports to management, and should designate specific
individuals with the responsibility.
• Recipient - organization responsible for review and approval of the report.
• Frequency - suggested or required frequency of reporting.
• Posting - location of where electronic version the report or data will be posted for public
access.
-------
PM2 5 CSN QAPP
Section No. 21
Revision No.. 1.2.0
Date: 11/11
Page 2 of 5
Table 21-1. Summary of Reports to Management for the CSN
Type of Report
21.1 Annual
Data Summary
Report
(Laboratory &
Field
operations)
21.2 Annual Q A
Report to
Management
(Network
Review)
21.3 Precision,
Bias and
Accuracy
Quarterly
Reports
21.3 Laboratory
Performance
Evaluation and
Technical
Systems Audit
Results
21.4SiteTSAs
(External)
2 1.5 Routine
Quality Control
records
21.6 Sampler
performance
Verifications
2 1.7 Data
Validation
Summaries
Contents
Executive summary.
Statistical summaries of
laboratory QC results;
corrective actions;
completeness summaries
by site
Precision, bias, and system
and performance audit
results.
Executive summary.
Precision, bias or network
data shifts, performance
audit results, Unusual
results of PE Tests and
results of special studies
Summary of precision,
accuracy and bias tests
(required for SLAMS/)
Evaluation and discussion
of PE audit results for
laboratories performing
PM2 5 chemical speciation
Evaluation of performance
at individual sampling
stations; design flow rate,
temperature, and
barometric pressure
accuracy
Operator's notebooks, site
records, calibration
records, etc.
Operator reports
Listing of data for all sites
operated by each
monitoring agency
Author
Laboratories
supporting the
CSN
OAQPS/
Monitoring
QA Staff
Monitoring
Agency
EPA
Monitoring
Agency
independent
assessors
Monitoring
Agency
Operations
Personnel
Monitoring
Site Operators
or
Technicians
Analytical
Laboratories
Recipient
OAQPS
OAQPS,
Regional EPA
Office CSN
Contacts
Regional EPA
Office
OAQPS
OAQPS
Monitoring
Agency QA
Management
Monitoring
Agency QA
Management
Monitoring
Agency
Frequency
Annual
Annual
Quarterly
Annual
Annual
Main-
tained
contin-
uously
Monthly or
upon
finding a
critical
failure
monthly
Posting
AMTIC
AMTIC
none
AMTIC
Flow rate
audits
posted on
AQS
none
AQS
posting is
Optional
but recom-
mended
none
-------
PM2 5 CSN QAPP
Section No. 21
Revision No.. 1.2.0
Date: 11/11
Page 3 of 5
21.8 Corrective
Action Reports
and other
performance
related records
Field or laboratory
activities.
Identification of problems,
proposed solution, and
results; results of invalid
tests
Monitoring
Agency
Operations
Personnel or
CSN support
laboratory
Monitoring
Agency QA
Management
at monitoring
agencies and
OAQPS
as needed
None,
unless
corrective
action is
necessary
on a
network
wide basis
21.1 Annual Quality Control (QC) Summary Report (Laboratory)
The EPA's contracted support laboratory for the CSN program produces an annual report that
details analytical quality measures as well as trip and field blank results, completeness
summaries, and corrective actions. It also includes a summary of performance verifications,
sampler audits and any noteworthy technical system audit (ISA) findings that appear to be
systemic network issues. These reports are posted on the Ambient Monitoring Technical
Information Center (AMTIC) website. Other laboratories performing analyses for the CSN are
urged to produce similar documentation on an annual basis and to provide this information to the
monitoring agencies.
21.2 Annual QA Report to Management (Network Review)
The Office of Air Quality Planning and Standards (OAQPS) Ambient Monitoring QA staff will
compile an assessment each year summarizing the following:
• Results from the collocated precision sites.
• Results of TSAs and sampler performance audits.
• Surprising or unusual shifts in ambient concentration results or blanks.
• Surprising or unusual results from the performance evaluation tests of the participating
network laboratories.
• Results of special studies pointed at specific operating procedures or performance
characteristics; e.g., shipping procedures, field blank collection, etc.
• Collocated Sampler Comparison Results.
21.3 Precision, Bias and Accuracy Reports
For the CSN, each monitoring organization must report to AIRS-AQS the results of all precision,
bias and accuracy tests as described in Section 14 on a quarterly basis as set forth in 40 CFR
Parts 58.16, and 40 CFR Part 58 Appendix A, Section 5. Precision from collocated samplers will
be calculated quarterly, (although the raw data from the 6 collocated samplers across the CSN is
submitted monthly). Since there are no independent measurements of ambient concentrations
with which we could compare CSN results, bias and accuracy can only be is represented by
surrogates of monthly flow checks and semiannual audits (or more frequently if chosen).
Reporting of monthly flow checks are not mandated by the regulations, however, they can be
-------
PM2 5 CSN QAPP
Section No. 21
Revision No.. 1.2.0
Date: 11/11
Page 4 of 5
invaluable if an audit reveals that a sampler is, and has been, operating at a flow rate that
invalidates the results.
21.4 Laboratory Performance Evaluation and Technical Systems Audit
Results
The EPA NAREL and OAQPS management perform periodic evaluations of the laboratories
performing chemical speciation analyses for the CSN. Performance audits are performed on an
annual basis. The schedule for completing these tests has varied from year to year due to the
multiple labs that participate. Refer to the Quality Assurance Project Plan and standing operating
procedures for EPA, ORIA, NAREL, Montgomery, Alabama.
21.5 Site TSAs (External)
For the SLAMS program, external systems audits are conducted for each monitoring
organization at least every 3 years by the EPA Regional Office as required by 40 CFR Part 58,
Appendix A, Section 2.5. At least one CSN site should be audited during the review. The choice
of sites in the PM2.s chemical speciation program should be discussed with the responsible EPA
Regional QA Coordinator, but first priority should be given to the speciation samplers at NCore
sites. Appendix A-6 is a PDF version of an Excel spreadsheet that has been created to guide
auditors through complete TSAs of the CSN monitoring sites and samplers, and electronically
report the audit results to the lab support contractor so that the flow audit data (on whatever
frequency the audits occur) can be posted to AQS. The national QA lead for the CSN or the
current support lab contractor will have information as to how this spreadsheet can be acquired
and the resulting report electronically transferred to the contract support lab. The worksheets and
directions for their acquisition and use are also posted at
http://www.epa.gov/ttn/amtic/specguid.html.
21.6 Routine Quality Control Records
Routine quality control records such as operators' notebooks, control charts, calibration records,
etc., are not considered reports to management; however, these materials must be maintained and
made available for audits and reviews. Sampler performance verification results, however, are
indicators of potential loss of data and should, therefore, be reported on a monthly basis with
immediate notification of the supervisor who can deploy technicians to correct the problem.
Appendix A-5 is a PDF version of an Excel spreadsheet that has been created to guide operators
through QC checks of the sampler and electronically report them to the lab support contractor so
that the monthly flow check data can be posted to AQS. The national QA lead for the CSN or the
current support lab contractor will have information as to how this spreadsheet can be acquired
and the resulting monthly worksheets reported. The worksheets and directions for their
acquisition and use are also posted at http://www.epa.gov/ttn/amtic/specguid.html.
-------
PM2 5 CSN QAPP
Section No. 21
Revision No.. 1.2.0
Date: 11/11
Page 5 of 5
21.7 Data Validation Summaries
According to 40 CFR part 58.16 subparagraphs (a)-(c), all QA data collected at CSN sites in
accordance with Appendix A will be reported in AQS. It will be consistent with the audit results
submitted for the PM2.s FRM/FEM network.
The EPA contract laboratory for the CSN produces, as part of the levels 0 and 1 validation,
creates a set of review data on a monthly basis for review by the individual agencies. The
agencies have 45 days to review their monitoring data, the contract lab's validation results. They
may conduct their own levels 2 and 3 validation and submit corrections or questions. Sixty days
after initial posting of the data (45 days plus 15), the data with any agency-required changes are
posted on the AQS data base system for public access. The data validation summaries produced
by the contract laboratory and the requested corrections and other correspondence are not
published, but are maintained so that they are available for audits and reviews. Validated air
quality data submitted for each reporting period will be entered into the AQS using the
procedures described in the AQS Data Coding Manual.
21.8 Corrective Action Reports and Performance-Related Records
A corrective action reporting system should be in place for all monitoring agencies. The
Corrective Action Report procedure should be followed whenever a problem is found ,such as a
safety issue that presents a risk to the site operators and auditors or an operational or procedural
problem that will result in loss of data or generated invalid data. The Response/Corrective Action
Report is one of the most important ongoing reports to operational management because it
documents primary QA activities, provides valuable records of QA activities that can be used in
preparing other summary reports, and may benefit other sites that encounter the same problem.
The Corrective Action Report form should identify the originator, state the problem, and may
suggest a solution. The form also indicates the name of the persons or persons who is/are
assigned to correct the problem, and the person responsible for verifying that the corrective
action has been completed. Copies of the Response/Corrective Action Report should be
distributed twice: first, promptly after the problem has been identified and corrective action has
been scheduled, and second, when the correction has been completed.
Corrective Action Reports (or equivalent) are normally not published but should be retained by
the agencies for audits and reviews. Significant corrective actions should be included in the
Annual Network Review discussed above, particularly when reportable data are affected.
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 1 of 17
22.0 Data Review, Validation, and Verification Requirements
This section describes the verification and validation process, which is used to decide the degree
to which each data item has met applicable quality specifications. The specific requirements for
verification and validation are developed by estimating the potential effect that each error
component may have on the usability of the associated data item, its contribution to the quality
of the reduced and analyzed data, and its effect on attainment of the data quality objectives
(DQOs).
Verification and validation are not the same as data quality assessment or evaluation of the
DQOs, processes which are described elsewhere in this Quality Assurance Project Plan (QAPP).
