Quality Assurance Guidance
Document
Model Quality Assurance Project Plan
For the National Air Toxics Trends Stations
-------
EPA 454/R-02-007
December 2002
Quality Assurance Guidance Document - Model Quality Assurance Project Plan
For the National Air Toxics Trends Stations
By:
Dennis K. Mikel
Office of Air Quality Planning and Standards
Emission, Monitoring and Analysis Division
Research Triangle Park, North Carolina
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Emission, Monitoring and Analysis Division
Research Triangle Park, North Carolina, 27711
-------
Forward
The following document represents the model Quality Assurance Project Plan (QAPP) for the National Air
Toxics Trends Stations (NATTS). The Office of Air Quality Planning and Standards (OAQPS) staff
developed this Model QAPP to serve as an example of the type of information and detail necessary for the
documents that will submitted by State, Local or Trib al Air Toxics Monitoring Programs (ATMP)
involved in the NATTS.
This model QAPP was generated using the EPA QA regulations and guidance as 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 QAPP regulations and
guidance are addressed in this model
Data in this QAPP should not be used by organizations to meet the data quality needs for ATMP, with the
exception of Chapter 7, which describes the Data Quality Objectives (DQO) for the national program.
Since all NATTS will be part of this trends network, OAQPS requires that the DQOs be identical. If an
agency wishes to add DQOs to Chapter 7 that go beyond the needs of the national trends objectives, then
this should be addressed in Chapter 7 and 24. Chapter 24 discusses the Data Quality Assessments (DQAs)
for a local objective, which is not based on the DQOs in Chapter 7. Please note that Chapter 24 states that
the national DQA will be performed by OAQPS.
The Standard Operation Procedures (SOPs) listed in the Table of Contents are a guidance document
developed for OAQPS for the NATTS. This Technical Assistance Document (TAD) was created by
Eastern Research Group, Morrisville, NC and is available at the following Internet web site:
http://www.epa.gov/ttn/amtic/airtxfil.html. It should not be used verbatim, but as a guideline for
development of an agency's SOPs. SOPs must be developed by the State, Local and Tribal Agencies.
Please note that both Toxa City and the Toxa City Air Pollution Control District are fictitious.
-------
Acknowledgments
This Model QAPP is the product of the EPA's Office of Air Quality Planning and Standards. The
following individuals are acknowledged for their contributions.
Principle Authors
Chapters 1-6 and 8-24, Dennis K. Mikel, OAQPS-EMAD-MQAG, Research Triangle Park, North
Carolina
Chapter 7, Battelle Incorporated
Reviewers
Office of Air Quality Planning and Standards
Sharon Nizich
Comments and questions can be directed to:
Dennis Mikel, OAQPS, RTP,NC mikel.dennisk@epa.gov
111
-------
Acronyms and Abbreviations
ATMP
ANSI
APTI
AQS
ASTM
CAA
CFR
COC
cv
DAS
DQA
DQOs
EDO
EPA
GIS
GLP
HVAC
IO
LAN
LIMS
MQOs
MSA
MSR
NAAQS
NATTS
NIST
OAQPS
PAMS
PC
PD
PM10
PTFE
QA/QC
QA
QAAR
QAD
QAM
QAPP
QMP
SIPS
SLAMS
SOP
TCAPCD
TO
ISA
UATS
VOC
Air Toxics Monitoring Program
American National Standards Institute
Air Pollution Training Institute
Air Quality System
American Society for Testing and Materials
Clean Air Act
Code of Federal Regulations
chain of custody
coefficient of variance
data acquisition system
data quality assessment
data quality objectives
environmental data operation
Environmental Protection Agency
geographical information systems
good laboratory practice
Heating and Ventilating Air Conditioning Unit
InOrganic
local area network
Laboratory Information Management System
measurement quality objectives
metropolitan statistical area
management system review
National Ambient Air Quality Standards
National Air Toxics Trends Stations
National Institute of Standards and Technology
Office of Air Quality Planning and Standards
Photochemical Assessment Monitoring Stations
personal computer
percent difference
particulate matter- lOmicorns
polytetrafluoroethylene
quality assurance/quality control
quality assurance
quality assurance annual report
quality assurance division director
quality assurance manager
quality assurance project plan
quality management plan
State Implementation Plans
state and local monitoring stations
standard operating procedure
Toxa City Air Pollution Control District
Toxic Organic
technical system audit
Urban Air Toxics Strategy
volatile organic compound
IV
-------
Tables
Number Description
3 i Distribution List xil
5j ListofHAPs 3
6 \ Design/Performance Specifications-Total Suspended Particulates - Toxics Metals 2
6 2 Design/Performance Specifications-Air Canister Sampler - Volatile Organic Compounds 2
53 Design/Performance Specifications - Carbonyl Sampler - Aldehyde and Ketone Compounds 2
6 4 Design/Performance Specifications-Aethalometer -Black and Organic Carbon 3
55 Assessment Schedule 3
6 6 Schedule of Critical Air Toxics Activities 5
7 i DQO input parameter for benzene at urban locations 9
7 2 DQO output parameter for benzene at urban locations 9
7 3 DQO input parameter for benzene at rural locations 10
74 DQO output parameter for benzene at rural locations 11
75 DQO input parameter for 1,3,buadiene at urban locations 12
76 DQO output parameter for 1,3 butadiene at urban locations 12
77 DQO input parameter for 1,3 butadiene at rural locations 13
78 DQO output parameter for 1,3 butadiene at rural locations 13
7 9 DQO input parameter for arsenic at urban locations 14
710 DQO output parameter for arsenic at urban locations 14
711 DQO input parameter for arsenic at rural locations 15
712 DQO output parameter for arsenic at rural locations 15
7.13 DQO input parameter for chromium 16
7.14 DQO output parameter for chromium 17
7.15 DQO input parameter for acrolein 17
716 DQO output parameter for acrolein 17
717 DQO input parameter for formaldehyde at urban locations 18
718 DQO output parameter for formaldehyde at urban locations 18
719 DQO input parameter for formaldehyde at rural locations 19
720 DQO output parameter for formaldehyde at rural locations 19
8.1 TCAPCD Training Requirements 1
8 2 Core Ambient Air Training Courses 3
9 i Air Toxics Reporting Package Information 2
101 Schedule of Air Toxics Sampling Related Activities 1
102 List of Collocated Samplers and Coordinates 2
HI Sample Set-up, Run and Recovery Dates 2
112 Supplies at Storage Shelters 4
113 Field Corrective Action 8
114 Temperature Requirements 9
11.5 Holding Times 2
131 Instruments used in the Toxa City Laboratory 5
141 Precision Acceptance Criteria 2
151 Inspections in the Laboratory 4
152 Preventive Maintenance in Weight Room 4
153 Preventive Maintenance in VOC Laboratory 5
154 Preventive Maintenance in Liquid Chromatography Laboratory 5
15.5 Preventive Maintenance in Inductively Coupled Plasma Laboratory 6
156 Preventive Maintenance on Field Instruments ~
161 Lab Instrument Standards
162 Standard Materials and/or Apparatus for Air Toxics Calibrations
171 Critical Field Supplies and Consumables
172 Critical Laboratory Supplies and Consumables 2
191 Validation Check Summaries 4
192 Data Transfer Operations 5
193 Data Reporting Schedule 6
194 Reporting Equations 8
195 Data Archive Policies 9
20.1 Assessment Summary 8
23 i Single Flag Invalidation Criteria for Single Sampler 2
-------
Figures
Number Page
4.1 Organizational Structure of TCAPCD 5
7.1 Power Curve for Benzene urban location 9
7.2 Power Curve for Benzene rural location 11
7.3 Power Curve for 1,3, butadiene urban location 11
7.4 Power Curve for 1,3 butadiene rural location 13
7.5 Power Curve for arsenic urban location 14
7.6 Power Curve for arsenic rural location 16
7.7 Power Curve for chromium 17
7.8 Power Curve for acrolein 18
7.9 Power Curve for formaldehyde urban location 19
7.10 Power Curve for formaldehyde rural location 20
10.1 Network Locations 4
19.1 Data Management and Sample Flow Diagrams 2
VI
-------
1.0 QA Project Plan Identification and Approval
Title: Toxa City Air Pollution Control District Quality Assurance Project Plan for the air toxics
ambient air monitoring program.
The attached QAPP for the ATMP is hereby recommended for approval and commits the
Department to follow the elements described within.
Toxa City Air Pollution Control District
1) Signature: Date:
Dr. Melvin Thomas - Air Pollution Control Officer
2) Signature: Date:
Russell Kuntz - QA Division Director
EPA Region 11
1) Signature: Date:
Dennis Mickelson-Technical Project Officer - Air Monitoring Branch
2) Signature: Date:
Benjamin T. Zachary - QA Officer - QA Branch
Vll
-------
2.0 Table of Contents
Section
Foreword
Acknowledgments
Acronyms and Abbreviations
Tables
Figures
Region Approval
A. PROJECT MANAGEMENT
1. Title and Approval Page
2. Table of Contents
3. Distribution List
4. Project/Task Organization
4.1 Roles and Responsibilities
5. Problem Definition/Background
5.1 Problem Statement and Background
5.2 List of Pollutants
5.3 Location of Interest for HAPs
6. Project/Task Description
6.1 Description of Work
6.2 Field Activities
6.3 Laboratory Activities
6.4 Project Assessment Techniques
6.5 Schedule of Activities
Page Revision Date
n
iii
iv
v
vi
vii
vn
viii
xi
1/10
1/4
2/4
3/4
1/6
2/6
3/6
4/6
5/6
11/04/02
11/04/02
11/04/02
11/04/02
11/04/02
11/04/02
11/04/02
11/04/02
11/04/02
11/04/02
11/04/02
11/04/02
11/04/02
7. Data Quality Objectives
7.1 The General DQO Process
7.2 State the Problem
7.3 Identify the Decision
7.4 Identify the Inputs to the Decision
7.5 Define the Study Boundaries
7.6 Develop the Decision Rule
7.7 Specify Tolerable Limits on the Decision Error
7.8 Optimize the Design
7.9 DQOs for the Six Study Compounds
7.10 DQOs for Measuring the Percent Decreases
8. Special Training Requirements/Certification
8.1 Training
8.2 Certification
11/04/02
1/19
2/19
3/19
4/19
5/19
5/19
5/19
6/19
6/19
6/19
9/19
1/4
4/4
11/04/02
vin
-------
Section Page
9. Documentation and Records
9.1 Information Included in the Reporting Package 1/5
9.2 Data Reporting Package and Documentation Control 4/5
9.3 Data Reporting Package Archiving and Retrieval 5/5
B. MEASUREMENT/ DATA ACQUISITION
10. Sampling Design
10.1 Scheduled Project Activities, Including Measurement 1/6
Activities
10.2 Rationale for the Design 1/6
10.3 Design Assumptions 2/6
10.4 Procedure for Locating and Selecting
Environmental Samples 4/6
10.5 Classification of Measurements as Critical/Noncritical 5/5
10.6 Validation of Any Non-Standard Measurements
5/6
11. Sampling Methods Requirements
11.1 Purpose/Background 1/9
11.2 Sample Collection and Preparation 1/9
11.3 Support Facilities for Sampling Methods 2/9
11.4 Sampling/Measurement System Corrective Action 3/9
11.5 Sampling Equipment, Preservation, and Holding Time 7/9
Requirements
12. Sample Custody
12.1 Sample Custody Procedure 1/3
13. Analytical Methods Requirements
13.1 Purpose/Background 1/4
13.2 Preparation of Samples 1/4
13.3 Analysis Methods 1/4
13.4 Internal QC and Corrective Action for Measurement System 2/4
13.5 Sample Contamination Prevention, Preservation and Holding 3/4
14. Quality Control Requirements
14.1QC Procedures 1/8
Revision Date
I 11/04/02
12/3/02
11/04/02
1 11/04/02
1 11/04/02
11/04/02
15. Instrument/Equipment Testing, Inspection, and Maintenance
Requirements
15.1 Purpose/Background
15.2 Testing
15.3 Inspection
15.4 Maintenance
16. Instrument Calibration and Frequency
16.1 Instrumentation Requiring Calibration
16.2 Calibration Methods
16.3 Calibration Standard Materials and Apparatus
16.4 Calibration Frequency
11/04/02
1/7
1/7
1/7
3/7
1/8
3/8
5/8
8/8
11/04/02
IX
-------
Section Page Revision Date
17. Inspection/Acceptance for Supplies and Consumables 1 11/04/02
17.1 Purpose 1/3
17.2 Critical Supplies and Consumables 1/3
17.3 Acceptance Criteria 3/3
17.4 Tracking and Quality Verification of Supplies and 3/3
Consumables
18. Data Acquisition Requirements (non-direct measurements) 1 11/04/02
18.1 Acquisition of Non-Direct Measurement Data 1/3
19. Data Management 1 11/04/02
19.1 Background and Overview 1/10
19.2 Data Recording 3/10
19.3 Data Validation 3/10
19.4 Data Transformation 5/10
19.5 Data Transmittal 5/10
19.6 Data Reduction 6/10
19.7 Data Summary 7/10
19.8 Data Tracking 8/10
19.9 Data Storage and Retrieval 9110
9/10
C. ASSESSMENT/OVERSIGHT
20. Assessments and Response Actions
20.1 Assessment Activities and Project Planning 1/8 1 11/04/02
20.2 External Assessment Schedule 8/8
21. Reports to Management 1 11/04/02
21.1 Frequency, Content, and Distribution of Reports 1/2
22. Data Review
22.1 Data Review Design 1/4
22.2 Data Review Testing 2/4 1 11/04/02
22.3 Procedures 3/4
D. VALIDATION AND USABILITY 1 11/04/02
23. Validation, Verification and Analysis Methods 1/3
23.1 Process for Validating and Verifying Data 3/3
23.2 Data Analysis
24. Reconciliation with Data Quality Objectives 2 11/04/02
24.1 Reconciling Results with DQOs 1/5
24.2 Five Steps of the DQA Process 1/5
Appendices
A. Technical Assistance Document For the National Ambient Air Toxics
Trends and Assessment Program
-------
3.0 Distribution
A hardcopy of this QAPP has been distributed to the individuals in Table 3-1. The document
is also available on the Internet at http://www.toxacitv.aped.gov.