Only after the data set has been verified and validated can it be fully assessed and/or used to
address the specific scientific and regulatory questions embodied in the DQOs.
Data validation summary reports should be included in regular quality control reports to
management. These reports should include the following information, at a minimum:
• Time interval covered by the report.
• Site identification(s).
• Sampler identification(s).
• Total number of valid observations sent to the Air Quality System (AQS).
• Number of flagged observations sent to AQS, categorized by analyte and flag.
• Number of invalid observations (not sent to AQS), by analyte.
• Statement of significant corrective actions taken.
22.1 Data Verification and Validation Responsibilities
Verification of data for the PM2.5 chemical Chemical Speciation Network (CSN) is the joint
responsibility of the Monitoring Organization, which runs the field component of the program,
and the CSN support laboratory, which analyzes the samples and calculates and reports the data.
Table 22-1 describes the respective responsibilities of the field and laboratory components for
data verification and validation.
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 2 of 17
Table 22-1 Data Verification Activities and Responsibilities for the CSN
Verification Activity
State, Local or Tribal Monitoring Organization
CSN Laboratory
Data source attribution. Verify that the
site, date, time, and channel assignments
are correct. Logbooks, reporting forms,
data custody sheets, and electronic data
transmittals should be checked for
consistency.
The Monitoring Organization should compare the
information on the chain of custody (COC) and field
data form with labels on modules received. Report
any discrepancies to the CSN support laboratory.
The COC forms should be corrected to reflect the
actual module used for a particular sampling
channel, if other than originally assigned by the
laboratory.
Discrepancies in module assignment due to
procedural errors in the field operation should be
corrected and documented.
The CSN support laboratory is responsible for generating the
COC sheets and placing corresponding labels on sample
modules sent to the field. The laboratory should respond
immediately to notifications from the field of shipped
modules that do not agree with the COC sheet.
The support laboratory database management system
(DBMS) will accommodate changes of module assignment
reported by the field.
Discrepancies in module assignment due to procedural errors
in the laboratory sample handling operation should be
corrected and documented.
Results for samples that cannot be positively identified must
be flagged as invalid in AQS.
Site selection and monitor placement.
The Monitoring Organization is responsible for
ensuring that all siting criteria have been met when
originally siting a CSN sampler. Continued
compliance with siting criteria should be verified at
least annually by the Monitoring Organization.
If the support laboratory is notified that a particular CSN
sampler is in violation of siting criteria, the corresponding
AQS data may have to be flagged or invalidated. This
decision should be made jointly by the Monitoring
Organization and EPA Regional Office with oversight
responsibility for that MO .
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 3 of 17
Verification Activity
State, Local or Tribal Monitoring Organization
CSN Laboratory
Integrity of sample handling. This
involves verification that the SOPs for
sample handling have been followed so
that the physical integrity of the sample
and its correct identification are ensured.
This requires that both the field and
laboratory operations follow the approved
SOPs and that any discrepancies are
followed up and resolved. System Audits
should include a step-by-step review of
sample handling procedures. Data audits
should include examination of original
custody sheets and other records to look
for discrepancies in sample identification.
The Monitoring Organization should develop and
follow SOPs for sample handling and/or follow
those provided by the contracted support laboratory.
It is very important that the PM2 5 sampler flow rate
be within ± 5 % of the design flow rate.
The field operation should conduct internal audits
and submit to external audits and reviews of its
sample handling and data processing systems.
The field operation should cooperate with
investigations of data integrity initiated by the CSN
support laboratory, U.S. Environmental Protection
Agency (EPA), or others.
The CAR process should be followed when
investigating isolated or systematic discrepancies.
Systematic problems should be addressed by
revision of the appropriate SOP.
The CSN laboratory should develop and follow SOPs for
sampling module shipment and sample handling.
The laboratory should conduct internal audits and submit to
external audits and reviews of its sample handling and data
processing systems.
The CSN support laboratory should cooperate with
investigations of data integrity initiated by a Monitoring
Organization, EPA, or others.
The CAR process should be followed when investigating
isolated or systematic discrepancies. Systematic problems
should be addressed by revision of the appropriate SOP.
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 4 of 17
Verification Activity
State, Local or Tribal Monitoring Organization
CSN Laboratory
Checks on sample containers and
preservation methods. Specific checks on
sample containers include examination
for physical integrity and comparison
with the COC and data sheet for correct
identification. The chief preservation
method for exposed sample modules is
the use of a chilled ice substitute in
insulated shipping containers to reduce
the temperature to 4°C.
The field operation is responsible for checking the
integrity of shipping containers and individual
modules upon receipt or deployment at the site. The
CSN support laboratory should be notified
immediately of any damage that may have occurred
in shipment.
The field operator is responsible for chilling the ice
substitute prior to shipping the exposed sample
modules back to the CSN support laboratory.
The CAR process should be followed if a systematic
problem is suspected.
The CSN support laboratory is responsible for properly
packaging the modules in shipping containers that include
the required number of ice substitutes.
The CSN support laboratory is responsible for notifying the
carrier and initiating claims for shipping damage.
The CSN support laboratory is responsible for checking the
condition of the shipping container, its interior temperature,
and individual modules upon receipt from the field.
The CSN laboratory must note any discrepancies that might
affect sample validity in the DBMS. The Monitoring
Organization should be notified if the problem might be due
to packaging procedures.
The CAR process should be followed if a systematic
problem is suspected.
Procedures to ensure that data were
generated as specified in the sampling or
analysis SOP.
This is typically ensured using systems audits, EPA
Regional reviews, and results of performance audits.
Failure on any of these checks could imply that some
samples acquired prior to the check are suspect or
invalid. It is very important that the PM2 5 sampler
flow rate be within ± 5 % of the design flow rate.
The CSN support laboratory and the EPA CSN
Regional Coordinator should be notified if serious
procedural issues are raised.
The CAR process should be followed to document
and rectify systematic procedural problems.
Integrity of laboratory analysis procedures is typically
ensured using internal and external systems audits and
results of performance audits. Failure on any of these could
imply that some samples analyzed prior to the check are
suspect or invalid.
Monitoring Organization(s) with affected samples and the
EPA Project Officer (PO) for the CSN support laboratory
contract should be notified if serious procedural issues are
raised.
The CAR process should be followed to document and
rectify systematic procedural problems.
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 5 of 17
Verification Activity
State, Local or Tribal Monitoring Organization
CSN Laboratory
Activities to determine how seriously a
sample deviated beyond the acceptable
limit so that the potential effect on the
validity of data can be evaluated.
Systematic problems that lead to unacceptably large
biases should be investigated and documented using
the CAR process.
A person within the organization should be
designated to investigate serious discrepancies
affecting sample data. This designation should be
formalized in the organization's quality management
plan (QMP) or in an appropriate SOP.
The CSN support laboratory should be notified when
a data discrepancy is first suspected and again when
the discrepancy is quantified and the investigation is
closed.
Although it would be impossible to prepare SOPs for
assessing every contingency, the organizational
responsibility and general approach for investigating
discrepancies should be clearly documented.
Systematic problems that lead to unacceptably large biases
should be investigated and documented using the CAR
process. The support laboratory supervisor is usually in a
position to investigate such occurrences, but this
responsibility can also be delegated to a qualified analyst or
to the project QA staff with the concurrence of the
laboratory supervisor and the Services Program Manager.
Data should be corrected, flagged, or invalidated based on
the best assessment of the individual situation.
Corrections to previously sent data should be transmitted to
AQS as soon as feasible. The EPA and the Monitoring
Organization should also be notified when its data in AQS
are changed.
Data corrections and flagging should be noted in regular QC
reports to management. [Detailed summaries are prepared
monthly and are reviewed by the CSN QA Officer. Annual
QC reports are also prepared, containing summaries of data
corrections and flagging, which are transmitted to EPA and
posted on the public AMTIC website.]
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 6 of 17
22.2 Corrective Action Reporting Process
Each Monitoring Organization and the CSN support laboratory should have a Corrective Action
Reporting (CAR) process in place. Ideally, this process should closely resemble corrective action
procedures used with other monitoring activities conducted by the organization. This process
should consist of the following elements:
• A description of the organizational responsibilities and procedures for instituting
corrective actions for the CSN program. This would be appropriately placed in the
organization's CSN standard operating procedures (SOPs) or in the QMP. If an
adequate process already exists in the organization, it is not necessary to develop new
procedures for the CSN.
• A reporting form (i.e., the CAR form) describing the event or problem along with a
suspected cause; a recommended solution is optional. Other information should
include the date and the submitter's name.
• A means for assigning responsibility for the corrective action investigation, as well as
for scheduling and appropriating resources to it.
• All active and just-completed CAR investigations should be reported in the next
scheduled quality control (QC) report to management or equivalent document. Dates
and sites of reportable data affected by the problem should be provided in detail.
Delayed or canceled CAR investigations should be identified and the reason for
delaying or canceling the investigation justified.
• The CAR files should be subject to regular audits and reviews.
22.3 Use of QC Information for Verification and Validation
The various QC samples used with the CSN are also potential sources for verification/validation
information. Table 22-2 summarizes QC information and its uses in data verification and
validation.
22.4 Use of Calibration Information for Verification and Validation
Calibrations can generate information that is useful in the verification and validation process.
Because this information is often not directly associated with a particular sample, procedures
must be in place to identify data that were dependent on a particular instrument when the
calibration data indicate that the instrument's performance was suspect. These considerations
apply to most analytical chemistry instruments; the balances used to weigh filters; and the field
calibrations of flow, temperature, and barometric pressure. It is very important that the PM2.5
sampler flow rate be within ± 5 percent of the design flow rate. Each SOP that involves the use
of quantitative calibrations must include the following considerations related to eventual use of
that data for data verification and validation:
• A mechanism for reporting calibrations that are out of specifications.
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 7 of 17
• A means for identifying all data that might be affected by the problem (all sample
data back to the last acceptable calibration should initially be considered suspect;
investigation may be required to estimate the maximum probable error and to decide
on the validation status of sample data).
• A set of acceptance limits on the calibration that address zero/offset, bias/gain/slope,
and linearity/noise.
• A procedure for verifying out-of-limits calibrations.
Table 22-2. Quality Control Data for CSN Data Verification and Validation
Type of QC
Data
Responsibility
Usage for Verification and Validation
Field Blanks
Joint lab and
field
High field blank values may indicate high levels of contamination
introduced by the sampler or the filter loading process. Data currently
reported from the CSN program are not numerically corrected for field
blank levels. The CSN support laboratory includes all field blank data in the
monthly data reports that are sent to the Monitoring Organizations for
review. Monitoring Organizations should raise questions with the CSN
support laboratory if unexpectedly high blanks are found. Data may be
flagged or invalidated based on the results of investigation.
Trip blank
Joint lab and
field (primarily
lab)
High trip blanks (with or without a corresponding high field blank) may
indicate a laboratory problem, or a problem with filter media, or with
shipping and handling. Corrective actions should begin in the laboratory.
Data currently reported on the CSN program are not numerically corrected
for field blank. The CSN support laboratory should investigate systematic
high trip blanks. Data may be flagged or invalidated based on results of
investigation.
Sampler leak
check failed
Field
Leak checks of the MetOne and the URG 3000N samplers have been
designed to indicate a pass or fail situation although either sampler can run
if a failing leak is occurring. If a failing leak is detected during a leak check
or audit, the person conducting the assessment should consult with the
manufacturer's operators manual for a troubleshooting guide or contact the
technical service department. The period of time and number of sampling
events that may have experienced the failing leak should be reported to the
CSN support laboratory immediately so that data can be flagged or
invalidated appropriately.
Sampler flow
rate check/audit
results outside of
acceptance
criteria
Field
The ability of a filter-based particle sampler to achieve the correct cut point
is critically dependent on the sampling flow rate. It is very important that
the PM2 5 sampler flow rate be within ± 10 % of its design flow rate. Flow
Rate CV's should be examined periodically and compared against the
advisory limits established for the particular sampler. High CVs are an
indicator that the flow rate may not have been within the ± 10 % limits.
Severe departure from the normal CV would be grounds for invalidating
results. Note advisory limits have been established at 5%.
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 8 of 17
Type of QC
Data
Responsibility
Usage for Verification and Validation
Sampler
temperature,
barometric
pressure sensor,
or flow rate
sensor failed
Field
Sensor failures should be investigated as soon as they are detected. Sensor
failures may directly affect the flow through one or more sampling modules
and therefore must be documented. The operator discovering the problem
should describe the error quantitatively, if possible. The operator should
also try to determine when the sensor began to malfunction. Data back to
the last successful check may be suspect. If it appears that the failure could
cause an error of 10% or more in the sampler's design flow rate, the
problem should be brought to the attention of the CSN laboratory so that
data can be flagged or invalidated appropriately and steps can be taken to
promptly correct the problem.
Filter integrity
inspections
Lab and field
Filters that fail the initial visual integrity inspection at the CSN laboratory
should not be used. If the filter has not been used for sampling, its number
can be voided in the laboratory's DBMS and no further action is necessary.
Failure of visual inspection after sampling usually results in data
invalidation or flagging. The person making the inspection should document
it on the COC and data form or in the analytical results data. Data entry
personnel should make the appropriate entry when the data are input into the
DBMS. The original analyst's notes or COC form may have to be consulted
during data validation to determine the seriousness of the problem and the
validity status of the reported data.
Repeated or systematic filter failures may be caused by many different
factors including rough handling, manufacturing defects, environmental
factors during sampling (rain, insects, etc.), defective packing materials, or
contamination in the sampler before or after sampling. Systematic problems
should be investigated using the CAR process and documented in QC
reports to management.
Laboratory QC
samples
Laboratory
A large number of routine QC samples are run in the gravimetric and
chemical laboratories. In many cases, instrumentation problems detected by
QC sample results can be corrected before any sample data are affected. The
specific acceptance criteria for laboratory QC samples are generally
determined by statistical methods (e.g., control charts or equivalent) or by
comparison against set limits defined in the method and given in the SOP.
These acceptance limits should be posted in the laboratory; each analyst is
responsible for identifying out-of-specification QC results.
Reserve aliquots or specimens of the original media should be retained until
successful post-analysis QC results have been obtained. (Aliquots and
specimens are also archived for a period of time under the CSN program, so
reanalysis may be possible at a later time; however, the results are generally
less accurate and more costly to obtain than immediate reanalysis.)
If QC results indicate an unacceptable uncertainty regarding the true value
and the analysis cannot be reproduced, the data must be invalidated or
flagged.
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 9 of 17
• Procedures for correcting data taken with defective calibrations. Errors other than
simple drift in the instrument's response factor, such as excess noise and calibration
nonlinearity, should generally not be corrected by recalculation; instead, the
instrument should be repaired and the samples rerun if possible. Recalculation of
results due to shifts in the instrument's response factor should be done only with the
approval of the laboratory supervisor, and these changes should be documented
appropriately by the laboratory.
Other considerations regarding calibrations that facilitate their use in the data verification/
validation process include the following:
• Calibrations should be performed within an acceptable time before and after analysis
of field samples as specified in the SOP. The acceptable time window is typically a
function of the instrument's drift characteristics.
• Calibrations, other QC samples, and field samples should be done in proper sequence;
for example; the positions of vials in an auto-sampler should be double-checked and
all vials should be labeled if possible.
• Calibration points must bracket the concentration range of interest and should be
spaced according to the needs of the method as defined in the SOP. Note that only
one-point flow rate calibrations are possible with the MetOne sampler. The URG
3000N provides for a three-point calibration. If either will not retain the calibration
the manufacturer should be contacted for guidance on troubleshooting and repairs.
The Contract Support Lab should be informed of sampling events that were affected
by incorrect flow rates due to poor calibration.
• Enough points should be included in a multipoint calibration to assess noise and
linearity (it may not be necessary to assess noise and linearity on a daily basis,
depending on the type of instrument).
22.5 Level 0 Verification and Validation
Basic review of data with respect to their origin, documentation, and critical criteria is referred to
as "level 0" verification and is performed by the CSN support laboratory. The criteria to be
applied during level 0 verification of the data set are summarized below:
• Data source attribution—Verify that the site, date, time, and channel assignments are
correct. Logbooks, reporting forms, data and chain of custody sheets, and electronic
data transmittals should be consulted if a problem of attribution is suspected.
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 10 of 17
COC verification—All COC forms for filters and denuders will be checked for
completeness at the CSN support laboratory before they are entered into the data base
and archived. Missing information should be identified by the data entry operator and
brought to the attention of the Sample Handling and Archival Laboratory (SHAL)
supervisor. The laboratory QC supervisor or designee should determine the validity of
any samples for which mandatory COC and field data information is missing. It may
be necessary to contact the state, local, and/or Tribal (SLT) Monitoring Organization
to attempt to fill in this information.
Holding times and conditions—The shipping and receiving documentation for all
PM2.s chemical speciation samples (filters and denuders) should be checked to verify
that holding times have been met and that required storage conditions such as
temperature met the requirements. Data should be flagged in AQS if holding times
and shipping/storage conditions were violated. The CSN support laboratory quality
assurance (QA) manager, in consultation with the EPA and the laboratory
supervisor(s), will decide the validity of any samples for which these conditions have
been violated.
Data transmission and recording integrity—Each error in data integrity that is
identified must be investigated and corrected and/or appropriate actions taken. The
CSN support laboratory QA manager should determine whether an uncorrectable data
transmission error affects data validity. The CAR process should be used with
problems of a systematic nature.
Calibration status of sampler and sensors—The Monitoring Organization must have a
means for verifying that PM2.5 speciation samplers were calibrated or checked within
the required windows of time. Samples taken when the instrument was past due on
any calibration or recertification interval must be appropriately flagged in AQS. It is
the responsibility of the Monitoring Organization to identify such data to the CSN
support laboratory because sampling records received by the laboratory do not
include all the necessary information.
Audit status—The Monitoring Organization must verify that all monthly, quarterly,
and other scheduled audits defined in the applicable SOP or QAPP have been
performed on time. Samples taken when a systems audit or transfer standard
recertification was past due must be appropriately flagged in AQS. It is the
responsibility of the Monitoring Organization to identify such data to the CSN
laboratory because audit records are typically not available to the CSN support
laboratory.
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 11 of 17
• Operational flags—All operational flags generated by the sampler electronically or
recorded on the COC data sheet by the operator will be entered into the data base. A
translation between sampler codes and AQS validation flags has been developed.
Section 19.5 of this QAPP presents the AQS data flags used by the CSN support
laboratory to qualify data. Operational flags can result in data flagging or
invalidation, depending on the severity.
22.6 Level 1 Data Validation
Validation (level 1) is the process of evaluating the correctness and acceptability of individual
items or groups of items within the data set using statistical methods and other screening
techniques. This process involves evaluating the impact of verification problems, QA or QC
problems, and statistically detected anomalies on the usability of the data for their intended
purpose.
Level 1 validation of field data will first involve the processing of verification results and data
screens into AQS data flags and then providing an overall assessment of the validity of the data
item or items. Based on the number and types of data flags and other information generated
during the verification and validation process, some data may be designated as invalid. Invalid
concentration data are not reported to AQS, although flagged records are uploaded in place of the
missing observation data to indicate that the data have been invalidated. The following items are
involved with validation of the PM2.5 chemical speciation field data:
• Operational data screening—temperature, barometric pressure, flow, and other
operational data are screened for compliance with acceptance limits established for
the CSN program.
• Filter inspections and other manual verification procedures performed by the site
operator and CSN support laboratory personnel.