Table 3.1 Distribution List
Name
Position
Division/Branch
Toxa City Air Pollution Control District
Dr. Melvin Thomas
Russell Kuntz
John Holstine
Thomas Sutherland
Daniel Willis
Holly J. Webster
James Courtney
Robert Kirk
Joe L. Craig
Kent Field
Alexander Barnett
Janet Hoppert
David Bush
Gary Arcemont
Lisa Killion
Robert Renelle
Mark Fredrickson
Air Pollution Control Officer
QA Division Director
QA Officer
QA Technician
Air Division Director
Ambient Air Monitoring Branch Chief
Field Technician
Field Technician
Field Technician
Data Manager
Program Support Division Director
Shipping/Receiving Branch Chief
Clerk
Laboratory Branch Chief
Lab Technician
Lab Technician
Lab Technician
TCAPCD
QA Division
QA Division
QA Division
Air Division
Technical/ Ambient Air Monitoring
Technical/ Ambient Air Monitoring
Technical/ Ambient Air Monitoring
Technical/ Ambient Air Monitoring
Technical/ Ambient Air Monitoring
Program Support
Program Support/Shipping &Rec.
Program Support/Shipping &Rec.
Technical/Laboratory
Technical/Laboratory
Technical/Laboratory
Technical/Laboratory
EPA Region 11
Dennis Mickelson
Benjamin T. Zachary
QA Officer
EPA Project Officer
Air/ Air Quality Monitoring
Air/Quality Assurance
XI
-------
Project: Model QAPP
Element No: 4
Re vision No: 1.0
Date: 11/04/02
Page 1 of 10
4.0 Project/Task Organization
4.1 Roles and Responsibilities
Federal, State, Tribal and local agencies all have important roles in developing and
implementing satisfactory air monitoring programs. As part of the planning effort, EPA is
responsible for developing National Ambient Air Quality Standards (NAAQS), and
identifying a minimum set of QC samples from which to judge data quality. The State and
local organizations are responsible for taking this information and developing and
implementing a quality system that will meet the data quality requirements. Then, it is the
responsibility of both EPA and the State and local organizations to assess the quality of the
data and take corrective action when appropriate. The responsibilities of each organization
follow.
4.1.1 Office of Air Quality Planning and Standards
OAQPS is the organization charged under the authority of the Clean Air Act (CAA) to protect
and enhance the quality of the nation's air resources. OAQPS sets standards for pollutants
considered harmful to public health or welfare and, in cooperation with EPA's Regional
Offices and the States, enforces compliance with the standards through state implementation
plans (SIPs) and regulations controlling emissions from stationary sources. OAQPS evaluates
the need to regulate potential air pollutants, especially air toxics and develops national
standards; works with State, Local and Tribal (S/L/T) agencies to develop plans for meeting
these standards. In addition, OAQPS provide the funding, through Section §103 and §105
funds.
Within the OAQPS Emissions Monitoring and Analysis Division (EMAD), the Monitoring
and Quality Assurance Group (MQAG) is responsible for the oversight of the NATTS.
MQAG has the following responsibilities:
•
• ensuring that the methods and procedures used in making air pollution measurements are
adequate to meet the programs objectives and that the resulting data are of satisfactory
quality
• evaluating the performance, through Technical Systems Audits (TSAs) and Management
System Reviews (MSRs), of organizations making air pollution measurements;
• implementing satisfactory quality assurance programs over EPA's Ambient Air Quality
Monitoring Network;
• ensuring that national regional laboratories are available to support toxics and QA
programs;
-------
Project: Model QAPP
Element No: 4
Re vision No: 1.0
Date: 11/04/02
Page 2 of 10
• rendering technical assistance to the EPA Regional Offices and air pollution monitoring
community.
4.1.2 EPA Region 11 Office
The EPA Regional Offices will address environmental issues related to the States within their
jurisdiction and to administer and oversee regulatory and congressionally mandated programs.
The major quality assurance responsibilities of EPA's Regional Offices, in regards to the
NATTS, are the coordination of quality assurance matters at the Regional levels with the State
and local agencies. This is accomplished by the designation of EPA Regional Project Officers
who are responsible for the technical aspects of the program including:
• reviewing QAPPs by Regional QA Officers who are delegated the authority by the
Regional Administrator to review and approve QAPPs for the Agency;
• supporting the air toxics audit evaluation program;
• evaluating quality system performance, through TSAs and network reviews whose
frequency is addressed in the Code of Federal Regulations;
acting as a liaison by making available the technical and quality assurance information
developed by EPA Headquarters and the Region to the State and local agencies, and
making EPA Headquarters aware of the unmet quality assurance needs of the State and
local agencies.
Toxa City will direct all technical and QA questions to Region 11.
4.1.3 Toxa City Air Pollution Control District
40 CFR Part 58 defines a State Agency as "the air pollution control agency primarily
responsible for the development and implementation of a plan under the Act (CAA)".
Section 302 of the CAA provides a more detailed description of the air pollution control
agency.
40 CFR Part 58 defines the Local Agency as "any local government agency, other than the
state agency, which is charged with the responsibility for carrying out a portion of the plan
(SIP)".
The major responsibility of State and local agencies is the implementation of a satisfactory
monitoring program, which would naturally include the implementation of an appropriate
quality assurance program. It is the responsibility of State and local agencies to implement
quality assurance programs in all phases of the environmental data operation (EDO),
including the field, their own laboratories, and in any consulting and contractor laboratories
-------
Project: Model QAPP
Element No: 4
Re vision No: 1.0
Date: 11/04/02
Page 3 of 10
which they may use to obtain data. An EDO is defined as work performed to obtain, use, or
report information pertaining to environmental processes or conditions.
Figure 4.1 represents the organizational structure of the areas of the Toxa City Air Pollution
Control District (TCAPCD or the District) that are responsible for the activities of the air
toxics ambient air quality monitoring program. The following information lists the specific
responsibilities of each individual and is grouped by functions of the Directors Office, and the
divisions related to Quality Assurance, Technical Support, and Program Support.
4.1.3.1 Directors Office
Air Pollution Control Director - Dr. Melvin Thomas
The Director has overall responsibility for managing the Toxa City Air Pollution Control
District according to policy. The direct responsibility for assuring data quality rests with
management. Ultimately, the Director is responsible for establishing QA policy and for
resolving QA issues identified through the QA program. Major QA related responsibilities of
the Director include:
• approving the budget and planning processes;
• assuring that the District develops and maintains a current and germane quality system;
• assuring that the District develops and maintains a current air toxics QAPP and ensures
adherence to the document by staff, and where appropriate, other extramural
cooperators;
• establishing policies to ensure that QA requirements are incorporated in all environmental
data operations;
• maintaining an active line of communication with the QA and technical managers;
• conducting management systems reviews.
The Director delegates the responsibility of QA development and implementation in
accordance with District policy to the Division Directors. Oversight of the District's QA
program is delegated to the QA Division Director.
-------
Project: Model QAPP
Element No: 4
Re vision No: 1.0
Date: 11/04/02
Page 4 of 10
4.1.3.2 QA Division
QA Division Director (QAD) - Russell Kuntz
The QA Division Director is the delegated manager of the District's QA Program. He has
direct access to the Director on all matters pertaining to quality assurance. The main
responsibility of the QAD is QA oversight, and ensuring that all personnel understand the
District's QA policy and all pertinent EPA QA policies and regulations specific to the
Ambient Air Quality Monitoring Program. The QAD provides technical support and reviews
and approves QA products. Responsibilities include:
• developing and interpreting District QA policy and revising it as necessary;
• developing a QA Annual Report for the Director;
• reviewing acquisition packages (contracts, grants, cooperative agreements, inter-agency
agreements) to determine the necessary QA requirements;
• developing QA budgets;
• assisting staff scientists and project managers in developing QA documentation and in
providing answers to technical questions;
• ensuring that all personnel involved in environmental data operations have access to any
training or QA information needed to be knowledgeable in QA requirements, protocols,
and technology of that activity;
• reviewing and approving the QAPP for the ATMP;
• ensuring that environmental data operations are covered by appropriate QA planning
documentation (e.g., QA project plans and data quality objectives);
• ensuring that Management System Reviews (MSRs), assessments and audits are scheduled
and completed, and at times, conducting or participating in these QA activities;
• tracking the QA/QC status of all programs;
• recommending required management-level corrective actions;
• serving as the program's QA liaison with EPA Regional QA Managers or QA Officers
and the Regional Project Officer.
The QAD has the authority to carry out these responsibilities and to bring to the attention of
the Director any issues associated with these responsibilities. The QAD delegates the
responsibility
-------
Project: Model QAPP
Element No: 4
Re vision No: 1.0
Date: 11/04/02
Page 5 of 10
EPA QA Officer
Dennis Mickleson
872 669-2299
EPA Project Officer
Benjamin T. Zachary
872 669-2378
Figure 4.1 Organization Chart for the TCAPCD
Quality Assurance Officer - John Holstine
The QA Officer is a main point of contact within the QA Division. The QA Officer's
responsibilities include:
• implementing and overseeing the District's QA policy within the division;
• acting as a conduit for QA information to division staff;
• assisting the QAD in developing QA policies and procedures;
• coordinating the input to the QA Annual Report (QAAR);
• assisting in solving QA-related problems at the lowest possible organizational level.
• ensuring that an updated QAPP is in place for all environmental data operations associated
with the ATMP;
• ensuring that technical systems audits, audits of data quality, and data quality; assessments
occur within the appropriate schedule and conducting or participating in these audits.
• tracking and ensuring the timely implementation of corrective actions;
• ensuring that a management system review occurs every 3 years;
• ensuring that technical personnel follow the QAPP
• review precision in the data;
-------
Project: Model QAPP
Element No: 4
Re vision No: 1.0
Date: 11/04/02
Page 6 of 10
• data validation;
• ensuring that all environmental data activities effectively follow the QA/QC requirements.
The QA officer has the authority to carry out these responsibilities and to bring to the
attention of his or her respective Division Director any issues related to these responsibilities.
The QA officer delegates the responsibility of QA development and implementation in
accordance with District policy.
Quality Assurance Technician - Thomas Sutherland
The QA technician is the staff QA contact appointed by the QA officer. Tom Sutherland is
the person who performs all field and laboratory audits. Mr. Sutherland's responsibilities
include:
remaining current on District QA policy and general and specific EPA QA policies and
regulations as it relates to the ATMP;
• scheduling and implementing technical systems audits;
• performing data quality assessments;
reviewing precision data;
providing QA training to Air and Program Support Division technical staff;
• ensuring timely follow-up and corrective actions resulting from auditing and evaluation
activities;
• facilitating management systems reviews implemented by the QA Officer.
4.1.3.3 Technical Division
The technical divisions are responsible for all routine environmental data operations (EDOs)
for the ATMP.
Air Division Director - Daniel Willis
The Air Division Director is the delegated manager of the routine ATMP which includes the
QA/QC activities that are implemented as part of normal data collection activities.
Responsibilities of the Director include:
communication with EPA Project Officers and EPA QA personnel on issues related to routine
sampling and QA activities;
understanding EPA monitoring and QA regulations and guidance, and ensuring subordinates
understand and follow these regulations and guidance;
understanding District QA policy and ensuring subordinates understand and follow the policy;
understanding and ensuring adherence to the QAPP;
-------
Project: Model QAPP
Element No: 4
Re vision No: 1.0
Date: 11/04/02
Page 7 of 10
reviewing 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 in QA requirements, protocols, and
technology;
recommending required management-level corrective actions.
The Air Director delegates the responsibility for the development and implementation of
individual monitoring programs, in accordance with District policy, to the Air Division
Branch Managers.
Air Monitoring Branch Manager - Holly J. Webster and Laboratory Branch Manager -
Gary Arcemont
These two branches are responsible for overseeing the routine field/lab monitoring and QA
activities of the Ambient Air Quality Monitoring Program. The Branch Manager's
responsibilities include:
• implementing and overseeing the District's QA policy within the branch;
• acting as a conduit for information to branch staff;
• training staff in the requirements of the QA project plan and in the evaluation of QC
measurements;
assisting staff scientists and project managers in developing network designs, field/lab
standard operating procedures and appropriate field/lab QA documentation;
• ensuring that an updated QAPP is in place for all environmental data operations
associated with the ATMP;
• ensuring that technical personnel follow the QAPP;
• assure that the laboratory and field staff adhere to the QA/QC requirements of the
specified analytical methods and Standard Operating Procedures (SOPs);
assure that the laboratory and field programs generate data of known and needed quality
to meet the programs Data Quality Objectives (DQOs);
• review and approve of modifications on the SOPs for the field and laboratory programs.
In addition, review and approval any new SOPs with the integration of new instruments.
Field Personnel - James Courtney, Robert Kirk, and Joe L. Craig
The field personnel are responsible for carrying out a required task(s) and ensuring the data
quality results of the task(s) by adhering to guidance and protocol specified by the QAPP and
SOPs for the field activities. Responsibilities include:
participating in the development and implementation of the QAPP;
participating in training and certification activities;
-------
Project: Model QAPP
Element No: 4
Re vision No: 1.0
Date: 11/04/02
Page 8 of 10
writing and modifying SOPs;
verifying that all required QA activities are performed and that measurement quality standards
are met as required in the QAPP;
performing and documenting preventative maintenance;
documenting deviations from established procedures and methods;
reporting all problems and corrective actions to the Branch Managers;
assessing and reporting data quality;
preparing and delivering reports to the Branch Manager;
flagging suspect data;
handling/transport of cartridges, filters, and other sampling needs in and out of the field;
maintain chain-of-custody records in the field;
calibration of samplers as specified by the QAPP and SOPs;
loading/unloading of samples;
packing, shipping or transporting the exposed samples in accordance with the SOPs and
QAPP;
maintain logbooks of the QA/QC activities and equipment preventive maintenance logs.