• Validation flags attached and reported—A limited number of flags are provided in
AQS to document exceptional conditions that apply to specific data items. Only valid
data are reported to AQS.
• Invalid data are identified—Based on the number and types of data flags and other
information generated during the verification and validation process, some data may
be designated as invalid. Invalid data are not reported to AQS.
22.7 Data Screening Techniques Used by the Contracted Support
Laboratory
The following procedures will be performed by the Contracted Support Laboratory before the
data are made available to the monitoring agencies.
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 12 of 17
Preliminary Crosstab Reports by Site, Parameter Occurrence Code (POC), and Scheduled
Dates
These are examined and anomalous results are investigated and corrected. The following
crosstab tables are routinely generated:
• Chain of Custody (COC) form number for each event - include field and trip blanks,
routine samples, unscheduled blanks. Missing cells in the table are investigated.
• Total counts of AQS-deliverable records - each sampler type should generate a
specific number of counts; exceptions are investigated.
• Counts of invalid or suspicious analyte records.
• Sampled date (i.e., the date actually sampled, as recorded on the COC form) - date
scheduled is compared against date sampled. Any event where these do not agree is
investigated.
Examination of Chain of Custody Forms
COC forms are inspected for completeness by QA personnel. Flags assigned by the site operator
or by the support laboratory's sample handling or data entry personnel are reviewed and
corrections are made if necessary.
Statistical Outlier Checks and Range Checks.
Limits used for outlier checking are provided in Tables 22-4 below. The Quality Assurance
Organization reviews these results and investigates problems. Outlier checks include the
following:
• Lower Limit on PM2.5 Mass - Routine samples only.
• Mass Balance - Routine samples only.
• Anion/Cation ratios - Routine samples only.
• Sulfur/Sulfate ratios - Routine samples only.
• Upper Limit on PM2.5 Mass - Blank samples only.
These steps are repeated until the data base is consistent and all exceptional conditions are
explained.
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 13 of 17
Entry and Verification of Data Changes from SLT Agencies and AQS Data Verification
The following step is done approximately 2 months after validation of the preliminary data set.
• Changes requested by the SLT agencies are entered into the support laboratory's data
base.
• The AQS text file is generated and is copied back into a QA table in the data base
which is used for reporting and automated comparisons.
• Crosstab tables are made from the AQS data table; these tables are compared with
previously generated crosstab tables for the reported batch. These are inspected as
described above.
• After any anomalies are corrected, the AQS text file is regenerated. This is repeated
until all problems are corrected. See Tables 22-3 through 22-6.
Table 22-3. Statistical Validation Limits for Blanks
Sampler/
Analyte(s)
MetOne/
PM25 Mass
MetOne/
PM25 Mass
MetOne/SHA
L
PM25 Mass
URG 3000N/
Carbon
URG 3000N/
Carbon
URG 3000N/
SHAL/Caibon
Percentile
5.0
5.0
2.75
2.75
2.75
2.75
Tail
Upper
Upper
Upper
Upper
Upper
Upper
Sample Type
Field Blank
Trip Blank
SHAL Blank
Field Blank
Trip Blank
SHAL Blank
Limit
fig/filter
20.00(a)
14.00(a)
TBD(b)
6.7
6.5
1.8
Flagged
Analytes
PM25 Mass
PM25 Mass
PM25 Mass
Total Caibon
Mass
Total Caibon
Mass
Total Caibon
Mass
Internal
Flag
QL1
QL1
QL1
QL1
QL1
QL1
(a) The orginal value for outliers was based on the MDLs for PM2 5 mass as measured by an FRM with a flow
rate of 16.67 liters per minute. These new values are using the same regulatory MDL, 2.0 ug/m3, multiplied
by the flow rate of the MetOne sampler, 9.7 m3/24 hr. event. The values are supported by data as recent as
2009 and 2010. See Annual Data Summary Report for the Chemical Speciation of PM25 Filter Samples
Project: January 1 through December 31, 2010. .
(b) The MetOne SHAL blank will have to be computed from data that will be collected over the first 3 years of
its collection, beginning in 2012.
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 14 of 17
Table 22-4. Statistical Validation Limits for Routine Data
Percentile
2.0
2.0
2.0
2.0
2.0
2.0
2.0
Tail
Lower
Lower
Upper
Lower
Upper
Lower
Upper
Analyte(s)
PM25 Mass Cone.
Anion/Cation Ratio
Anion/Cation Ratio
Mass Ratio
Mass Ratio
Sulfur/Sulfate Ratio
Sulfur/Sulfate Ratio
Sample Type
Routine
Routine
Routine
Routine
Routine
Routine
Routine
Limit
^g/m3
2.98
0.86
2.82
0.60
1.32
0.25
0.45
Flagged
Analytes
PM2s Mass Cone.
all ions
all ions
all analytes
all analytes
ions, XRF
ions, XRF
Internal
Flag
QL1
QAC
QAC
QMB
QMB
QCR
QCR
Table 22-5. Mapping of Outlier Flags Onto AQS Codes
Objective Cause Found for Level 1 Outlier
Lab Error
Filter Damage
Module Assignment Error
Sampler Malfunction [4]
Unusual Conditions noted by operator [4]
Unknown Cause [4]
AQS codes
If NOT Invalid
(Suspicious)
[1]
[1]
[1]
(N/A)
[2]
T
If Invalid
AR
AJ
AQorAR
AN
[3]
AS
Notes:
[1] - No appropriate AQS validity status code exists in current AQS.
[2] - Use the applicable AQS validity status code, or T.
[3] - Use the applicable AQS null value code, or AM.
[4] - Other flags may be assigned in consultation with the Monitoring Organization to more precisely define the
cause of the problem.
(N/A) - Not Applicable
Table 22-6. Automated Range Checks
Parameter
Exposure Duration, texp
Holding Time before removal
from sampler, (FHT)
Average Flow Rate, Favg
Temperature Reasonableness
(all temperature channels)
Barometric Pressure (all)
Limits
23 96 hrs IF the sampling
schedule is not based on the
sequential samplers, then 48 hours.
within 10% of target flow rate
-20 < T < 45 (could vary by site
and season)
550
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
_ Page 15 of 17
22.8 Treatment of Deviations From Requirements
Deviations from requirements call for a variety of response activities that are summarized below:
Flag Data in ^(AS*— Chemical speciation data should be marked with a data validity flag only if
the data are considered valid for most purposes and uses.
Invalidate Data in AQS — Data of uncertain origin and data with unacceptable levels of
uncertainty should not be included in AQS. The corresponding concentration data will be
reported as missing with an AQS missing value code substituted.
CAR Process — The Monitoring Organization's CAR process or equivalent should be followed in
cases of systematic problems or problems affecting a significant number of data points The
CAR process is described in Section 22.2.
Revision of SOP s and Other Project Documentation — One of the most significant of the possible
outcomes of the CAR process is the identification of the need for revising SOPs and other project
documentation. Procedural changes to overcome identified problems are a key element in the
continuous improvement of the CSN program.
Consultation to Determine Impact of Deviation — The cognizant QA supervisor in an
organization is typically charged with the responsibility of determining the impact of a deviation
from requirements on data quality. This process should involve consideration of the primary
DQO of trend detection described in Section 7.0 as well as other potential uses of the data such
as source attribution.
Notification of EPA or Other Stakeholders — The investigation of a serious or systematic problem
should consult operators, analysts, and other personnel involved with the situation being
investigated, as well as stakeholders who might be impacted by the decision to validate or
invalidate data. Field organizations should contact the CSN laboratory (through channels
approved by the delivery order project office) to provide documentation of corrective actions that
might affect the data validation status of reportable data. The state Monitoring Organizations
should include significant QA problems in their annual QA reports to management.
22.9 Verification and Validation Criteria: Field Component
Table 22-7 summarizes the verification and validation criteria applicable to the field program.
The verification and validation (level 0 and level 1) criteria are combined into a single table
because the criteria overlap.
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 16 of 17
Table 22-7. Data Verification and Validation Summary
Item
Sampler and site identification
Date and time identification
Filter or denuder channel
assignment
COC and Data Form records for
filters and denuders
Holding times and shipping and
storage conditions:
• unexposed samples,
holding time, and storage
temperature
• exposed samples,
residence time in sampler
before retrieval
• exposed samples, on-site
holding time plus shipping
time, and shipping
temperature
• holding time and storage
temperature after receipt
in the laboratory prior to
analysis
Electronic data transmission
Sensors checked or calibrated
within required time frame
Transfer standards recertified
within required time frame
Criteria
Must be
correct
Must be
correct
Must be
correct
Must be
present
See criteria
tables in
Section 20.0.
No errors
reported
All
temperature,
barometric
pressure, and
flow rate
sensors
All transfer
standards
routinely
used by site
operator
Applicable
to
Data forms
Data forms
Data forms
COC forms,
sample labels
All filters
Electronically
transmitted
data
Routine
calibration
records
Audit and
recertification
records
Comment
Investigate and correct, if
possible; invalidate data
if identification cannot
be established
Review and determine
impact on data validity
Review and determine
impact on data validity
Review and determine
impact of the error on
data validity
Dates of required sensor
checks should be verified
before data are reported
Dates of recertifications
should be verified before
data are reported
Flag in
AQSif
Violated
0
(see
notes)
•
o
o
Invalidate
if Violated
•
(see notes)
•
•
0
0
o
-------
PM2 5 CSN QAPP
Section No. 22
Revision No.: 1.2.0
Date: 11/11
Page 17 of 17
Item
Sampler-generated flags
(electronic data):
• filter/ambient temperature
difference
• flow rate out of
specification
• total sampling time out of
specification
• sampling start or stop time
out of specification
Operator flags and other
information from field data
forms:
• filter inspections
• other conditions noted by
operator that could impact
the sample
Statistical screening of
operational data:
• temperature
• barometric pressure
• flow
• other operational data
Criteria
See criteria
table
See field
data form
See criteria
tables
Applicable
to
Electronic
data
Field data
forms
Data
summaries
Comment
Data should be reviewed
and evaluated to
determine validity for
reporting to AQS
Operator's notes should
be evaluated regarding
probable impacts on data
validity
Development of
statistical data screens
will be coordinated with
the laboratory
Flag in
AQS if
Violated
•
0
0
Invalidate
if Violated
0
0
0
Notes:
Flagging or invalidation is optional. Monitoring Organization should review the circumstances and the
potential impact on the data. When both columns are marked with this symbol, there are three choices: (1)
report to AQS without a flag if the data need not be qualified to the data user, (2) flag the data and report to
AQS if the user might need to know that the data are qualified but may be usable for certain purposes, and
(3) invalidate the data (invalid data are not reported to AQS).