Laboratory Personnel - Lisa Killion, Robert Renelle, Mark Fredrickson
Laboratory personnel are responsible for carrying out a required task(s) and ensuring the data
quality results of the task(s) by adhering to guidance and protocol specified by the air toxics
QAPP and SOPs for the lab activities. Their responsibilities include:
participating in the development and implementation of the QAPP;
participating in training and certification activities;
participating in the development of data quality requirements (overall and laboratory) with the
appropriate QA staff;
writing and modifying SOPs and good laboratory practices (GLPs);
verifying that all required QA activities were performed and that measurement quality
standards were met as required in the QAPP;
following all manufacturers' specifications;
performing and documenting preventative maintenance;
documenting deviations from established procedures and methods;
reporting all problems and corrective actions to the Branch Manager;
assessing and reporting data quality;
preparing and delivering reports to the Branch Manager;
flagging suspect data;
-------
Project: Model QAPP
Element No: 4
Re vision No: 1.0
Date: 11/04/02
Page 9 of 10
In addition, the laboratory personnel will perform the following duties:
sample receiving and inspection from vendor;
• Di-nitro-phenyl-hydrazine (DNPH) cartridge preparation;
• preparing the chain-of-custody forms for field use;
• post-sampling receiving of samples and processing of samples (i.e., refrigeration of DNPH
cartridges);
sample preparation, extraction, and clean-up;
• Analysis of the VOC, metals and aldehydes according to accepted SOPs.
Information Manager- Kent Field
The Information Manager is responsible for coordinating the information management
activities of the ATMP. The main responsibilities of the Information Manager include
ensuring that data and information collected for the ATMP are properly captured, stored, and
transmitted for use by program participants. Responsibilities include:
developing local data management standard operating procedures;
ensuring that information management activities are developed within reasonable time frames
for review and approval;
maintenance and upkeep of the Laboratory Information Management System (LIMS);
storage of raw data for the analysis data, i.e., chromatograms from the various laboratory
instrumentation;
long term storage of data to Compact Disk (CD) or other digital storage media;
upkeep of LIMS software and upgrading when needed;
ensuring the adherence to the QAPP where applicable;
ensuring access to data for timely reporting and interpretation processes;
ensuring the development of data base guides (data base structures, user guidance documents);
ensuring timely delivery of all required data to the AIRS system.
4.1.3.4 Program Support
The Program Support Division include the areas of human resources, facilities maintenance,
and shipping and receiving.
Program Support Division Director - Alexander Barnett
Responsibilities of the Director include:
communication with QA and Air Monitoring Division on specific needs;
understanding EPA monitoring and QA regulations and guidance, and ensuring subordinates
understand and follow these regulations and guidance;
-------
Project: Model QAPP
Element No: 4
Re vision No: 1.0
Date: 11/04/02
Page 10 of 10
understanding District QA policy and ensuring subordinates understand and follow the policy;
understanding and ensuring adherence to the QAPP as it relates to program support activities;
ensuring that all support personnel have access to any training or QA information needed to
be knowledgeable in QA requirements, protocols, and technology.
Shipping/Receiving Branch Manager - Janet Hoppert
This branch is responsible for shipping and receiving equipment, supplies and consumables
for the routine field/lab monitoring and QA activities of the ATMP. The Branch Managers
responsibilities include:
implementing and overseeing the District's QA policy within the branch
acting as a conduit for information to branch staff;
training staff in the requirements of the QA project plan as it relates to shipping/receiving;
assisting staff in developing standard operating procedures;
coordinating the Branch's input to the Quality Assurance Annual Report
ensuring that technical personnel follow the QAPP;
reviewing and evaluating staff performance and conformance to the QAPP.
Clerk -David Bush
Mr. Bush supports for all shipping/receiving of all equipment and consumable supplies for the
ATMP. Responsibilities include:
assisting in the development of standard operating procedures for shipping/receiving;
following SOPs for receiving, storage, chain-of-custody and transfer of filters, canisters and
cartridges;
informing appropriate field /lab staff of arrival of consumables, equipment, and samples;
documenting, tracking, and archiving shipping/receiving records.
-------
Project: Air Toxics Model QAPP
Element No: 5
Re vision No: 2.1
Date: 11/04/02
Page 1 of 4
5.0 Problem Definition/Background
5.1 Problem Statement and Background
5.1.1 Background
There are currently 188 hazardous air pollutants (HAPs), or air toxics, regulated under the
Clean Air Act (CAA) that have been associated with a wide variety of adverse health effects,
including cancer, neurological effects, reproductive and developmental effects, as well as
ecosystem effects. These air toxics are emitted from multiple sources, including major
stationary, area, and mobile sources, resulting in population exposure to these air toxics as
they occur in the environment. While in some cases the public may be exposed to an
individual HAP, more typically people experience exposures to multiple HAPs and from
many sources. Exposures of concern result not only from the inhalation of these HAPs, but
also, for some HAPs, from multi-pathway exposures to air emissions.
5.1.2 The National Air Toxics Trends Stations and the Role of the TCAPCD
EPA finalized the Urban Air Toxics Strategy (UATS) in the Federal Register on July 19,
19991. The UATS states that emissions data are needed to quantify the sources of air toxics
impacts and aid in the development of control strategies, while ambient data are then needed
to understand the behavior of air toxics in the atmosphere after they are emitted. Since
ambient measurements cannot practically be made everywhere, modeled estimates are needed
to extrapolate our knowledge of air toxics impacts into locations without monitors. Exposure
assessments, together with health effects information, are then needed to integrate all of these
data into an understanding of the implications of air toxics impacts and to characterize air
toxics risks. The EPA proposed the National Air Toxics Assessment (NATA). There are four
activities which are key to the success of the NATA.
Source-specific standards and sector-based standards, including section 112 standards, i.e.
Maximum Achievable Control Technology (MACT), Generally Achievable Control
Technology (GACT), residual risk standards, and section 129 standards. - National,
regional, and community-based initiatives to focus on multi-media and;
Cumulative risks, such as the Integrated UATS, Great Waters, Mercury initiatives,
Persistent Bio-accumulative Toxics (PBT) and Total Maximum Daily Load (TMDL)
initiatives, and Clean Air Partnerships.
• NATA activities that will help EPA identify areas of concern, characterize risks and track
progress. These activities include expanded air toxics monitoring, improving and
periodically updating emissions inventories, national- and local scale air quality and
exposure modeling, and continued research on effects and assessment tools, leading to
-------
Project: Air Toxics Model QAPP
Element No: 5
Re vision No: 2.1
Date: 11/04/02
Page 2 of 4
improved characterizations of air toxics risk and reductions in risk resulting from ongoing
and future implementation of air toxics emissions control standards and initiatives.
• Education and outreach.
The success of the NATA critically depends on our ability to quantify the impacts of air
toxics emissions on public health and the environment. All of these activities are aimed at
providing the best technical information regarding air toxics emissions, ambient
concentrations, and health impacts to support the development of sound policies for NATA.
Specifically, these activities include:
• The measurement of air toxics emission rates from individual pollution sources;
the compilation of comprehensive air toxics emission inventories for local, State, and
national domains;
• the analysis of patterns and trends in ambient air toxics measurements;
• the estimation of ambient air toxics concentrations from emission inventories using
dispersion modeling;
the estimation of human and environmental exposures to air toxics, and;
• the assessment of risks due to air toxics:
• and the measurement of ambient concentrations of air toxics at trends monitoring
sites throughout the nation.
Analysis was performed by OAQPS to ascertain the size and features of a national trends
network that would satisfy the goals as stated above. This analysis illustrated that a number
urban and rural locations would provide the needed coverage for estimates of national trends.
TCAPCD was contacted by the EPA to support one of these national trends sites. TCAPCD
has agreed to provide the support to this network. TCAPCD will support one monitoring
station as agreed through the Section §103 and §105 funds received from the Regional Office.
This QAPP focuses on the Quality Assurance (QA) and Quality Control (QC) that will be
instituted by the TCAPCD to fulfill it's obligation to the NATTS. In order to better focus the
data collection activities on the final use of the data, a DQO process was performed in Chapter
7 of this QAPP.
5.2 List of Pollutants
There are 33 HAPs identified in the draft Integrated Urban Air Toxics Strategy (UATS). They
are a subset of the 188 toxics identified in Section 112 of the CAA which are thought to have
the greatest impact on the public and the environment in urban areas. The TCAPCD staff
reviewed the 33 HAPs list and consulted with EPA and State Agency staff. After several
consultations, a final list of compounds were selected. The list is based on:
The EPA's Concept Paper2
-------
Project: Air Toxics Model QAPP
Element No: 5
Re vision No: 2.1
Date: 11/04/02
Page 3 of 4
• A major portion of the 33 Unified Air Toxics Strategy (UATS) HAPs can be measured
with 3 field and lab systems;
• The limitations of the State-of-the-Science instruments.
A number of compound on the UATS list are difficult to characterize or the methods have not
been developed yet. These compounds will not be included in the pollutant list. If at some
time in the future methods are developed for these compounds, the District may, at some point
include these compounds. See Table 5-1. However, EPA is only requiring the TCAPCD to
report to the national Air Quality System (AQS) six compound that are listed in the Required
Section of Table 5-1. Since the collection and analysis of samples will also provide data on
other compounds, the TCAPCD will report values to AQS that can be quality assured and
accepted by the procedures detailed in this QAPP.
Table 5.1 List of HAPs
Benzene, chromium, acrolein, and
formaldehyde
Required
Core
Benzene, 1,3-butadiene, carbon
tetrachloride, chloroform, 1,2-
dichloropropane,
dichloromethane,
tetrachloroethylene,
trichloroethylene, vinyl chloride,
arsenic, beryillium, cadmium,
chromium, lead, manganese,
formaldehyde and acrolein
Max
Acrylonitrile, benzene, 1,3-butadiene,
:arbon tetrachloride, chloroform, 1,2
dibromomethane, 1,3-dichloropropene,
1,2-dichloropropane, ethylene
dichloride, ethylene oxide,
dichloromethane, tetrachloro ethane,
tetrachloroethylene, trichloroethylene,
vinyl chloride, arsenic, beryillium,
cadmium, chromium, lead, mercury,
manganese, nickel, acetaldehyde,
formaldehyde and acrolein, 2,2,7,8
tetrachlorobenzo -p-dioxin, coke oven
emissions, hexachlorobenzene,
liydrazine, polycyclic organic matter,
polychloronated biphenyls, quinoline
As can be seen from Table 5-1, there are a number of additional HAPs in the Core and Max
lists. Many of these are HAPs that the current analytical systems can measure. The
TCAPCD will report the compounds that are on the Core and Max list if these can be detected
and analyzed while collecting the data on the required list.
-------
Project: Air Toxics Model QAPP
Element No: 5
Re vision No: 2.1
Date: 11/04/02
Page 4 of 4
5.3 Locations of Interest for HAPs
The main objective for the TCAPCD NATTS is to provide data for the national trends, as
determined in Chapter 7 of this QAPP. However, the TCAPCD will also be operating four
other air toxics monitoring stations. Further information on air toxics is needed for both
industrial/downtown and suburban areas. The major manufacturing and industrial areas are
also near the mouth of the bay. There are several neighborhoods that surround this area. The
TCAPCD has decided to target these several areas in addition to the NATTS. The other
locations are suburban-oriented sites needed to characterize general exposure and temporal
and spatial variability.
References
1. National Air Toxics Program: The Integrated Urban Strategy-Report to Congress, EPA Document No.
453/R-99-007, July 2000, URL Address: http://www.epa.gov/ttn/atw/urban/urbanpg.html
2. Air Toxics Monitoring Concept Paper, Draft, February 29, 2000, URL address:
http ://www. epa. gov/ttn/amtic/airtxfil.html
-------
Air Toxics Model QAPP
Element No: 6
Re vision No: 1.0
Date: 11/04/02
Page 1 of6
6.0 Project Task/Description
6.1 Description of Work
The measurement goal of the NATTS is to estimate the concentration, in units of micrograms
per cubic meter (ug/m3) and parts per billion/volume (ppbv) of air toxic compounds of
particulates and gases. This is accomplished by three separate collection media: canister
sampling with passivated canisters, DNPH cartridges and Paniculate Matter - 10 micron
(PMio) high volume sampling on an 8 x 10" quartz glass filter and diesel emissions through
continuous spectrophotometry.
The following sections will describe the measurements required for the routine field and
laboratory activities for the network.
6.2 Field Activities
Table 6.1, 6.2, 6.3 and 6.4 summarizes some of the more critical performance requirements.
Table 6.1 Design/Performance Specifications - Particulate Matter — 10 micron - Toxic Metals
Equipment
Filter Design Specs.
Size
Medium
Filter thickness
Collection efficiency
Stability Temperature
Sampler Performance
Specs.
Sample Flow Rate
Flow Regulation
Flow Rate Precision
Flow Rate Accuracy
Clock/Timer
Frequency
1 in 6 days
1 in 6 days
Acceptance Criteria
See Reference 1
8.5" x 11"
Quartz Glass Fiber Filter
0.50mm
99.95%
temperatures > 540°C
1.13 m/min.
0.1 m/min.
+7
±2
24 hour + 2 min accuracy
Reference
See Reference 1
"IO- 2.1 Sec 8.1
"Sec 3. 2
"Sec 6.2.4
"Sec 6.2.2
"Sec 6. 2. 2
"Sec 6.1. 4
See Reference 4, Appendix J
See Reference 5 Section 2. 1 1.7
See Reference 6, Section 2.11.2
-------
Air Toxics Model QAPP
Element No: 6
Re vision No: 1.0
Date: 11/04/02
Page 2 of6
Table 6.2 Design/Performance Specifications - Air Canister Sampler - Volatile Organic Compounds
Equipment
Canister Design Specs.
Size
Medium
Max Pressure
Max. pressure drop
Collection efficiency
Lower Detection Limit
Sampler Performance
Specs.
Sample Flow Rate
Flow Regulation
Flow Rate Precision
Flow Rate Accuracy
External Leakage
Internal Leakage
Frequency
1 in 6 days
1 in 6 days
Acceptance Criteria
See Reference 2
6 liters spherical
Passivated SUMMA electro -
polished Stainless Steel Canister
30 psig
14psig.
99%
compound specific, usually >0. 1
ppbv
1 80 cc/min.
1.0 cc/min.
+10%
+10%
Vendor specs
Vendor specs
24 hour + 2 min accuracy
Reference
See Reference 2
"Vender Spec.
"
"
"
See TO-14A
"Vender Spec.
See Reference 2
TO-14A
NA
NA
"Sec 6. 1.8
Table 6.3 Design/Performance Specifications - Carbonyl Sampler - Aldehyde and Ketone Compounds
Equipment
Filter Design Specs.
Size
Medium
Sampler Performance
Specs.
Sample Flow Rate
Flow Regulation
Flow Rate Precision
Flow Rate Accuracy
External Leakage
Internal Leakage
Clock/Timer
Frequency
1 in 6 days
1 in 6 days
Acceptance Criteria
See Reference 4
100 mm Cylindrical Silica Gel
cartridge
coated with
2,4-Dinitro-phenyl hydrazine
0.20 m/min.
0.2 m /min.
+10%
+10%
Vendor specs
Vendor specs
24 hour + 2 min accuracy
Reference
See Reference 3
"TO-llASec 7.1
"Vender Spec.
"
NA
NA
"Vender Spec.
-------
Air Toxics Model QAPP
Element No: 6
Re vision No: 1.0
Date: 11/04/02
Page 3 of6
Table 6.4 Design/Performance Specifications — Aethalometer — Black and Organic Carbon
Equipment
Sampler Performance Specs.
Sample Flow Rate
Size
medium
Data Recording
Power
Temperature Range
Wavelength
Time Resolution
Sensitivity
Frequency
Continuous
Continuous
Acceptance Criteria
2 to 6 1/min.
19" by 10.5"
Quartz Tape
0-5 VDC
60watts/110VAC
0-40deg. C
880 nm and 370 nm
Ihour or 1 minute
0. 1 ug/m3
Reference
"Vender Spec.
The District assumes the sampling instruments to be adequate for the sampling for air
toxics. All of the instruments operated in the field are vendor supplied. The descriptions
of the samplers are similar to the instruments described in the references noted above.
6.2.1 Field Measurements
Table 6.1, 6.2, 6.3 and 6.4 represents the field measurements that must be collected.
These tables presented in the Compendia of Organic and Inorganic Methods listed in
References 1-3. At the urging of the EPA, the TCAPCD will also measure Elemental
Carbon (EC) and Organic Carbon (OC) using the Aethalometer. This is a continuous
instrument that draws samples through quartz tape. The OC/EC particle are trapped on
the tape and analyzed via spectrophotometry at 880 and 370 nm. The data are stored on
the internal 3.5 inch drive and can be retrieved during site visits. All other instruments
collect discreet and are stored in the instrument for downloading by the field operator
during routine visits.
6.3 Laboratory Activities
Laboratory activities for the air toxics program include preparing the filters, canisters and
cartridges for the routine field operator, which includes three general phases:
Pre-Sampling
• Receiving filters, canisters or cartridges from the vendors;
• Checking sample integrity;
• Conditioning filters, storing canisters and cartridges;
• Weighing filters;
Storing prior to field use;
Packaging filters, canisters and cartridges for field use;
Associated QA/QC activities;
Maintaining microbalance and analytical equipment at specified environmental
conditions;
-------
Air Toxics Model QAPP
Element No: 6
Re vision No: 1.0
Date: 11/04/02
Page 4 of6
Equipment maintenance and calibrations.
Shipping/Receiving
• Receiving filters, canisters and cartridges from the field and logging into
database;
Storing filters, canisters and cartridges;
Associated QA/QC activities.
Post-Sampling
• Checking filter, cartridge and canister integrity;
• Stabilizing/weighing filters;
extraction of VOCs from canisters;
extraction of metals from quartz filter using hot acid/microwave extraction;
• extraction of DNPH compounds;
• Analysis of samples extracted;
• Data downloads from field samplers;
Data entry/upload to AIRS;
Storing filters/archiving;
Cleaning canisters;
Associated QA/QC activities.
The details for these activities are included in various sections of this document as well as
References 1-3.
6.4 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: performance evaluation (PE), MSRs, TSAs, peer review,
inspection, or surveillance. Section 20 will discuss the details of the District's
assessments.
Table 6.5 will provide information on the parties implementing the assessment and their
frequency.
-------
Air Toxics Model QAPP
Element No: 6
Re vision No: 1.0
Date: 11/04/02
Page 5 of6
Table 6.5 Assessment Schedule
Agency
NAREL
ERG
OAQPS-EMAD
Regional Offices
Regional Offices
Type of Assessment
TSA and PEs, round robin
inter-laboratory samples
PEs
MSRs, TSAs
Network Reviews
TSAs and IP As
Agency Assessed
ERG
S/L/T agencies
ERG, NAREL,
EPA Regional and
S/L/T agencies
S/L/T agencies
S/L/T agencies
Frequency
Annually
Annually
As needed by
EMAD
determination
Once every 5
years
Annually *
6.5 Schedule of Activities
Table 6.6 contains a list of the critical activities required to plan, implement, and assess
the air toxics program.
Table 6.6 Schedule of Critical Air Toxics Activities
Activity
Network development (EPA)
Sampler order
Laboratory upgrade
Laboratory procurement
QAPP development
Samplers arrive
Sampler siting/testing
QAPP Submittal
Field/Laboratory Training
QAPP Approval
Routine Sampling Begins
Due Date
June 2002
August 2002
August 2002
September 2002
November 2002
November 2002
November 2002
December 2002
December 2002
January 2003
January 1,2003
Comments
Preliminary list of sites and samplers required
Samplers ordered from National contract
Listing of laboratory requirements
Ordering/purchase of all laboratory and miscellaneous field
equipment
Development of the QAPP
Received in Shipping and Receiving District
Establishment of sites and preliminary testing of
samplers
QAPP submittal to EPA
Field and laboratory training activities and certification.
Approval by EPA
Routine activities must start
-------
Air Toxics Model QAPP
Element No: 6
Re vision No: 1.0
Date: 11/04/02
Page 6 of6
References
1. Compendium Method for the Determination of Inorganic Compounds in Air, United States
Environmental Protection Agency, June 1999, Section IO-2.1.
2. Compendium Method for the Determination of Toxic Organic Compounds in Air, United States
Environmental Protection Agency, Section TO-11A, January 1999
3. Compendium Method for the Determination of Toxic Organic Communes in Air, United States
Environmental Protection Agency, Section TO-15, January 1999
4. Code of Federal Regulations, Chapter 40, Part 50, Appendix J, Section 4.1
5. Quality Assurance Handbook for Air Pollution Measurement Systems Volume II, Section 2.11.7, April
1989
6. Quality Assurance Handbook for Air Pollution Measurement Systems Volume II, Section 2.11.2, April
1989
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Pagel of 19
7.0 Data Quality Objectives
Since the TCAPCD has decided to participate in the NATTS, staff realized that the
objectives of the non-trend stations would be different from the NATTS. Since EPA-
OAQPS has developed the DQOs for the trends sites, the TCAPCD has decided to state
the objectives for the non-trends sites. Below is the non-trends objective statement.
Determine the highest concentrations expected to occur in the area covered by the
network, i.e., to verify the spatial and temporal characteristics of HAPs within the
city.
Please note that these are the identical monitoring objective to those of the pilot study,
which were developed in the TCAPCD Pilot City QAPP. However, the NATTS
objectives are different. Therefore, the DQOs developed for the NATTS are described
below.
The DQO process described in EPA's QA/G-4 document provides a general framework
for ensuring that the data collected by EPA meets the needs of the intended decision
makers and data users. The process establishes the link between the specific end use(s)
of the data with the data collection process and the data quality (and quantity) needed to
meet a program's goals. This process was applied to one of the primary goals of the
National Air Toxics Trends Network, namely to establish trends and evaluate the
effectiveness of HAP reduction strategies. This section describes the results of the DQO
process for the local monitoring data requirements for: benzene, 1,3-butadiene, arsenic,
chromium, acrolein, and formaldehyde. In addition, the objectives for the rest of the
network should also be stated clearly.
The technical approach used followed the conceptual model developed for the PM2.5
FRM DQOs. This conceptual model was followed mainly due to its success in use with
PM2.5 and the flexibility of the conceptual model. It is a quite general model for
simulating the characterization of ambient concentrations in terms of annual or multi-year
averages from 1 in n day sampling. The model incorporates several sources of
variability: seasonal variability, natural day-to-day variability, sampling incompleteness,
and measurement error. The measurement error was restricted to a precision component
without a bias component because the final mathematical form of the assessment of
trends is robust to multiplicative bias. Pollutant specific parameters were used in the
modeling. The parameters describing the natural variation of the pollutants were based
on data analyses of the Pilot City data and the Air Toxics Archive. Finally, separate
urban and rural DQOs were established for the pollutants that were sufficiently measured
in rural locations of the Pilot Study.
A workgroup organized by EPA/OAQPS/EMAD provided representatives of data users,
decision makers, state and local parties, and monitoring and laboratory personnel.
Battelle provided technical statistical support throughout the process with examples and
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 2 of 19
data analyses. The workgroup guided the DQO development and made the decisions that
were not driven by data analyses in the DQO development during a series of conference
calls. These decisions included items such as establishing a specific mathematical form
for measuring trends and establishing limits on the sampling rate. Battelle and EPA also
held a meeting in Research Triangle Park, North Carolina, on June 17, 2002 to discuss
the development details.
7.1 THE GENERAL DQO PROCESS
This section presents an overview of the seven steps in EPA's QA/G-4 DQO process as
applied to one of the primary goals of the National Air Toxics Monitoring Network,
namely to establish trends and evaluate the effectiveness of HAP reduction strategies (see
www.epa.gov/quality/qa docs.html). The purpose of this section is to provide general
discussion on the specific issues that were used in developing the DQOs as they relate to
the general DQO process.
The DQO process is a seven-step process based on the scientific method to ensure that
the data collected by EPA meet the needs of its data users and decision makers in terms
of the information to be collected, in particular the desired quality and quantity of data. It
also provides a framework for checking and evaluating the program goals to make sure
they are feasible and that the data are collected efficiently. The seven steps are usually
labeled as:
State the Problem
• Identify the Decision
Identify the Inputs to the Decision
• Define the Study Boundaries
Develop a Decision Rule
• Specify Tolerable Limits on the Decision Errors
• Optimize the Design.
This section has general discussion for each of these items. The pollutant specific
outcomes of the DQO process are contained in Section 3.
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 3 of 19
7.2 State the Problem
Characterize the ambient concentrations in the region represented by the monitor
to establish any significant downward trend (measured by a per cent change
between successive 3-year means of the concentrations).
The ability to characterize the trends was statistically modeled. The statistical model was
designed by starting with a model similar to the one used for PM2.5 FRM data. The
ambient concentrations are modeled as deviations from a sine curve, where the sine curve
represents seasonality. This sine curve represents long-term daily averages of the
concentrations that one would observe at the site. The form used is as follows:
A
r-\) day
1+ sin —2 n
(r+ iJ 1365
where
A = the long term annual average and
r = the ratio of the highest point on the sine curve to the lowest point. A value
of r = 1 indicates no seasonality.)
The natural deviations from the sine curve are assumed to follow a lognormal distribution
with a mean that is given by the particular point on the sine curve. (For example, the
value of the sine curve for Day 100 is the mean for all Day 100s across many years.) The
coefficient of variation (CV) of the lognormal distribution is assumed to be a constant.
The general model considered also allows for the day-to-day deviations from the sine
curve to be correlated, but the current DQOs are based on a correlation of zero. (The
correlation effectively measures how quickly the concentrations can change from one
deviation from the sine curve to another. A correlation of zero indicates that it can
change fast enough that values measured on consecutive days would be completely
independent. A value of 0.2 would say that a positive deviation from the curve is
somewhat more likely to be followed by another positive deviation than a negative
deviation. A value of 0.9 would indicate that positive deviations are almost always
followed by another positive deviation.) Finally, the measured values are modeled with a
normally distributed random measurement error with a constant coefficient of
variation (CV). The specific values for the various parameters are pollutant specific.
The population parameters (the degree of seasonality, the autocorrelation, and the CV of
the deviations from the sine curve) were estimated from the Pilot City data (and in the
case of benzene compared with estimates from the Air Toxics Data Archive). A near
worst-case choice was made for each of the parameters. The power curves and decision
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 4 of 19
errors are established via Monte-Carlo simulation of the model with the particular
parameters for various combinations of truth and observed percent changes in three-year
mean concentrations. The power curves are plotted as functions of the true percent
change in the three-year annual means for compound specific combinations of the
sampling frequency, completeness, and precision. Decision errors are stated for these
worst-case scenarios.
Note: It was decided by the workgroup from budgetary considerations that the proposed
DQOs should be constrained to no more than one in six day sampling.
7.3 Identify the Decision
The decision statement should provide a link between the principal study question and
possible actions. The potential actions associated with achieving or failing to achieve a
particular percent decrease in the observed three-year mean concentration were not
defined by the workgroup. However, it was decided that any decision would be based on
whether or not a 15 percent decrease was observed. Hence the form of the decision was
fixed, and may be specified as follows:
Significant decreases (15percent or more) between successive three-year mean
concentration levels will result in ... Insignificant decreases, (increases, or
decreases of less than 15 percent) will trigger alternate actions of.
7.4 Identify the Inputs to the Decision
Only six HAPs (benzene, 1,3-butadiene, arsenic, chromium, acrolein, and formaldehyde)
were considered in the DQO development. It is assumed that the other pollutants will be
represented by at least one of these six. The statements included here apply implicitly to
the other HAPs.
It is assumed that the analytical techniques used in the Pilot study will be used throughout
the program. Most importantly for the DQOs the Method Detection Limits (MDLs) will
not increase. The pollutant specific MDLs assumed are listed in Section 3. Those values
were identified as pollutant-site maximums that were achieved by at least two of the pilot
sites in each pollutant's case.