Flagging or invalidation is mandatory. When this symbol is shown in the "Flagging" column and the
"Invalidation" column is blank, the data must be flagged and must be reported to AQS.
-------
PM2 5 CSN QAPP
Section No. 23
Revision No.: 1.2.0
Date: 11/11
Page 1 of 5
23.0 Validation and Verification Methods
The processes for verifying and validating the measurement phases of the chemical Speciation
Trends Network data collection operation have been discussed in the previous section. If these
processes are followed, the quality of data should be achieved to meet the data quality objectives
for trend detection. This section describes the organizational implementation of the validation
procedures, the applicability of corrective actions, and the reporting requirements and schedules.
The process of data validation and verification is a cooperative effort between the reporting
organization responsible for field sampling and the Chemical Speciation Network (CSN)
laboratory, which conduct the chemical analyses and reports the validated concentration values
to the Air Quality System (AQS). This section focuses on the verification and validation methods
applied by the reporting organizations, Monitoring Agencies, after the first stages of validation
have been completed by the CSN laboratory.
The Monitoring Agencies are encouraged to review and utilize similar procedures and tools for
validating the CSN data. To this end a document EPA under wrote the book "Data Validation
Process for the PM2.s Chemical Speciation Network" developed by RTI under contact. See it at
http://www.epa.gov/ttn/amtic/files/ambient/pm25/spec/05datval.pdf RTI was also commissioned
to develop an automated data review tool, the Speciation Data Validation and Analysis Tool
(SDVAT < http://www.epa.gov/ttn/amtic/sdvat.html>. The CSN contract support laboratory
provides technical support for this tool as well. The tool accepts the monthly-produced
spreadsheets posted on the secure CSN website managed by the lab service contractor. The
includes several of the data review methods listed below.
Data completeness.
Time series analyses.
Outlier tests.
Mass concentration reconstruction.
Pie chart species distributions.
Spatial and temporal variability analyses.
23.1 Interorganizational Responsibilities for Data Validation
The sequence of data verification and validation steps and the corresponding organizational
responsibilities are as follows:
1. The CSN laboratory issues unexposed sample module kits to the Monitoring
Agencies along with a partially completed Custody And Field Data Form (CAFDF).
2. The Monitoring Agency exposes the filters, fills in the necessary data about the
exposure on the CAFDF form, and returns them to the CSN laboratory.
3. The CSN laboratory enters the data recorded on the CAFDF, performs the data
analyses, and enters the results into the data system.
-------
PM2 5 CSN QAPP
Section No. 23
Revision No.: 1.2.0
Date: 11/11
Page 2 of 5
4. The CSN laboratory performs level 0 and level 1 data validation based on the data to
which it has access. This includes validation of laboratory results and entry of any
validation flags associated with problems noted on the CAFDF received from the
field.
5. After all validation is completed for a data set, the CSN laboratory transmits the data
and associated validation flags in hardcopy and electronic form to the delivery order
project officer (DOPO) who, in turn, distributes the data to the respective Monitoring
Agency.
6. The Monitoring Agency completes the verification and validation process at levels 2
and 3 based on information compiled from previous PM2.5 research programs,
guidance from the Clean Air Scientific Advisory Committee and Expert Panel on
PM2.5 Speciation, its own internal records, audit results, and other information
available from site operators and other local sources.
7. The Monitoring Agency forwards the validated data back to the CSN laboratory for
submission to AQS.
8. Further statistical validation may be performed by the EPA and others as needed.
23.2 Personnel Responsibilities Within the Reporting Organization
The Monitoring Agency should assign data validation responsibilities according to its
organizational structure and the needs of this program. It is recommended that the following
roles and responsibilities for CSN data validation be defined within the Monitoring Agency.
Program Manager—The Monitoring Agency's manager for the PM2.5 chemical CSN is
ultimately responsible for meeting schedules, for ensuring that qualified staff perform data
management and validation functions, and for delivering valid data to AQS. The Monitoring
Agency's CSN Manager should ensure that data validation and data management responsibilities
have been assigned to individuals with the appropriate educational background, training, and
knowledge of the CSN program.
Data Validation Specialist—The Data Validation Specialist is the person who conducts the actual
examination of data received from the CSN laboratory via the DOPO. This person should be
knowledgeable about quality assurance principles, particulate measurement methods, and basic
statistics and chemistry and should be familiar with network operations including sampler
operation and shipping and receiving of sample modules. Often, the quality assurance supervisor
responsible for the CSN program will be assigned to validate the data. The Data Validation
Specialist should be supported by data management personnel, chemists, environmental
scientists, and statisticians to deal with questions that arise during the validation process. It is
critically important that field personnel, particularly the operator responsible for CSN sampling
and his/her supervisor, be available to answer questions. The Data Validation Specialist should
also have ready access to project files and relevant data sets.
-------
PM2 5 CSN QAPP
Section No. 23
Revision No.: 1.2.0
Date: 11/11
Page 3 of 5
The following sections describe the verification and validation procedures that should be applied
by Monitoring Agencies to the partially validated data received from the CSN laboratory.
23.3 Completion of Level 0 Data Verification and Validation
The first step in the Monitoring Agency's data verification and validation process should be to
confirm that the data are correctly accounted for according to the Monitoring Agency's own
records. The CSN laboratory will have checked the data according to its shipping and receiving
records and the chain of custody data form and will have flagged any discrepancies in data
attribution. The Data Validation Specialist should examine each discrepancy that the CSN
laboratory has flagged as suspicious due to uncertain attribution or problems with chain of
custody. The Monitoring Agency should examine its copy of the CAFDF, shipping and receiving
records, operator logbooks, and any other relevant records to address the problem. (The CSN
support laboratory may have already contacted the Monitoring Agency, via the DOPO, regarding
problems of data attribution, so a repeated check of a previously investigated problem may not
be necessary.) The Monitoring Agency can deal with this type of flagged data in a number of
different ways, as follows:
• If the level 0 flag can be explained and the data are correct, the flag may be removed
or changed to a more appropriate flag.
• If the CSN laboratory has flagged data as misassigned (i.e., assigned to the wrong
sampling date, site, analyzer, or channel), the Monitoring Agency may edit the data
so that monitoring data are correctly assigned. This may only be done if verifiable
information can be found in the Monitoring Agency's operating records that allow the
correct data assignments to be made.
• If concentration values that have been flagged can be corrected numerically (e.g., by
correcting an incorrect sample volume recorded on the CAFDF sheet), the Monitoring
Agency can make this correction directly, but it is recommended that this information
be passed back to the CSN laboratory, via the DOPO, so that the calculation can be
verified.
• If the Monitoring Agency cannot find an explanation for a flagged discrepancy, the
data flag should be allowed to stand.
23.4 Identification of Outliers and Data Flagging Techniques
23.4.1 Manual Methods
To fully complete the validation process, the Monitoring Agency must examine the data set, both
flagged and unflagged data, for validation criteria based on information sources available within
the organization. The validation tables in Section 22.0 of this document should be consulted for
specific validation criteria. Some validation criteria that are not easily automated are listed
below:
-------
PM2 5 CSN QAPP
Section No. 23
Revision No.: 1.2.0
Date: 11/11
Page 4 of 5
Manual Data Inspection—The purpose of manual data inspection is to spot unusually
high (or low) values that might indicate a gross error in the data collection system. It
is often helpful to plot data in a time series.
Systems Audit Report Results—Audits occasionally turn up serious deficiencies that
could affect the validity of the data. For example, if it is found that a field operator is
mounting sample modules incorrectly, it would be necessary to flag or invalidate all
the data corresponding to that sample module when that operator was working.
Performance Audit Results—Large deviations from acceptance or advisory limits for
critical measurements such as sample volume accuracy could result in data
invalidation or flagging back to the last previous acceptable audit or control check
result. Poor performance audit results should be investigated further, and if a specific,
identifiable problem is uncovered that affects reportable data, that data should be
flagged or invalidated. In general, reportable data should not be invalidated unless a
specific, identifiable cause for the discrepancy can be found.
Collocated Sample Results—Performance on collocated duplicates should be
evaluated if information is available. Collocated sampling equipment may include
FRM samplers operated for the PM2.5 mass network. Poor collocated sample results
should be investigated further, and if a specific, identifiable problem is uncovered that
affects reportable data, that data should be flagged or invalidated. In general,
reportable data should not be invalidated unless a specific, identifiable cause for the
discrepancy can be found; however differences greater than a factor of 2 can be
invalidated at the discretion of the Monitoring Agency using a flag such as AM -
Miscellaneous Void.
Operator's Notes and Site-Specific Information—Operator's notes can contain
information that would call for data invalidation due to lack of sampling
representativeness. Examples include meteorological events such as sand storms,
temporary violations of siting criteria such as nearby construction, or operational
difficulties with the sampling equipment. The Monitoring Agency should use its best
judgment about the impact of site conditions on the acceptability of the data and may
consult with EPA via the DOPO or the Regional Office if there are questions.
Shipping Records—Shipping records can be compared with CAFDF records to
identify exposed sample sets that were held too long before shipping. Unexposed
sampling media should not be used if more than 30 days have elapsed since the initial
weighing of the Teflon™ filters. Sample module sets should be used at a designated
site in the order they were received from the CSN support laboratory.
-------
PM2 5 CSN QAPP
Section No. 23
Revision No.: 1.2.0
Date: 11/11
Page 5 of 5
• Corrective Action Requests that Affect Data Quality—The Data Validation Specialist
should review any corrective action requests in effect when the samples were
acquired. Any corrective action requestions that could affect the data should be
evaluated and appropriate actions taken with regard to flagging or invalidating the
data.
23.4.2 Automated Methods
Because the data will be downloaded from the CSN laboratory's website to the Monitoring
Agencies in Excel spreadsheet format, automated checking methods have been implemented.
Some automated screening methods of most use for assessing the CSN datasets include the
following:
• Interparameter Checks—These include ion ratio and mass balance checks that use
data from a number of different channels. Note that each sampling site will have a
pattern or cycle of relationships that are set up by seasonal patterns, local sources of
PM2.5, and geography. Samples with atypical results could be examined more closely
as part of the validation process.
• Time Series Analysis—This analysis is typically the examination of a set of data for a
single observable (e.g., a particular chemical species at a certain site) acquired over a
period of time. Time series data are often best examined graphically, and it is often
helpful to chart related variables together on the same graph.
• Outlier Checks—Statistical outlier checks for screening PM2.5 chemical speciation
concentration measurements (the actual environmental measurements, rather than
quality assurance/quality control data) are another means of identifying potential
problems. An environmental observation should never be invalidated simply
because it is identified as a possible outlier by statistical techniques. Observed
environmental concentration distributions tend to be somewhat skewed, so that a
small number of concentrations significantly higher than the long-term average
should be expected. Selecting the top 2 to 5 percent of values in a data set for
investigation; however, is often a good rule of thumb for data assessment because
high data values are sometimes the result of analytical, procedural, or calculation
errors.
.Once potential outliers or other suspicious data points are identified, a more in-depth
investigation may be performed to determine it they are invalid.
-------
PM2 5 CSN QAPP
Section No. 24
Revision No.: 1.2.0
Date: 03/12
Page Iof8
24.0 Reconciliation with Data Quality Objectives (DQOs)
Results obtained must be reconciled with the requirements defined by the data user or decision-
maker, as specified by the DQOs for the project (U.S. EPA, 1994). The methods initially
envisioned to analyze the data were based on the statistical model described fully in Data
Quality Objectives for the Trends Component of the PM2.5 Speciation Network (U.S. EPA,
1999a. See http://www.epa.gov/ttn/amtic/specgen.html). This section discusses measurements of
the MQOs and components of uncertainty that were made from 2000-2010 and compare them to
the values that were used as preliminary assumptions to develop the original DQOs. Several
analyses of speciation data will be summarized to illustrate if the original DQOs have been met.
The tools that were used will be briefly discussed and references supplied so that future analyses
can be constructed and DQOs retained or modified to suit the needs of environmental decision-
makers.
24.1 DQO for Chemical Speciation Trends
The primary DQO for the Chemical Speciation Network (CSN) was stated to be the ability to
detect a +5 percent trend within 5 years with statistical power of 0.80. To assess whether this
DQO has been met, it is necessary to determine if a significant time trend can be detected (or
rejected, when there is no trend) with the requisite statistical power after 5 years of data have
been collected. Note that by satisfying the DQO it is not necessary to show that a trend
definitely exists or does not exist; a legitimate finding is that a trend cannot be either diagnosed
or rejected with the required certainty. In fact, this is a likely outcome when a trend of
intermediate size (e.g., +2 to 3 percent per year) exists or when a trend is present but highly
variable over time. On the other hand, if all sources of error and uncertainty are particularly well-
controlled, it may be possible to diagnose a trend smaller than +5 percent over the designated 5-
year period. To satisfy the primary DQO, it is only necessary to demonstrate that the 5-year data
set is capable of detecting a 5 percent trend (or failing to find a trend when none exists) with the
requisite level of confidence. This is done on a species-by-species basis. All the species used in
the study, however, were present at a relatively high level; it is not likely that the trend detection
DQO will be met for all chemical species. Reasons for failing to meet the trend detection DQO
for a particular species include the following:
• Low concentration (typical of certain uncommon elements reported by energy-
dispersive X-ray fluorescence [EDXRF]).
• Large background variability near the analytical method's detection limit.
• Large proportion of "non-detects", thus weakening the statistical power of the
analysis.
• High variability in concentration of the species in the environment.
• Seasonal variability.
• The presence of local sources (ocean, nearby construction, industrial or residential
sources).
-------
PM2 5 CSN QAPP
Section No. 24
Revision No.: 1.2.0
Date: 03/12
Page 2 of 8
• Method or equipment changes that prematurely interrupt any five year sequences may
introduce enough bias or other uncertainty to essentially reset the baseline year.
• Moving a site prior to 5 years of operation probably precludes the determination of a
trend at the initial site. The baseline year for a new trend determination is established
by the restart of the sampler at the new location.
24.2 Interim Evaluations of Data Quality
The CSN and its constituent monitoring agencies should perform interim evaluations of data
quality annually to assess whether the goal of meeting the DQO for trend detection can be met
within 5 years. Here are two primary ways of performing interim evaluations of data quality:
comparison of the project's quality control (QC) statistics against measurement quality objectives
(MQOs) and direct modeling using the method used by EPA in the DQO study. The following
sections describe these two approaches and suggest corrective actions that can be taken if interim
analysis indicates that the DQO may not be met after 5 years.
24.2.1 Evaluations Based on MQOs
Estimating the measurement error being achieved based on available QC data is perhaps the
quickest interim measure of assessing progress toward meeting the DQO for trend detection.
Objectives for total measurement error based on the DQO study are provided in Section 7.0 of
this document. This quick but inexact method of data quality assessment is appropriate after
1 year of sampling; however, analysis that is more exhaustive should be done after 2 or 3 years
of data have been collected, as described in the next section.
The CSN program has various measures of bias and precision available as a result of the QC data
being taken. Measurement error must be assessed as the total end-to-end error. Some QC
samples assess only part of the total system and thus underestimate the total error. Some of the
QC samples that are useful in assessing total measurement error are as follows:
Flow Rate Checks—these checks are carried out at various intervals using different independent
flow standards. Total volume is directly proportional to flow rate, and calculated concentration is
inversely related to flow rate. Thus, a 5 percent bias error in flow rate will result in a 5 percent
error in calculated concentration (approximately, not accounting for the effect of such an error on
the particle size range distribution, that is, a change in the cut-point of the inlet due to a change
in flow rate)1.
Another statistical check on flow rate uncertainty is its coefficient of variation (CV) during a
sampling event. There are several forms of the metric; however, they generally involve
averaging the differences in polled flow rates at specified intervals; calculating a standard
deviation of the differences and dividing the standard deviation by the average. The CV metric is
always stated in absolute value; consequently, the data analyst should examine the polled flow
rates if a CV of greater than 4% is reported. If it is directional in nature in flow rate could result
in a calculation error in concentration. If a direction is not discernable then the calculated
1 Error due to particle size distribution will only occur if there is a distribution. If for example all the PM is PM25,
the separator will have no effect on the mass that is removed—there will be none.
-------
PM2 5 CSN QAPP
Section No. 24
Revision No.: 1.2.0
Date: 03/12
Page 3 of 8
volume over the duration of the sampling event may be considered as an accurate estimate. If
CV >5% and there is a significant PMcoarse (PM 10-2.5) concentration in the ambient air sampled
by that monitor, then the effect of the variable flow should be estimated. The amount of time that
the flow is less than 10% of the design rate will be biasing the apparent PM2.5 mass on the high
side. But the chemical constitution of the ambient air will affect other values. For example, if
geo-crustal components dominate the PMcoarse fraction, the PM.25 mass and the XRF results
may be the only values biased by the aberrant flow rate. Finally flow rate is only one factor in
the estimate of total measurement error because it does not include sampling representativeness
and analytical errors.
Analytical QC Samples Including Analysis of Standards, Duplicates, and Matrix Spikes—These
samples provide information about the analytical component of accuracy and precision.
Theoretical estimates of uncertainty are also available for the EDXRF data (elemental analysis).
These estimates do not completely characterize the total measurement system because they omit
field errors, including flow rate and sampling representativeness.
Field Blank Results—Field blanks provide information on contamination due to handling
operations within the laboratory and in the field, plus any of the species of interest that might
have been present in the clean filter, for example. Normally, field blank levels should be kept
quite small by early corrective actions and should never be allowed to become a significant
component of total measurement error.
Laboratory Blank Results— Laboratory blanks provide information on contamination due to
handling operations within the laboratory plus any of the species of interest that might have been
present in the clean filter, for example. This allows for separation of the sources of error
associated with the laboratory versus the field portion of the process. Normally, laboratory blank
levels should be kept quite small by early corrective actions and should never be allowed to
become a significant component of total measurement error. Filters high in the analyte of interest
also should never be allowed to become a significant component of total measurement error, if
appropriate QC is being conducted.
Collocated Sampling Results— Collocated sampling may be utilized for assessing two different
types of measurement variability. One type is when a replicate sampler is placed side-by-side
with the primary network sampler and operated by the same technician using the same
procedures. Data from several sampling events using this tactic provides a estimate of precision
for a given site and sampler. If data from several sites that use the replicate pairs of the same
sampler is compiled, it provides an estimate of general precision using that particular sampler
along with the analytical measurements associated with the sampling.
If the network employs several makes and models of samplers that are designed to generate the
same data, a measurement of relative bias of each make of sampler must be ascertained. It may
be done once as was the case for the CSN samplers that were originally approved for use in the
network. By having all the samplers together each sampler's value would be compared to the
-------
PM2 5 CSN QAPP
Section No. 24
Revision No.: 1.2.0
Date: 03/12
Page 4 of 8
average of the entire group's values on each day they sample. EPA determined that the
originally approved samplers would meet the needs of the CSN's monitoring and quality
objectives.