Among the key decisions made as a part of the DQO process was that each pollutant will
need to be measured on a schedule of at least once every six days with a quarterly
completeness of 85 percent for six consecutive years. The completeness criterion was
checked against the pilot data, and was generally achieved. All valid measurements
count toward the completeness goal, including non-detects. The analysis of the trends at
the site level will be based on a percent difference between the mean of the first three
annual concentrations and the mean of the last three annual concentrations. Hence for
each year the annual average concentration, X^ needs to be found, i = 1, 2, ... 6. Next
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 5 of 19
find the mean, X, for the first three years and the mean, Y, for years 4 through 6 as
follows:
r =
Then the downward trend, T, is the percent decrease from the first three-year period to
the second three-year period. Namely,
x
The Action Level is the cutoff point that separates different decision alternatives.
Based on the assumed budgetary constraint of one in six day sampling and the natural
variation exhibited by the six compounds considered, an action level of 15 percent was
chosen. Hence at least a 15 percent decrease between the two distinct three-year mean
concentrations will need to be observed in order to be considered a significant decrease.
This assumes that the mean concentrations are above the health standards, and hence it
makes sense to consider trends. (Note that characterizing the mean concentrations is a
separate goal of the Air Toxics program that has not yet been considered and could result
in different DQOs.)
7.5 Define the Study Boundaries
It is desired that the specific location of the monitors be constrained so that they represent
neighborhood scale assessment for each of the two three-year periods under
consideration. The details of how to ensure this goal have not yet been determined.
Some guideline is provided by the Air Toxics Monitoring Concept Paper (see
http://www.epa.gov/ttn/antic/airtxfil.html).
7.6 Develop a Decision Rule
The decision rule is an "if ... then" statement for how the various alternatives will be
chosen. As noted above the specific alternative actions have not been formalized yet, just
the form of the decision rule.
If the percent change between successive three-year average
concentration levels is greater than or equal to 15 percent, then
...Otherwise ...
7.7 Specify Tolerable Limits on the Decision Errors
Since the program will not generate complete, error-free data, there will be some
probability of making a decision error. The main goal of the DQO process is to find a
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 6 of 19
workable balance between how complete and error free the data are with acceptable
levels of decision errors. To find the balance, the possible errors need to be carefully
defined. This usually needs to be done with the recognition that there will be a range,
often called the gray zone, where it is impractical to control decision errors.
The QA/G-4 guidance recommends using 0.01 as the starting point for setting decision
error rates. However, such a limit would generally require a sampling rate that is not
feasible. The workgroup decided on the following limits:
If there is no true decrease in the three-year average concentrations, then
the probability of observing a mean concentration for years four through
six that is at least 15 percent below the observed mean concentration from
years one through three should be no more than 10 per cent.
If there is a true decrease in the three-year average concentrations of at
least 30 percent, then the probability of observing a mean concentration
for years four through six that is less than 15 percent below the observed
mean concentration from years one through three should be no more than
10 percent.
Equivalently, the second statement could read that:
If there is a true decrease in the three-year average concentrations of at
least 30 percent, then the probability of observing a mean concentration
for years four through six that is at least 15 percent below the observed
mean concentration from years one through three should be at least 90
percent.
The power curves shown in Section 3 show the probability of observing at
least a 15 percent decrease as a function of the true decrease. In terms of the
above goals this means that the power curve graphs should start below 10 percent
for a true percent change of 0 and end above 90 percent for a true percent change
of 30 percent. Since there is a particular interest in the error rates for no true
change and for a true change of a 30 percent decrease, this associated x-axis
(horizontal axis) range is shown for each curve. Also, it is sometimes useful to
know when the two target error rates are achieved. The range of "truth" between
these values is referred to as the gray zone, i.e., the range of true percent
decreases that cannot be reliably detected by the sampling scheme. These are also
given for each curve (and indicated with vertical dotted lines).
7.8 Optimize the Design
In each pollutant's case, a sampling schedule of once every six days is set forth with a
quarterly completeness criteria of 85 percent. Pilot City study participants were surveyed
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 7 of 19
and almost all were collecting and obtaining valid data values at a rate that exceeded 85
percent for each of the six compounds considered (valid non-detects counted toward
completeness). Hence, the target rate of 85 percent was selected, instead of the more
common 75 percent completeness goal. This should make the power curves more
representative of the network's expected monitoring conditions.
7.9 DQOS For the Six Study Compounds
This section states the design values, namely it gives the expected maximum error rates,
gray zones, and power curves for each of the six compounds considered explicitly. The
parameters describing the natural state of the ambient conditions used to construct the
power curves, error rates and gray zone are compound specific based on data from the
Pilot Study. In each case, the Pilot City data yielded a range of estimates. The specific
values used were the extremes (or nearly so) that would make detecting a downward
trend more difficult. Actual performance in almost all cases should be better than that
indicated by the power curves, since specific sites would not be characterized by these
extremes in each of these parameters. However, since the sensitivity to the different
parameters is not the same, the DQOs need to protect against a combined set of extremes.
Hence, the use of extremes for network design purposes is conservative.
Since the rural sites can be quite different from urban sites, separate DQOs are shown in
those cases where there were sufficient data to support investigating a separate set of
DQOs. In the case of formaldehyde, the urban and rural DQOs are essentially the same.
There are twelve input parameters shown in each section. They are:
1. Tl. This is the target error rate for when there is no change. It is always
10 percent.
2. T2. This is the target error rate for when there is a 30 percent decrease. It is
always 10 percent.
3. The action limit. This is the minimum observed percent change from the
mean concentration of the first three years to the mean concentration from the
last three years that would be used to indicate that the concentrations have
decreased. Decreases less than this amount would not be considered
significant decreases in the mean concentration.
4. The sampling rate. It is set to one in six day sampling in each case.
5. The quarterly completeness criterion. This was set to 85 percent based on the
recommendation of ERG and a review of the Pilot Study data completeness.
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 8 of 19
6. Measurement error Coefficient of Variation (CV). This was assumed to be
15 percent for each compound. (A sensitivity analysis showed that the DQOs
are robust to moderate changes in this value.)
7. Seasonality ratio. This is a measure of the degree of seasonality.
Specifically, it is the ratio of the highest point on the seasonal curve to the
lowest point. A value of 1 indicates no seasonality. Larger values make it
more difficult to estimate an annual or three-year mean concentration, and
hence larger values make it more difficult to measure the percent change.
8. Autocorrelation. This is a measurement of how quickly day-to-day deviation
from the seasonal curve can occur. A value of 0 indicates that changes occur
quickly enough that each day is independent of the preceding day. Values
greater than 0 indicate that the changes are generally slower, so that days with
concentrations above the seasonal curve are more likely to be followed by
another day above the seasonal curve. Values greater than 0 increase the
precision of the three-year means and the percent change between the three-
year means. Hence, a value of 0 is the most conservative choice for the
DQOs. Zero was used in all cases, because many daily measurements are
required to obtain a reliable estimate of this parameter.
9. Population CV. This is a measurement of the natural variation about the
seasonal curve. Larger values decrease the precision of the three-year mean
concentration estimates and the percent change between them. The power
curves are strongly dependent on this parameter, but the estimates can be
strongly influenced by a few outlier values. Generally the 90th percentile of
the estimates from the Pilot study was used as a balance between these
competing forces. This value was then rounded up to be a multiple of 5
percent for the urban DQOs. For the rural DQOs an additional 5 percent was
added, since there were fewer rural sites on which to base the estimates.
10. MDL. This is the MDL used in the simulations. The value was chosen to be
a reasonably attainable maximum for a site and compound.
11. Initial mean concentration. This is the mean concentration of the first three
years in the simulations. Values closer to the MDL decrease the precision of
the percent change estimate. The value chosen was approximately equal to
the 25th percentile of the site-compound means from the Pilot study.
12. Health Risk Standard. This value is shown for reference only. It was not
used in the simulations.
In addition to the power curves, there are three sets of output values.
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 9 of 19
1. Error0 is the percent of the simulations with no change in the true three-year
means that in fact generated at least a 15 percent decrease in the observed
three-year means.
2. Error30 is the percent of the simulations with a 30 percent decrease in the true
three-year means that generated less than a 15 percent decrease in the
observed three-year means.
3. The gray zone is the interval of the true decreases that cannot be detected with
confidence by the study design. In this range, the probability of observing at
least a 15 percent decrease is greater than 10 percent, but less than 90 percent.
In summary, based on variability and uncertainty estimates from the ten-city Pilot Study,
the following Sections 3.1 through 3.10 suggest that the specified air toxics trends DQOs
will be met for monitoring sites that satisfy the goals of 1 in 6 day sampling, 85 percent
completeness, and 15 percent measurement CV. These results were explicitly developed
for benzene (urban and rural); 1,3-butadiene (urban and rural); arsenic (urban and rural);
chromium (urban only); acrolein (urban only); and formaldehyde (urban and rural).
7.10 DQOs for Measuring the Percent Decrease
7.10.1 Benzene at Urban Locations
Table 3.1.1 shows the input parameters used in the simulation model in developing the DQOs for
measuring the percent decrease between three-year mean concentrations of benzene at urban locations.
Table 3.1.2 shows the output values from the simulations. Figure 3.1.1 shows the associated power curve,
which is the probability of observing a 15 percent difference between successive three-year means as a
function of the true percent difference in the distinct three-year means. In summary, based on variability
and uncertainty estimates from the ten-city Pilot Study data, Table 3.1.2 suggests that the specified air
toxics trends DQOs will be met for benzene at urban monitoring sites that satisfy the goals of one in six-
day sampling, 85 percent completeness, and 15 percent measurement CV. (See section 3.0 for definitions
of the input parameters and output values.)
Table 7.1
DQO input parameters for benzene at urban locations
T1
10%
T2
10%
Action Limit
15%
Measurement CV
15%
Sampling Rate
1 in 6 day
Completeness
85%
Seasonality
4.5
Autocorrelation
0
Population CV
85%
MDL (Mg/m3)
0.044
Initial
Concentration (u.g/m3)
1.0
Risk Standard (u.g/m3)
0.128
Table 7.2
DQO output parameters for benzene at urban locations
I Error rate for no true change
6%
Error rate for 30% decrease
97%
Gray zone I
3% - 26% |
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 10 of 19
10 15 20
True percent difference
25
30
Figure 7.1
Power curve for detecting a 15 percent decrease between successive three-year
means of benzene concentrations based on the data variation found in urban
locations of the Pilot Study
7.10.2 DQOs for Measuring the Percent Decrease of Benzene at Rural Locations
Table 3.2.1 shows the input parameters used in the simulation model in developing the
DQOs for measuring the percent decrease between three-year mean concentrations of
benzene at rural locations. Table 3.2.2 shows the output values from the simulations.
Figure 3.2.1 shows the associated power curve, which is the probability of observing a 15
percent difference between successive three-year means as a function of the true percent
difference in the distinct three-year means. In summary, based on variability and
uncertainty estimates from the ten-city Pilot Study data, Table 3.2.2 suggests that the
specified air toxics trends DQOs will be met for benzene at rural monitoring sites that
satisfy the goals of one in six-day sampling, 85 percent completeness, and 15 percent
measurement CV. (See section 3.0 for definitions of the input parameters and output
values.)
Table 7.3
DQO input parameters for benzene at rural locations
T1
10%
T2
10%
Action Limit
15%
Measurement CV
15%
Sampling Rate
1 in 6 day
Completeness
85%
Seasonality
4.0
Autocorrelation
0
Population CV
60%
MDL (Mg/m3)
0.044
Initial
Concentration (u.g/m3)
1.0
Risk Standard (u.g/m3)
0.128
Table 7.4
DQO output parameters for benzene at rural locations
I Error rate for no true change
2%
Error rate for 30% decrease
99%
Gray zone I
7% - 23% |
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 11 of 19
'•3
Figure 7.2
l
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
10 15 20
True percent difference
25
30
Power curve for detecting a 15 percent decrease between successive three-year
means of benzene concentrations based on the data variation found in rural
locations of the Pilot Study
7.10.3 DQOs for Measuring the Percent Decrease of 1,3-Butadiene at Urban
Locations
Table 3.3.1 shows the input parameters used in the simulation model in developing the
DQOs for measuring the percent decrease between three-year mean concentrations of
1,3-butadiene at urban locations. Table 3.3.2 shows the output values from the
simulations. Figure 3.3.1 shows the associated power curve, which is the probability of
observing a 15 percent difference between successive three-year means as a function of
the true percent difference in the distinct three-year means. In summary, based on
variability and uncertainty estimates from the ten-city Pilot Study data, Table 3.3.2
suggests that the specified air toxics trends DQOs will be met for 1,3-butadiene at urban
monitoring sites that satisfy the goals of one in six-day sampling, 85 percent
completeness, and 15 percent measurement CV. (See section 3.0 for definitions of the
input parameters and output values.)
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 12 of 19
Table 7.5
DQO input parameters for 1,3-butadiene at urban locations
T1
10%
T2
10%
Action Limit
15%
Measurement CV
15%
Sampling Rate
1 in 6 day
Completeness
85%
Seasonality
7.0
Autocorrelation
0
Population CV
100%
MDL (jig/m3)
0.02
Initial
Concentration (u.g/m3)
0.1
Risk Standard (u.g/m3)
10'°
Table 7.6
DQO output parameters for 1,3-butadiene at urban locations
Error rate for no true change
10%
Error rate for 30% decrease
94%
Gray zone
0% - 28%
CO
'•3
•§
o.
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
10 15 20
True percent difference
25
30
Figure 7.3 Power curve for detecting a 15 percent decrease between successive
three-year means of 1,3-butadiene concentrations based on the data variation found
in urban locations of the Pilot Study
7.10.4 DQOs for Measuring the Percent Decrease of 1,3-butadiene at Rural
Locations
Table 3.4.1 shows the input parameters used in the simulation model in developing the
DQOs for measuring the percent decrease between three-year mean concentrations of
1,3-butadiene at rural locations. Table 3.4.2 shows the output values from the
simulations. Figure 3.4.1 shows the associated power curve, which is the probability of
observing a 15 percent difference between successive three-year means as a function of
the true percent difference in the distinct three-year means. In summary, based on
variability and uncertainty estimates from the ten-city Pilot Study data, Table 3.4.2
suggests that the specified air toxics trends DQOs will be met for 1,3-butadiene at rural
monitoring sites that satisfy the goals of one in six-day sampling, 85 percent
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 13 of 19
completeness, and 15 percent measurement CV. (See section 3.0 for definitions of the
input parameters and output values.)