Another way to determine bias is with a longer term program that periodically collocates every
sampler in the network with the single make and model of sampler, which is believed to achieve
excellent precision and to most accurately represent the parameter being measured. The so-called
reference sampler would be operated by an independent technician and the sample would be
analyzed by an independent laboratory, which has demonstrated rigorous quality control and
quality assurance procedures. The results provide a measure of relative bias and if the results are
widely disparate other issues may be revealed. This strategy was not employed by the CSN
because the costs of running the independently samplers and the chemical speciation analyses
that would be necessarily performed on the acquired filter samples of each sampling event.
When starting up a new chemical speciation sampler at a site, it is a good idea to collect a
number of duplicate samples with a replicate collocated sampler. Approximately ten valid
duplicate samples should be considered the minimum for estimating the total error. These should
be collected over a given season.
Continued collocations should be done on a regular schedule so that data can be developed over
all seasons of the year. Although the collocated sampling results provide a relatively complete
picture of end-to-end measurement error, two components of measurement are not included.
The first is related to the fact that collocated samples are generally analyzed in the same
laboratory on the same day. The collocated sampling error will underestimate the true total
measurement error by an amount related to the laboratory's day-to-day variability. This error can
be controlled, or at least estimated, by tracking the results of daily analyses of laboratory
standards. The variability in repeat analyses of laboratory standards should be kept small with
respect to the targeted MQO. The CSN laboratory will provide the EPA with regular quality
assurance (QA) summary reports that will include the necessary information for assessing this
and other components of analytical error. This error is expected to be small, compared to the
total measurement error.
The second error that collocated sampling omits is due to field sampling representativeness
caused when the paired samplers are not equivalently sited. This factor can be controlled by
careful observance of the siting requirements (for example, separation of sampler inlets by at
least 1 meter but no more than 4 meters) as well as by using common sense in making the two
sampler locations as equivalent as possible. Problems with siting might also be detected by
assessing the duplicate precision for total mass. Consistent relative differences in total mass of
more than about 10 percent may indicate a siting problem or a problem with one or both of the
samplers exists. Low-level chemical species are likely to show larger relative variability than
mass measurements—this is to be expected and may not indicate a problem. If siting and sampler
Evaluation of PM2.5 Chemical Speciation Samplers for Use in the EPA National PM2.5 Chemical Speciation
Network; EPA-454/R-01-005, JulylS, 2000. http://www.epa.gov/ttn/amtic/files/ambient/pm25/spec/fourctv.pdf.
-------
PM2 5 CSN QAPP
Section No. 24
Revision No.: 1.2.0
Date: 03/12
Page 5 of 8
operation are carried out correctly, this "error" should not be an issue with regard to total
measurement uncertainty.
Finally to get a better characterization of precision across the network the collocations should
occur at multiple sites. Through the first 12 years of the CSN only 6 sites were designated for
collocated measurements. By 2013 collocated sampling at half of these sites will be suspended
and the replicate samplers will be rotated among other CSN sites for 1 year collocations.
The recommended interim method for estimating total measurement error is to calculate the total
error based on ten or more collocated measurements for each chemical species and mass. The
following method of calculation is adapted from the method given in 40 Code of Federal
Regulations (CFR) Part 58, Appendix A, Section 5.5.2.1:
1. Calculate the duplicate difference for each observation, as follows:
Y — X
xlOO
where
di = percent difference for observation /'
YJ = primary (station) sampler concentration value
Xi = duplicate (reference) sampler concentration value.
Note: Omit data for which the average concentration in the denominator is less than 3 times the
method detection limit. This may result in inadequate data for evaluating some trace species.
2. Calculate the CV for a single check. The following equation for calculating the CV is
provided in 40 CFR Part 58, Appendix A, Section 5.5.2.2:
where
CVi = CV for observation /
di = duplicate percentage difference for observation /'.
3. Calculate the single sampler precision. The following equation is adapted from 40
CFR Part 58, Appendix A, Section 5.5.2.3, which calculates precision for collocated
samplers of identical type CVj represents an average error value over all observations
-------
PM2 5 CSN QAPP
Section No. 24
Revision No.: 1.2.0
Date: 03/12
Page 6 of 8
within a given time period for a particular chemical species (or total mass) designated
by the subscript/
I crj
where
j = pooled CV for speciesy over the specified time period
ij = CV for speciesy, observation /
= number of paired observations made for speciesy over the time period.
Compare the CVj results against the MQO. If the CV does not meet established
criteria for a species of interest, corrective actions such as those discussed in
Section 24.2.3 should be considered. Corrective actions should be taken promptly
after identification of a problem; do not allow a problem to linger, otherwise
significant amounts of data may be invalidated.
24.2.2 Evaluations Based on Direct Assessment of the Monitoring Data
Using the statistical model described in the DQO (U.S. EPA, 1999a), an assessment of progress
toward meeting the DQO can be made with less than 5 years' worth of data but with reduced
statistical power. The statistical power of the tests should be extrapolated to 5 years, based on the
error levels computed for the preliminary data. If this extrapolation indicates that the DQO for
detecting a trend will not be satisfied for a critical species of interest, corrective actions should
be taken as described in the next section. It is important to do a careful assessment of the actual
monitoring data as early in the program as possible, typically after the second or third full year of
sampling, because the assumptions used in the DQO study may not hold at any particular CSN
site.
24.2 3 Interim Corrective Actions
If an interim assessment indicates that the DQO for trend detection is not being met,
modification to the experimental design should be considered after any obvious measurement
quality problems have been resolved to. Experimental design factors include site selection,
frequency of sampling, frequency and precision of QC measurements, and frequency of
equipment maintenance. The experimental design changes most likely to improve data quality
are listed below in decreasing order of effectiveness:
-------
PM2 5 CSN QAPP
Section No. 24
Revision No.: 1.2.0
Date: 03/12
Page 7 of 8
• Increase the Frequency of Sampling—The DQO study (U.S. EPA, 1999a) showed
that l-in-3 day sampling was adequate to meet the DQO based on the IMPROVE data
set, while daily sampling was unnecessary and l-in-6 day sampling was not adequate
to meet the DQO. Increasing the sampling frequency may be helpful when
measurement error and unexplained variability are larger than expected. Increasing
the frequency of sampling is unlikely to affect variability attributable to seasonality.
• Add Additional Samplers—Locating one or more additional samplers in the same
impact area may help decrease the statistical uncertainty in the same way that
increased sampling frequency does. In addition, locating samplers at some distance
from one another may be effective in reducing unexplained variability due to local
sources and siting variables.
• Remedy Siting Problems—Factors such as proximity of local sources or shielding by
nearby buildings and other objects should be eliminated as potential sources of
excessive variability.
• Improve QC and Maintenance—Measurement errors can, in principle, be reduced by
increasing the frequency of QC checks, audits, and maintenance. Increasing the
intensity of the QC program, however, may not be the most effective approach for
reducing total uncertainty for two reasons: the relative contribution of measurement
error to the total uncertainty is small, and a point of diminishing returns may be
reached after which little improvement in measurement quality can be achieved.
Purchasing more precise or accurate standards is unlikely to make a meaningful
difference in overall data uncertainty. On the other hand, increasing the frequency of
QC and maintenance can sometimes be effective at sites where a excessive numbers
of operational problems and malfunctions are being seen and will certainly be
effective in the latter years of network operation as the equipment ages and major
maintenance is required.
24.3 Assessing and Reporting Chemical Speciation Trends
At the end of the first 5 years of monitoring, every monitoring agency should assess whether or
not the DQOs and MQOs for trend detection has been met and will apply suitable statistical tests
to test for a trend in the concentration data for all chemical species of interest. Detailed
description of the trend assessment method is outside the scope of this QAPP. Network and
reporting agency personnel should use methods similar to those in the DQO document (U.S.
EPA, 1999a) as a point of departure for their analysis of concentration trends, but they are
encouraged to use the most appropriate statistical model for their individual situations. Specific
assistance can be obtained from OAQPS in Research Triangle Park, North Carolina.
The interim DQO analyses should address the following questions for each analyte:
• Was an annual trend of+5 percent or greater indicated by the analysis?
• Was an annual trend of as much as +5 percent excluded by the analysis?
• Was the statistical test inconclusive about the existence of a trend?
-------
PM2 5 CSN QAPP
Section No. 24
Revision No.: 1.2.0
Date: 03/12
Page 8 of 8
• If the test was inconclusive, was the data of sufficient quality to make the assessment
with the requisite power if a trend had been present?
24.4 Reconciling Other CSN Research Objectives
There are several important research objectives for the CSN data other than trend identification
(U.S. EPA, 1999b). These include model development and validation, source attribution, spatial
and seasonal characterization of aerosol distributions, state implementation plan attainment and
strategy development, and emissions inventory. The ultimate users of the data include
environmental researchers, regulators, and state and federal policymakers. The DQO for trend
detection focuses on changes in concentrations of individual species over time. Other data uses,
however, may rely on different characteristics, such as concentration ratios between species,
seasonal variations in concentrations or concentration ratios, or the absolute concentration of
certain chemical species at a particular point in time. Meeting the DQO for trend detection does
not guarantee suitability of the data for another purpose. To be useful for objectives other than
trend detection, the primary data set must be accompanied by a complete set of supporting data
so that the user can derive information that might be applicable to other research objectives.
24.5 References
U.S. EPA (Environmental Protection Agency). February 2006. Guidance on Systematic
Planning Using the Data Quality Objectives Process: EPA QA/G-4, Report No. EPA/240/B-
06/001, U.S. EPA, Washington, DC.