Table 7.7
DQO input parameters for 1,3-butadiene at rural locations
T1
10%
T2
10%
Action Limit
15%
Measurement CV
15%
Sampling Rate
1 in 6 day
Completeness
85%
Seasonality
6.0
Autocorrelation
0
Population CV
75%
MDL (jig/m3)
0.02
Initial
Concentration (u.g/m3)
0.1
Risk Standard (u.g/m3)
10'°
Table 7.8
DQO output parameters for 1,3-butadiene at rural locations
Error rate for no true change
4%
Error rate for 30% decrease
98%
Gray zone
4% - 25%
ct
o
•a
10 15 20
True percent difference
25
30
Figure 7.4 Power curve for detecting a 15 percent decrease between successive
three-year means of 1,3-butadiene concentrations based on the data variation found
in rural locations of the Pilot Study
7.10.5 DQOs for Measuring the Percent Decrease of Arsenic at Urban Locations
Table 3.5.1 shows the input parameters used in the simulation model in developing the
DQOs for measuring the percent decrease between three-year mean concentrations of
arsenic at urban locations. Table 3.5.2 shows the output values from the simulations.
Figure 3.5.1 shows the associated power curve, which is the probability of observing a 15
percent difference between successive three-year means as a function of the true percent
difference in the distinct three-year means. In summary, based on variability and
uncertainty estimates from the ten-city Pilot Study data, Table 3.5.2 suggests that the
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 14 of 19
specified air toxics trends DQOs will be met for arsenic at urban monitoring sites that
satisfy the goals of one in six-day sampling, 85 percent completeness, and 15 percent
measurement CV. (See section 3.0 for definitions of the input parameters and output
values.)
Table 7.9
DQO input parameters for arsenic at urban locations
T1
10%
T2
10%
Action Limit
15%
Measurement CV
15%
Sampling Rate
1 in 6 day
Completeness
85%
Seasonality
5.0
Autocorrelation
0
Population CV
85%
MDL (ng/m3)
0.000046
Initial
Concentration (u.g/m3)
0.002
Risk Standard (u.g/m3)
0.0043
Table 7.10 DQO output parameters for arsenic at urban locations
Error rate for no true change
8%
Error rate for 30% decrease
95%
Gray zone
2% - 27%
i
0.9
0.3
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
10 15 20
True percent difference
25
30
Figure 7.5 Power curve for detecting a 15 percent decrease between successive
three-year means of arsenic concentrations based on the data variation found in
urban locations of the Pilot Study
7.10.6 DQOs for Measuring the Percent Decrease of Arsenic at Rural Locations
Table 3.6.1 shows the input parameters used in the simulation model in developing the
DQOs for measuring the percent decrease between three-year mean concentrations of
arsenic at rural locations. Table 3.6.2 shows the output values from the simulations.
Figure 3.6.1 shows the associated power curve, which is the probability of observing a 15
percent difference between successive three-year means as a function of the true percent
difference in the distinct three-year means. In summary, based on variability and
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 15 of 19
uncertainty estimates from the ten-city Pilot Study data, Table 3.6.2 suggests that the
specified air toxics trends DQOs will be met for arsenic at rural monitoring sites that
satisfy the goals of one in six-day sampling, 85 percent completeness, and 15 percent
measurement CV. (See section 3.0 for definitions of the input parameters and output
values.)
Table 7.11
DQO input parameters for arsenic at rural locations
T1
10%
T2
10%
Action Limit
15%
Measurement CV
15%
Sampling Rate
1 in 6 day
Completeness
85%
Seasonality
4.0
Autocorrelation
0
Population CV
65%
MDL (jig/m3)
0.000046
Initial
Concentration (u.g/m3)
0.001
Risk Standard (u.g/m3)
0.0043
Table 7.12 DQO output parameters for arsenic at rural locations
Error rate for no true change
3%
Error rate for 30% decrease
99%
Gray zone
5% - 24%
'•S
•§
DL,
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
10 15 20
True percent difference
25
30
Figure 7.6 Power curve for detecting a 15 percent decrease between successive
three-year means of arsenic concentrations based on the data variation found in
rural locations of the Pilot Study
7.10.7 DQOs for Measuring the Percent Decrease of Chromium
Table 3.7.1 shows the input parameters used in the simulation model in developing the
DQOs for measuring the percent decrease between three-year mean concentrations of
chromium. Table 3.7.2 shows the output values from the simulations. Figure 3.7.1
shows the associated power curve, which is the probability of observing a 15 percent
difference between successive three-year means as a function of the true percent
difference in the distinct three-year means. In summary, based on variability and
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 16 of 19
uncertainty estimates from the ten-city Pilot Study data, Table 3.7.2 suggests that the
specified air toxics trends DQOs will be met for chromium at monitoring sites that satisfy
the goals of one in six-day sampling, 85 percent completeness, and 15 percent
measurement CV. (See section 3.0 for definitions of the input parameters and output
values.)
Table 7.13
DQO input parameters for chromium
T1
10%
T2
10%
Action Limit
15%
Measurement CV
15%
Sampling Rate
1 in 6 day
Completeness
85%
Seasonality
5.0
Autocorrelation
0
Population CV
90%
MDL (jig/m3)
0.00018
Initial
Concentration (u.g/m3)
0.0015
Risk Standard (u.g/m3)
0.012
Table 7.14
DQO output parameters for chromium
Error rate for no true change
7%
Error rate for 30% decrease
96%
Gray zone
2% - 27%
10 15 20
True percent difference
25
30
Figure 7.7 Power curve for detecting a 15 percent decrease between successive
three-year means of chromium concentrations based on the data variation found in of
the Pilot Study
7.10.8 DQOs for Measuring the Percent Decrease of Acrolein
Table 3.8.1 shows the input parameters used in the simulation model in developing the
DQOs for measuring the percent decrease between three-year mean concentrations of
acrolein. Table 3.8.2 shows the output values from the simulations. Figure 3.8.1 shows
the associated power curve, which is the probability of observing a 15 percent difference
between successive three-year means as a function of the true percent difference in the
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 17 of 19
distinct three-year means. In summary, based on variability and uncertainty estimates
from the ten-city Pilot Study data, Table 3.8.2 suggests that the specified air toxics trends
DQOs will be met for acrolein at monitoring sites that satisfy the goals of one in six-day
sampling, 85 percent completeness, and 15 percent measurement CV. (See section 3.0
for definitions of the input parameters and output values.)
Table 7.15 DQO input parameters for acrolein
T1
10%
T2
10%
Action Limit
15%
Measurement CV
15%
Sampling Rate
1 in 6 day
Completeness
85%
Seasonality
4.0
Autocorrelation
0
Population CV
105%
MDL (jig/m3)
0.14
Initial
Concentration (u.g/m3)
0.4
Risk Standard (u.g/m3)
-
Table 7.16
DQO output parameters for acrolein
Error rate for no true change
10%
Error rate for 30% decrease
91%
Gray zone
0% - 29%
•-8
(8
cu
0 5 10 15 20 25
True percent difference
Figure 7.8 Power curve for detecting a 15 percent decrease between successive three-year means of
acrolein concentrations based on the data variation found in the Pilot Study
7.10.9 DQOs for Measuring the Percent Decrease of Formaldehyde at Urban
Locations
Table 3.9.1 shows the input parameters used in the simulation model in developing the
DQOs for measuring the percent decrease between three-year mean concentrations of
formaldehyde at urban locations. Table 3.9.2 shows the output values from the
simulations. Figure 3.9.1 shows the associated power curve, which is the probability of
30
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 18 of 19
observing a 15 percent difference between successive three-year means as a function of
the true percent difference in the distinct three-year means. In summary, based on
variability and uncertainty estimates from the ten-city Pilot Study data, Table 3.9.2
suggests that the specified air toxics trends DQOs will be met for formaldehyde at urban
monitoring sites that satisfy the goals of one in six-day sampling, 85 percent
completeness, and 15 percent measurement CV. (See Section 3.0 for definitions of the
input parameters and output values.)
Table 7.17
DQO input parameters for formaldehyde at urban locations
T1
10%
T2
10%
Action Limit
15%
Measurement CV
15%
Sampling Rate
1 in 6 day
Completeness
85%
Seasonality
7.0
Autocorrelation
0
Population CV
90%
MDL (Mg/m3)
0.014
Initial
Concentration (u.g/m3)
2.5
Risk Standard (u.g/m3)
1.3 10'°
Table 7.18
DQO output parameters for formaldehyde at urban locations
Error rate for no true change
8%
Error rate for 30% decrease
95%
Gray zone
2% - 27%
'-a
Figure 7.9
i
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
10 15 20
True percent difference
25
30
Power curve for detecting a 15 percent decrease between successive three-year
means of formaldehyde concentrations based on the data variation found in urban
locations of the Pilot Study
7.10.10 DQOs for Measuring the Percent Decrease of Formaldehyde at Rural
Locations
Table 3.10.1 shows the input parameters used in the simulation model in developing the
DQOs for measuring the percent decrease between three-year mean concentrations of
-------
Air Toxics Model QAPP
Element No: 7
Revision No: 1.0
Date: 11/04/02
Page 19 of 19
formaldehyde at rural locations. Table 3.10.2 shows the output values from the
simulations. Figure 3.10.1 shows the associated power curve, which is the probability of
observing a 15 percent difference between successive three-year means as a function of
the true percent difference in the distinct three-year means. In summary, based on
variability and uncertainty estimates from the ten-city Pilot Study data, Table 3.10.2
suggests that the specified air toxics trends DQOs will be met for formaldehyde at rural
monitoring sites that satisfy the goals of one in six-day sampling, 85 percent
completeness, and 15 percent measurement CV. (See Section 3.0 for definitions of the
input parameters and output values.)
Table 7.19
DQO input parameters for formaldehyde at rural locations
T1
10%
T2
10%
Action Limit
15%
Measurement CV
15%
Sampling Rate
1 in 6 day
Completeness
85%
Seasonality
7.0
Autocorrelation
0
Population CV
90%
MDL (jig/m3)
0.014
Initial
Concentration (u.g/m3)
2.1
Risk Standard (u.g/m3)
1.3 10'°
Table 7.20 DQO output parameters for formaldehyde at rural locations
Error rate for no true
change
8%
Error rate for 30% decrease
95%
Gray zone
1 % - 27%
••8
ctf
(D
10 15 20
True percent difference
25
30
Figure 7.10 Power curve for detecting a 15 percent decrease between successive three-year means of
formaldehyde concentrations based on the data variation found in rural locations of the
Pilot Study
-------
Air Toxics Model QAPP
Element No: 8
Revision No: 1.0
Date: 11/04/02
Page 1 of4
8.0 Special Training Requirements/Certification
8.1 Training
Personnel assigned to the NATTS will meet the educational, work experience, responsibility,
personal attributes, and training requirements for their positions. Records on personnel
qualifications and training will be maintained in personnel files and will be accessible for review
during audit activities.
Adequate education and training are integral to any monitoring program that strives for reliable
and comparable data. Training is aimed at increasing the effectiveness of employees and the
District. Table 8.1 represents the general training requirements for all employees, depending
upon there job classification.
Table 8.1 TCAPCD Training Requirements.
Job Classification
Training Title
Time/Fre quency
Requirement
Directors
Executive Development Program
As available
Branch Chief and above
Framework for Supervision
Keys to Managerial Excellence
EEO for Managers and Supervisors
Sexual Harassment
Contract Administration for Supervisors
40 hours of developmental activities
1st 6 months
After comp. of above
As available
Project Officers and Above
Contract Administration
Contract Administration Recertification
EEO for Managers and Supervisors
Grants Training
Project Officer Training (contract/grants)
Ethics in Procurement
Work statements for Negotiated Procurement
Prior to responsibility
Every three years
As available
Prior to responsibility
Field Personnel
24-Hour Field Safety
8- hour Field Safety Refresher
8-hour First Aid/CPR
Blood borre pathogens
1st time
Yearly
Yearly
1st time
Laboratory Personnel
24- Hour Laboratory Safety
4- Hour Refresher
Safety Video/Discussion
Chemical Spill Emergency Response
Blood borne pathogens
1st time
Yearly
Yearly
1st time
1st time
-------
Air Toxics Model QAPP
Element No: 8
Re vision No: 1.0
Date: 11/04/02
Page 2 of 4
8.1.1 Ambient Air Monitoring Training
Appropriate training is available to employees supporting the monitoring program, commensurate
with their duties. Such training may consist of classroom lectures, workshops, tele-conferences,
and on-the-job training.
Over the years, a number of courses have been developed for personnel involved with ambient air
monitoring and quality assurance aspects. Formal QA/QC training is offered through the
following organizations:
• Air Pollution Training Institute (APTI) http://www.epa.gov/oar/oaq.apti.html
• Air & Waste Management Association (AWMA) http://awma.org/epr.htm
• American Society for Quality Control (ASQC) http://www.asqc.org/products/educat.html
• EPA Institute
• EPA Quality Assurance Division (QAD) http://es.inel.gov/ncerqa/qal
• EPA Regional Offices
Table 8.2 presents a sequence of core ambient air monitoring and QA courses for ambient air
monitoring staff, and QA managers. The suggested course sequences assume little or no
experience in QA/QC or air monitoring. Persons having experience in the subject matter
described in the courses would select courses according to their appropriate experience level.
Courses not included in the core sequence would be selected according to individual
responsibilities, preferences, and available resources.
-------
Air Toxics Model QAPP
Element No: 8
Re vision No: 1.0
Date: 11/04/02
Page 3 of 4
Table 8.2. Core Ambient Air Training Courses
Sequence
1*
2*
3*
4*
5*
6*
7*
8
9
10
*
*
*
*
*
Course Title (SI = self instructional)
Air Pollution Control Orientation Course (Revised), SL422
Orientation to Quality Assurance Management
Introduction to Ambient Air Monitoring (Under Revision), 81:434
General Quality Assurance Considerations for Ambient Air Monitoring (Under
Revision); 81:471
Quality Assurance for Air Pollution Measurement Systems (Under Revision),
470
Data Quality Objectives Workshop
Quality Assurance Project Plan
Atmospheric Sampling (Under Revision), 435
Analytical Methods for Air Quality Standards, 464
Chain-of-Custody Procedures for Samples and Data, 81:443
Data Quality Assessment
Management Systems Review
Beginning Environmental Statistical Techniques (Revised), SL473A
Introduction to Environmental Statistics, SL473B
Statistics for Effective Decision Making
AQS Training
Source
APTI
QAD
APTI
APTI
APTI
QAD
QAD
APTI
APTI
APTI
QAD
QAD
APTI
APTI
ASQC
OAQPS
* Courses recommended for QA Managers
8.2 Certification
For the NATTS, the QA Division will issue training certifications for the successful completion
of field, laboratory, sample custody and data management training. Certification will be based
upon the qualitative and quantitative assessment of individual's adherence to the SOPs.
-------
Air Toxics Model QAPP
Element No: 9
Re vision No: 1.0
Date: 11/04/02
Page 1 of 5
9.0 Documentation and Records
For the ATMP, there are number of documents and records that need to be retained. A document,
from a records management perspective, is a volume that contains information which describes,
defines, specifies, reports, certifies, or provides data or results pertaining to environmental
programs. As defined in the Federal Records Act of 1950 and the Paperwork Reduction Act of
1995 (now 44 U.S.C. 3101-3107), records are: "...books, papers, maps, photographs, machine
readable materials, or other documentary materials, regardless of physical form or characteristics,
made or received by an agency of the United States Government under Federal Law or in
connection with the transaction of public business and preserved or appropriate for preservation
by that agency or its legitimate successor as evidence of the organization, functions, policies,
decisions, procedures, operations, or other activities of the Government or because of the
informational value of data in them..." The TCAPCD follows the guidelines to ensure the public
that the District's procedures are being performed within the guidelines of the Paper Reduction
Act.
The following information describes the Air Pollution Control's document and records procedures
for the NATTS. In this QAPP regulation and guidance, the District uses the term reporting
package. It is defined as all the information required to support the concentration data reported to
EPA and the State, which includes all data required to be collected as well as data deemed
important by the District under its policies and records management procedures. Table 9-1
identifies these documents and records.
9.1 Information Included in the Reporting Package
9.1.1 Routine Data Activities
The TCAPCD has a structured records management retrieval system that allows for the efficient
archive and retrieval of records. The air toxics information will be included in this system. It is
organized in a similar manner to the EPA's records management system (EPA-220-B-97-003)
and follows the same coding scheme in order to facilitate easy retrieval of information during
EPA technical systems audits and network reviews. Table 9.1 includes the documents and records
that will be filed according to the statute of limitations discussed in Section 9.3. In order to
archive the information as a cohesive unit, the air toxics information will be filed under the
individual codes depending on the chemical makeup of the compound. Please see Table 9.1.
-------
Project: Model QAPP
Element No: 9
Re vision No: 1.0
Date: 12/3/02
Page 2 of 5
Table 9.1 Air Toxics Reporting Package Information
Categories
Management
and
Organization
Site
Information
Environmental
Data
Operations
Raw Data
Data Reporting
Record/Document Types
State Implementation Plan
Reporting agency information
Organizational structure
Personnel qualifications and
training
Training Certification
Quality management plan
Document control plan
EPA Directives
Grant allocations
Support Contract
Network description
Site characterization file
Site maps
Site Pictures
QA Project Plans
Standard operating procedures
(SOPs)
Field and laboratory notebooks
Sample handling/custody records
Inspection/Maintenance records
Any original data (routine and QC
data) including data entry forms
Electronic deliverables of
summary analytical and associated
QC and calibration runs per
instrument
Air quality index report
Data summary reports
Journal
articles/papers/presentations
File Codes
AIRP/217
AIRP/237
ADMI/106
PERS/123
AIRP/482
AIRP/216
ADMI/307
DIRE/007
BUDG/043
CONT/003
CONT/202
AIRP/237
AIRP/237
AIRP/237
AUDV/708
PROG/185
SAMP/223
SAMP/502
TRAN/643
AIRP/486
SAMP/223
SAMP/224
AIRP/484
AIRP/484
PUBL/250
-------
Project: Model QAPP
Element No: 9
Re vision No: 1.0
Date: 12/3/02
Page 3 of 5
Categories
Data
Management
Quality
Assurance
Record/Document Types
Data algorithms
Data management
plans/flowcharts
Air toxics Data
Data Management Systems
Good Laboratory Practice
Network reviews
Control charts
Data quality assessments
QA reports
System audits
Response/Corrective action
reports
Site Audits
File Codes
INFO/304
INFO/304
INFO/160 -
INFO/173
INFO/304 -
INFO/170
COMP/322
OVER/255
SAMP/223
SAMP/223
OVER/203
OVER/255
PROG/082
OVER/658
9.1.2 Annual Summary Reports Submitted to EPA
The TCAPCD shall submit to EPA Region 11 Office, included in the standard annual summary
report, all the air toxics data collected within that calendar year. The report will be submitted by
April 1 of each year for the data collected from January 1 to December 31 of the previous year.
The report will contain the following information:
Site and Monitoring Information
• City name;
• county name and street address of site location;
• AQS site code;
• AQS monitoring method code.
Summary Data
Annual arithmetic mean, and
Sampling schedule used as once every 6-day schedule.
Dr. Melvin Thomas, as the senior air pollution control officer for the District, will certify that the
annual summary is accurate to the best of his knowledge. This certification will be based on the
various assessments and reports performed by the organization, in particular, the Quality
-------
Project: Model QAPP
Element No: 9
Re vision No: 1.0
Date: 12/3/02
Page 4 of 5
Assurance Annual Report (QAAR). Section 21 documents the quality of the air toxics data and
the effectiveness of the quality system.
9.2 Data Reporting Package Format and Documentation Control
Table 9-1 represents the documents and records, at a minimum, that must be filed into the
reporting package. The details of these various documents and records will be discussed in the
appropriate sections of this document.
All raw data required for the calculation of air toxics concentrations, the submission to the AQS
database, and QA/QC data, are collected electronically, on data forms or in notebooks that are
included in the field and analytical methods sections. All hardcopy information will be filled out
in indelible ink. Corrections will be made by inserting one line through the incorrect entry,
initialing this correction, the date of correction and placing the correct entry alongside the
incorrect entry, if this can be accomplished legibly, or by providing the information on a new line.
9.2.1 Notebooks
The District will issue notebooks to each field and laboratory technician. This notebook will be
uniquely numbered and associated with the individual and the NATTS. Although data entry
forms are associated with all routine environmental data operations, the notebooks can be used to
record additional information about these operations. All notebooks will be bound as well as
paginated so that individual pages cannot be removed unnoticeably.
Field notebooks - Notebooks are issued for each sampling site. The TCAPCD NATTS will also
have a sampling site notebook. This will be 3-ring binders that will contain the appropriate data
forms for routine operations as well as inspection and maintenance forms and SOPs.
Lab Notebooks - Notebooks will also be issued for the laboratory. These notebooks will be
uniquely numbered and associated with the ATMP. One notebook will be available for general
comments/notes; others will be associated with, the temperature and humidity recording
instruments, the refrigerator, calibration equipment/standards, and the analytical balances and
instruments used for this program.
Sample shipping/receipt- One notebook will be issued to the shipping and receiving facility.
This notebook will be uniquely numbered and associated with the ATMP. It will include standard
forms and areas for free form notes.
-------
Project: Model QAPP
Element No: 9
Re vision No: 1.0
Date: 12/3/02
Page 5 of 5
9.2.2 Electronic data collection
In order to reduce the potential for data entry errors, automated systems will be utilized where
appropriate and will record the same information that is found on data entry forms. In order to
provide a back-up, a hardcopy of automated data collection information will be stored for the
appropriate time frame in project files. The Information Manager will back-up analytical data
acquired by each laboratory instrument including tuning, calibrations and QC sample runs
associated with samples.
9.3 Data Reporting Package Archiving and Retrieval
In general, all the information listed in Table 9-1 will be retained for 5 years from the date the
grantee submits its final expenditure report unless otherwise noted in the funding agreement.
However, if any litigation, claim, negotiation, audit or other action involving the records has been
started before the expiration of the 5-year period, the records will be retained until completion of
the action and resolution of all issues which arise from it, or until the end of the regular 5-year
period, whichever is later. The District will extend this regulation in order to store records for
three full years past the year of collection. For example, any data collected in calendar year 2003
(1/1/03 - 12/31/03) will be retained until, at a minimum, January 1, 2009, unless the information
is used for litigation purposes.
-------
Project: Model QAPP
Element No: 10
Revision No: 1.0
Date: 11/04/02
Page 1 of6
Section 10.0 Sampling Design
10.1 Scheduled Project Activities, Including Measurement Activities
TCAPCD will be monitoring concentrations at five locations. Therefore, this section will discuss
the operation and installation of all samplers including the NATTS. Since the NATTS is a
nationally recognized monitoring location each system (with the exception of the Aethalometer),
will be collocated. Table 10.1 represents the activities associated with the ordering and deployment
of the primary and collocated samplers. Please note that since the TCAPCD had participated in the
National Air Toxics Pilot Program, many of the samplers were already in place. However, the
NATTS site will be expanded to accommodate the collocated samplers.
Table 10.1. Schedule of Airtoxics Sampling-Related Activities
Activity
Receive samples
Remove QA PM10 sampler from
TC3.
Install collocated PM10 sampler at site
TC5
Receive Aethalometer
Begin testing laboratory equipment
Perform pre-run sampling
Calibrate all field samplers
Begin routine sampling
Due Date
September 1,2002
September 2002
November 2002
November 2002
November 2002
November 2002
December 2002
January 1,2003
Comments
After receipt, begin conditioning of filters
Collocated samplers installed. VOC, ALD, PM10
Test in laboratory. After two weeks, install at TC5
Bring analytical equipment on line
Test samplers and analytical laboratory systems
Perform official calibration for samplers
Begin NATTS sampling
10.2 Rationale for the Design
10.2.1 Primary Samplers
The purpose of the ATMP operated by Toxa City is to ascertain the long term trends as described
in Chapter 7. To determine whether these characteristics are quantified with sufficient confidence,
Toxa City must address sampler type, sampling frequency, and sampler siting. By employing
samplers that are described in the appropriate compendia1'2'3, the data collected will be comparable
-------
Project: Model QAPP
Element No: 10
Revision No: 1.0
Date: 11/04/02
Page 2 of6
to standard EPA methods. By complying with the sampling frequency requirements of Network
Design and Site Exposure Criteria for Selected Noncriteria Air Pollutants4,Toxa City assumes that
the sampling frequency is sufficient to attain the desired confidence in the annual 95th percentile
and annual mean of concentrations in the vicinity of each monitor. By selecting sampler locations
using the rules in Network Design and Site Exposure Criteria for Selected Noncriteria Air
Pollutants, Toxa City can be confident that the concentrations within its jurisdiction are adequately
characterized. Sampler type, frequency, and siting are further described in section 10.4.
10.2.2 QA Samplers
The purpose of collocated samplers and the performance evaluation is to estimate the precision of
the various systems samplers. The goal of the District is to have concentrations measured by a
sampler be within ±10% of the true concentration and that the precision have a coefficient of
variation less than 10% for each monitoring system. To estimate the level of precision being
achieved in the field, the NATTS site will operate collocated samplers for PMi0. The VOC and
Aldehyde samplers have duel channel configuration, which allows collocated canisters and DNPH
cartridges to be loaded and allowed to collect samples on the same instrument as the primary
sample. The QA samples will be set, run and collected on a 1 in 12 day schedule. Please see Table
11.1 for details on setup and recovery of collocated samples. Chapter 24 outlines the equations
that will be used to determine precision. There will be 2 analytes from each instrument that will be
used to determine the precision.
Field accuracy will be estimated using flow, temperature sensor and barometric checks.
Laboratory accuracy will be determined by the analysis of known reference analytes prepared by
independent laboratories submitted to the TCAPCD laboratory. If a sampler and laboratory
equipment are operating within the required precision and accuracy levels, then the decision maker
can proceed knowing that the decisions will be supported by unambiguous data. Thus the key
characteristics being measured with the QA samplers are precision.
10.3 Design Assumptions
The sampling design is based on the assumption that following the rules and guidance provided in
CFR and Network Design and Site Exposure Criteria for Selected Noncriteria Air Pollutants will
result in data that can be used to measure compliance with the national standards. The only issue
at Toxa City's discretion is the sampler siting, and to a degree, sampling frequency. The siting
assumes homogeneity of concentrations within the MSA. Boundaries will be regularly reviewed,
as part of the network reviews (Section 20). The basis for creating and revising the boundaries is
described in the following section.
-------
Project: Model QAPP
Element No: 10
Revision No: 1.0
Date: 11/04/02
Page 3 of6
10.4 Procedure for Locating and Selecting Environmental Samples
10.4.1 Sampling Design
The design of the air toxics network must achieve the monitoring objective. For the NATTS, the
objectives are stated in Statement 1 and for the rest of the ATMP, the objective is stated in
Statement 2.
1. Detect a percent difference change between successive three-year average
concentration levels that is greater than or equal to 15 percent
2. Determine the highest concentrations expected to occur in the area covered by the
network, i.e., to verify the spatial and temporal characteristics of HAPs within the city.
The procedure for siting the samplers to achieve the objective is based on judgmental sampling, as
is the case for most ambient air monitoring networks. Judgmental sampling uses data from
existing monitoring networks, knowledge of source emissions and population distribution, and
inference from analyses of meteorology to select optimal sampler locations. In addition, a
Geographic Information System (GIS) software package was also utilized to help locate the
samplers. Figure 10-1 illustrates the use of GIS for locating the samplers. Figure 10.1 shows that
the highest population in the area is in the northwest and just southeast of the bay. Between these
residential areas are the port facilities, power plants and the majority of the industrial sources.
This knowledge was used to locate the sampling areas. The exact locations are discussed in
Section 10.4.2.
10.4.2 Sampling Locations
Toxa City is situated in 2 counties: Hillsburg and Pine Lake. The boundaries were determined
based on (1) the 2000 census data by census tract, (2) the boundaries of the existing MSAs, and (3)
the surrounding geography. Figure 10-1 shows the population for which TCACPD is responsible.