U.S. EPA (Environmental Protection Agency). February 2006. Data Quality Assessment: A
Reviewer's Guide, EPA QA/G-9S, Report No. EPA/240/B-06/002,
U.S. EPA (Environmental Protection Agency). 1999a. Data Quality Objectives for the Trends
Component of the PM2.5 Speciation Network, U.S. EPA, Research Triangle Park, NC, 1999.
(Available online on the Ambient Monitoring Technical Information Center [AMTIC] at
http://www.epa.gov/ttn/amtic/files/ambient/pm25/spec/dqo3.pdf)
U.S. EPA (Environmental Protection Agency). 1999b. Particulate Matter (PM2.s) Speciation
Guidance Document (FinalDraft), U.S. EPA, Research Triangle Park, NC. October 7, 1999.
(Available at: http://www.epa.gov/ttn/amtic/files/ambient/pm25/spec/specfml.pdf)
U.S. EPA (Environmental Protection Agency). JulylS, 2000. Evaluation of PM2.5 Chemical
Speciation Samplers for Use in the EPA National PM2.5 Chemical Speciation Network; EPA-
454/R-01-005. (Available at:
http://www.epa.gov/ttn/amtic/files/ambient/pm25/spec/fourcty.pdf)
-------
PM2 5 CSN QAPP
Appendices
Revision No.: 1.2.0
Date: 06/12
Page 1 of4
Appendices
The original appendices A-l through A-7 have been either deleted or incorporated into the following
appendices A-lthrough A-4 as appropriate due to the consolidation of the network to Metone SASS or
SuperSASS and the URG 3000N carbon samplers. They are incorporated in this QAPP by reference to
reduce redundancy of electronic files as well as the size of this file. Each may be reviewed and
downloaded from EPAs Ambient Monitoring Technical Information Center on the Technology Transfer
Network. The web address is http://www.epa.gov/ttn/amtic/spectraining.html.
Appendix A-l: Met One SASS Field Operation Manual (PDF) (79pp, 648 kb) - 12/27/2001
Appendix A-2: Met One SASS Standard Operating Procedure (SOP) (PDF) (12pp, 89 kb) - 7/27/2011
Appendix A-3: Standard Operating Procedure (SOP) for the URG 3000N Sequential Particulate
Speciation System (Sampler) (PDF) (82pp, 2.4 MB) - 8/11/2011
Appendix A-4: The URG3000N Operator's Manual Version 5.6 (PDF) (75pp, 5.8 MB) - 7/17/2009
The current QAPP for the national contract service laboratory is also located at
http://www.epa.gov/ttn/amtic/specguid.html.
National Contract Service Laboratory Quality Assurance Project Plan: Chemical Speciation of PM2.5
Filter Samples (PDF) (95 pp. 1.1 MB) 05/11/2012
The SOPs associated with all activities covered by the national service contract are integral to the
operation of the Chemical Speciation Network. These SOPs are identified below and may be accessed for
review and download at http://www.epa.gov/ttn/amtic/specsop.html.
Note the national contract service laboratory's QAPP and associated SOPs could change upon award of
the next service contract.
-------
PM2 5 CSN QAPP
Appendices
Revision No.: 1.2.0
Date: 06/12
Page 2 of4
U.S. ENVIRONMENTAL PROTECTION AGENCY
AMTtC Home
Basic Information
SLAMS Networks
Other Networks &
Partners
Air Monitoring Methods
Quality Assurance
Regulations &
Guidance
Training & Calendars
Program Review &
Oversight
Related Links
Ambient Monitoring
Mailing List
Technology Transfer Network
Ambient Monitoring Technology
Information Center
Search: OAII EPA ©AMTIC
You are here: EPA Home » Air S. Radiation » Technology TranErer 'i£!i'.'C.r< -"nziient Monitoring Technology
fnFormation Center » S-_A'vlS i.et.-.cr<= »• Chemical Sedation » Laboratcr/ Standard Operating Procedures
Chemical Speciation -
Laboratory Standard Operating Procedures
(SOPs)
The Chemical Speciation Network (CSN) routinely measures
PM2.5 mass, elements, ions, and carbon species. Several SOPs
are used to support the laboratory operations for the analysis of
these species and the list of routine SOPs are provided below.
In addition, there are SOPs that are used to support laboratory
operations for optional or supplemental analyses and they are
listed at the bottom of this page. The optional SOPs are available
for use in special projects as needed and not used in the routine
CSN.
Routine CSN SOPs
SOP for Procurement and Acceptance Testing for Teflon. Nylon,
and Quartz Filters (PDF) (6pp, 303 kb) - 2011
SOP for Cleaning Nylon Filters Used for Collection of PM-, .,
Material (PDF) (8pp, 338 kb) - 2009
SOP for Coating Aluminum Honeycomb Denuders with Mad (PDF)
(9pp, 298 kb) - 2009
SOP for PM Gravimetric Analysis (PDF) (23pp, 391 kb) - 2008
Chemical Speciation
Navigation
Chemical Speciaton Home
General Information
Field Standard Operating
Procedures
Quality Assurance
Data Management and
Reporting
Special Studies
Laboratory Standard
Operating Procedures
Newsletters
DIG Evaluation of the
Speciation Program
Semi-continuous Speciation
URG3000N Carbon
Conversion Project
SDVAT Speciation Data
Validation Tool
Sunset OC/EC Evaluation
Project
SOP for
SOP for
Cation Analysis (PDF) (llpp, 350 kb) - 2009
Anion Analysis (PDF) (12pp, 355 kb) - 2009
SOP for X-Ray Fluorescence Analysis of Particulate Matter Deposits on Teflon Filters (PDF) (17pp,
400 kb) -2009
Chester Lab Net SOP AD - 008.05 - Sample receipt and log in (PDF) (15pp, 935 kb) - 2008
Chester Lab Net SOP XR - 002.04 - Analysis of Elements in Air Particulates (PDF) (26pp, 1.8 MB)
2009
-------
PM2 5 CSN QAPP
Appendices
Revision No.: 1.2.0
Date: 06/12
Page 3 of4
Chester Lab Net SOP XR - 004.01 - Kevex XRF Spectrometer Calibration (PDF) (22pp, 1.5 MB) -
2008
Chester Lab Net SOP XR - 005.01 - Kevex Spectrometer Data Generation. Interpretation, and
Reporting (PDF) (18pp, 1.3 MB) - 2009
DRI SOP - DRI model 2001 - Thermal/Optical Carbon Analysis (PDF) (86pp, 2.4 MB) - 2008
SOP for Corrective Action for the PM2.5 Chemical Speciation Program (PDF) (4pp, 277 kb) - 2008
SOP for Assigning Data Validation Flags for the Chemical Speciation Network (PDF) (13pp, 652 kb)
- 2008
SOP for Database Operations (PDF) (18pp, 318 kb) - 2008
SOP for Long-Term Archiving of PM? ^ Filters and Extracts (PDF) (6pp, 280 kb) - 2009
SOP - Speciation Data Processing Disaster Recovery Plan (PDF) (7pp, 284 kb) - 2008
SOP for Document Control for the PM2.5 Chemical Speciation Program (PDF) (6pp, 291 kb) - 2009
SOP for Sample Handling and Archiving Laboratory (PDF) (31pp, 1.7 MB) - 2009
SOP for Shipping Filters to and from an Off-Site Laboratory (PDF) (Spp, 393 kb) - 2009
SOP for Training for Staff Working on the PM2.5 Chemical Speciation Program (PDF) (6pp, 292 kb)
- 2008
Optional or Supplemental CSN SOPs
SOP for the X-5eries ICP-M5 for the Analysis of Particulate Deposits on Teflon Filters (PDF) (2008)
(10 pp, 293 kb)
SOP for Temperature Calibration of the Sample Thermocouple in a Sunset or DRI 2001 Carbon
Aerosol Analyzer (PDF) (2009) (13 pp, 350 kb)
SOP for the Determination of Carbon Fractions in PM using the Improve - A Heating Protocol on a
DRI 2001 Analyzer (PDF) (2009) (26 pp, 490 kb)
SOP for the Determination of Carbon Fractions in PM Using the Improve - A Heating Protocol on a
Sunset Dual-Mode Analyzer (PDF) (2009) (26 pp, 460 kb)
SOP for Organic, Elemental, and Total Carbon in PM Using a Thermal/Optical Transmittance
Carbon Analyzer (PDF) (2009) (23 pp, 459 kb)
SOP for Coating R&P Speciation Sampler Chemcomb Denuders with Sodium Carbonate (PDF)
(2008) (6 pp, 326 kb)
SOP for Sample Preparation and Analysis of PM10 ad PM2.5 by Scanning Electron Microscopy
(PDF) (2008) (7 pp, 369 kb)
-------
PM2 5 CSN QAPP
Appendices
Revision No.: 1.2.0
Date: 06/12
Page 4 of4
SOP for Coating and Extracting Annular Denuders with Sodium Carbonate (PDF) (2009) (6 pp, 295
kb)
SOP for Coating Annular Denuders with XAD-4 Resin (PDF) (2008) (9 pp.. 288 kb)
DRI SOP Procedure for Light Transmission Analysis (PDF) (2008) (17 pp, 90S kb)
DRI SOP Analysis of Semi-Volatile Organic Compounds by GC/MS (PDF'i (2008) (26 pp, 1.3 MB)
SOP for Coating and Extracting Compact Parallel - Plate Denuders for Ammonia (PDF) (2010) (12
pp, 378 kb)
SOP for Coating and Extracting Denuders for Ammonia (PDF'i (2008) (8 pp, 302 kb)
EPA Home Privacy and Security Notice Contact
Us
nttp:,'Vwww. epa.gov/ttn/amtic/specsop, htm I
Print As-Is
Last updated on Tuesday, April ID, 2012
-------
United States Office of Air Quality Planning and Standards EPA-454/B-12-003
Environmental Protection Air Quality Assessment Division June 2012
Agency Research Triangle Park, NC
------- |