According to the 2000 census, the Hillsburg County has a population of 834,054 while Pine Lake
county has a population of 851, 659. The population is evenly distributed through the MSA
except in the downtown area (see Figure 10-1). As can be seen from figure 10-1, the two counties
surround a coastal bay.
One site, TCI will be the upwind/background and will be located to quantify the background
concentrations. The siting of TCI is best suited for background concentrations. Site TC3 is located
on the bay near the industrial center. Again, this site well suited for highest concentration. This is
a middle scale monitoring station sited to capture maximum concentrations. Site TC4 is a
downwind neighborhood scale site.
-------
Project: Model QAPP
Element No: 10
Revision No: 1.0
Date: 11/04/02
Page 4 of6
10
10
2O Miles
Major Roadways
'opultation Density (persons/sq.mi.)
O - 1634
1635 -3354
3355 -5OO2
50O3 -6976
6977 -9658
9659 -1368O
13681 - 23338
23339 - 1O9932
Figure 10-1 Network Locations
-------
Project: Model QAPP
Element No: 10
Revision No: 1.0
Date: 11/04/02
Page 5 of6
The latitude/longitude coordinates for the five monitoring sites are listed in Table 10-2. It was
decided by the TCAPCD staff that since TC5 was a downwind/suburban monitoring location, it
would be selected as the NATTS.
Table 10.2 List of Collocated Samplers and Coordinates
Site Name
TCI
TC2
TC3
TC4
NATTS/TC5
Samplers Operated
VOC, PMio, Aldehydes
VOC, PM10
Aldehydes, VOC
VOC, PMio
VOC, PMio, Aldehydes,
Aethalometer
Collocated
VOC, PMio, Aldehydes
Coordinates
(Lat./Long.)
27.89/-82.80
28.12/-82.61
27.96/-82.S9
28.03/-82.16
27.71/-82.36
10.5 Classification of Measurements as Critical/Noncritical
The ambient concentration and site location data will be provided to AQS. The information
collected at collocated samplers is the same as that presented in Tables 6-1, 6-2, 6-3 and 6-4 for
primary samplers. All of the measurements in these tables are considered critical because it forms
the basis for estimating precision, which are critical for evaluating the ability of the decision
makers to make decisions at desired levels of confidence.
10.6 Validation of Any Non-Standard Measurements
At this time there are no NAAQS for the air toxics compounds, with the except for lead. Toxa
City is deploying and operating instruments according to descriptions in the applicable EPA
guidance documents.
References
1. Compendium Method for the Determination of Inorganic Compounds in Air, United States
Environmental Protection Agency, June 1999, Section IO-2.1.
2. Compendium Method for the Determination of Toxic Organic Compounds in Air, United States
Environmental Protection Agency, Section TO-11 A, January 1999
-------
Project: Model QAPP
Element No: 10
Revision No: 1.0
Date: 11/04/02
Page 6 of6
3. Compendium Method for the Determination of Toxic Organic Communes in Air, United States
Environmental Protection Agency, Section TO-15, January 1999
4. Network Design and Site Exposure Criteria For Selected Noncriteria Air Pollutants, EPA
Document Number, EPA 450/4-84-022, September 1984.
-------
Project: Model QAPP
Element No: 11
Revision No: 1.0
Date: 11/04/02
Page 1 of9
11.0 Sampling Methods Requirements
11.1 Purpose/Background
The methods described herein provide for measurement of the relative concentration of HAPs in
ambient air for a 24-hour sampling period. The method requirements for the Aethalometer are
described in Section 5.
Since three of the instruments are discreet samplers that collect HAPs inside or on a substrate,
there are three separate analytical techniques. General QA handling requirements are crucial for
sampling however the handling regimes are similar.
11.2 Sample Collection and Preparation
Sample preparation is an essential portion of the AMTP. The following functions are required for
sample preparation:
- filter receipt and inspection, filter numbering, conditioning and storage;
• VOC - cleaning, testing, verification and storage of canisters;
• Aldehydes - receipt and storage of DNPH cartridges in the laboratory refrigerator.
Sample set-up of the air toxics samplers in the Toxa City network takes place any day after the
previous sample has been recovered. For instance, on a Sunday - Thursday sample day set-up
when 1 in 6 day sampling is required, the pickup occurs the day after the run. However, on Friday
and Saturday run dates, the pick up is on the following Monday. It is important to recognize that
the only holding time that affects sample set-up is the 30 day window from the time a samples are
pre-weighed/processed to the time it is installed in the monitor. Since the NATTS has collocated
samplers, the second monitor will be set up to run at a sample frequency of 1 in 12 days; however,
sample set-up will take place on the same day as the primary sampler. Detailed sample set-up
procedures are available from the Toxa City SOPs.
11.2.2 Sample Recovery
Sample recovery of any individual sample from the air toxics instruments sampler in the Toxa City
network must occur within 72 hours of the end of the sample period for that sampler. For 1 in 6
day sampling this will normally be the day after a sample is taken. The next sample would also be
set-up at this time. See Table 11.1.
-------
Project: Model QAPP
Element No: 11
Revision No: 1.0
Date: 11/04/02
Page 2 of9
Table 11.1 Sample Set-up, Run and Recovery dates
Sample
Frequency
Iin6
Weekl
Iin6
Week 2
Iin6
Week 3
Iin6
Week 4
Iin6
WeekS
Iin6
Week 6
Sunday
Sample
Davl
Monday
Recovery
& Set-up
Recovery
& Set-up
Recovery
& Set-up
Sample
Day 7
Tuesday
Sample
Day 6
Collocated
Samples Run
Recovery &
Set-up
Wednesday
Sample
Day 5
Recovery &
Set-up
Thursday
Sample
Day 4
Collocated
Samples Run
Recovery &
Set-up
Friday
Sample
Day 3
Recovery
& Set-up
Saturday
Sample
Day 2
Collocated
Samples Run
11.3 Support Facilities for Sampling Methods
The main support facility for sampling is the sample trailer or shelter. At each sample location in
the Toxa City network there is a climate controlled sample trailer. The trailer has limited storage
space for items used in support of air toxic sampling. Table 11.2 lists the supplies that are stored at
each sample location trailer
-------
Project: Model QAPP
Element No: 11
Revision No: 1.0
Date: 11/04/02
Page 3 of9
Table 11.2 Supplies at Storage Trailers
Item
Powder Free Gloves
Fuses
Temperature standard
Flow rate standard
Sampler Operations Manual
Sampling SOPS
Flow rate verification filter
Tools
Filter Cassettes
Motor Brushes
Various 1/8'" and 1/4"
fittings
Pumps
Data Download Cable
Teflon end caps
aluminum foil
ice chests
Minimum
Quantity
Box
2
1
1
1 per model
1
2
1
1
1 set of 2
IBox
IBox
1
IBox
IBox
2
Notes
Material must be inert and powder free
Of the type specified in the sampler manual
In the range expected for this site andNIST
traceable
Calibrated from at least 15. 0 LPM to 18. 4
LPM and NIST Traceable
For PMio sampler
One Tool kit with various wrenches,
screwdrivers, etc..
For use with flow rate check filter or non-
permeable membrane
For PMio samplers
For Carbonyl and VOC samplers
For use with laptop computer
For capping the DNPH cartridges
For Carbonyl samplers
Spare ice chests for transporting samples
-------
Project: Model QAPP
Element No: 11
Revision No: 1.0
Date: 11/04/02
Page 4 of9
Since there are other items that the field operator may need during a site visit that are not expected
to be at each site, the operator is expected to bring these items with him/her.
11.4 Sampling/Measurement System Corrective Action
Corrective action measures in the ATMP will be taken to ensure the data quality objectives are
attained. There is the potential for many types of sampling and measurement system corrective
actions. Table 11.3 is an attempt to detail the expected problems and corrective actions needed for
a well-run network.
Table 11.3 Field Corrective Action
Item
Filter
Inspection
(Pre-sample)
Filter
Inspection
(Post- sample)
Flow rate
erratic
Problem
Pinhole(s) or torn
Torn or otherwise
suspect particulate
by-passing 46.2 mm
filter.
Heavy loading or
motor/motor brushes
are worn..
Action
1.) If additional filters have
been brought, use one of
them. Void filter with pinhole
or tear.
2.) Use new field blank filter
as sample filter.
3.) Obtain a new filter from
lab.
1.) Inspect area downstream
of where filter rests in sampler
and determine if particulate
has been by-passing filter.
2.) Inspect in-line filter before
sample pump and determine if
excessive loading has
occurred. Replace as
necessary.
Replace brushes or motor.
Re-calibrate
flowrate.
Notification
1.) Document on field
data sheet.
2.) Document on field
data sheet.
3.) Notify Field
Manager
1.) Document on field
data sheet.
2.) Document in log
book.
Document in log
book
-------
Project: Model QAPP
Element No: 11
Revision No: 1.0
Date: 11/04/02
Page 5 of9
Item
Problem
Action
Notification
Sample Flow
Rate
Verification
Out of Specification
{+ 10% of transfer
standard)
1.) Completely remove mass
flow controller and perform
flow rate check.
2.) Perform leak test.
3.) Check flow rate at 3 points
to determine if flow rate
problem is with zero bias or
slope.
4.) Re-calibrate flow rate
1.) Document on data
sheet.
2.) Document on data
sheet.
3.) Document on data
sheet. Notify Field
Manager
4.) Document on data
sheet. Notify Field
Manager.
Leak Test
VOC canisters will
not hold pressure.
1.) Replace fitting on nut on
sampler line.
2.) Inspect connections to the
mass flow controller and re-
perform leak test.
1.) Document in log
book.
2.) Document in log
book, notify Field
Manager, and flag
data since last
successful leak test.
Sample Flow
Rate
Consistently low
flows documented
during sample run
1.) Check programming of
sampler flowrate of
VOC/Carbonyl Sampler.
2.) Check flow with a flow
rate verification filter and
determine if actual flow is
low.
1.) Document in log
book.
2.) Document in log
book.
-------
Project: Model QAPP
Element No: 11
Revision No: 1.0
Date: 11/04/02
Page 6 of9
Item
Ambient
Temperature
Verification,
and Filter
Temperature
Verification.
Ambient
Pressure
Verification
Elapsed
Sample Time
Problem
Out of Specification
{+ 1'Cof standard)
Out of Specification
(±10 mm Hg)
Out of Specification
(10 min/day)
Action
l.)Make certain
thermocouples are immersed
in same liquid at same point
without touching sides or
bottom of container.
2.) Use ice bath or warm
water bath to check a different
temperature. If acceptable, re-
perform ambient temperature
verification.
3.) Connect new
thermocouple.
4.) Check ambient
temperature with another
NIST traceable thermometer.
1.) Make certain pressure
sensors are each 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
monitor's sensor. Pressure
correction may be required
3.) Connect new pressure
sensor
Check Programming, Verify
Power Outages
Notification
1.) Document on data
sheet.
2.) Document on data
sheet.
3.) Document on data
sheet. Notify Field
Manager.
4.) Document on data
sheet. Notify Field
Manager.
1.) Document on data
sheet.
2.) Document on data
sheet.
3.) Document on data
sheet. Notify Field
Manager
Notify Field Manager
-------
Project: Model QAPP
Element No: 11
Revision No: 1.0
Date: 11/04/02
Page 7 of9
Item
Elapsed
Sample Time
Power
Power
Data
Downloading
Problem
Sample did not run
Power Interruptions
LCD panel on, but
sample not working.
Data will not
transfer to laptop
computer or there is
no printout from the
Carbonyl/VOC
samplers
Action
1 .) Check Programming
2.) Try programming sample
run to start while operator is at
site. Use a flow verification
filter.
Check Line Voltage
Check circuit breaker, some
the VOC and Carbonyl
samplers have battery back-up
for data but will not work
without AC power.
Document key information on
sample data sheet. Make
certain problem is resolved
before data is written over in
sampler microprocessor.
Notification
1.) Document on data
sheet. Notify Field
Manager
2.) Document in log
book. Notify Field
Manager.
Notify Field Manager
Document in log
book
Notify Field
Manager.
In addition to these corrective actions, the samplers will also be calibrated: when installed, after
any major repairs, or when an audit flow rate shows that the samplers is outside of the +/- 10%
relative to the audit flow value.
11.5 Sampling Equipment, Preservation, and Holding Time Requirements
This sections details the requirements needed to prevent sample contamination, the volume of air
to be sampled, how to protect the sample from contamination, temperature preservation
requirements, and the permissible holding times to ensure against degradation of sample integrity.
11.5.1 Sample Contamination Prevention
The quality system has rigid requirements for preventing sample contamination. Powder free
gloves are worn while handling filter cassettes, canisters and DNPH cartridges. Filter and
cartridges are to be held in storage containers (static resistant zip lock bags) as provided by the
sampler manufacturer during transport to and from the laboratory. Once samples have been
analyzed they, are stored in static resistant zip lock bags.
-------
Project: Model QAPP
Element No: 11
Revision No: 1.0
Date: 11/04/02
Page 8 of9
11.5.2 Sample Volume
The volume of air to be sampled is specified in manufacturer's and the method specifications. The
different methods specify that certain minimum volumes must be collected Samples are expected
to be 24 hours, therefore the site operators must set the flow rates to collect sufficient sample to
obtain the minimum sample volume. In some cases a shorter sample period may occur due to
power outages. A valid sample run should not to be less than 23 hours. If the sample period is less
than 23 hours or greater than 25 hours, the sample will be flagged and the Branch Manager
notified.
11.5.3 Temperature Preservation Requirements
The temperature requirements of the samples vary between methods. During transport from the
laboratory to the sample location there are no specific requirements for temperature control with
the exception of DNPH cartridges. Filters will be located in their protective container and in the
transport container. Excessive heat must be avoided (e.g., do not leave in direct sunlight or a
closed-up car during summer). DNPH cartridges need to be stored at 4° C until they are loaded
into the sampler. The temperature requirements are detailed in Table 11.4.
Table 11.4 Temperature Requirements
Item
PM10 filter temperature
control during sampling and
until recovery.
DNPH Cartridge Filter
temperature control pre- and
post- sampling .
VOC canister Pre and post
sampling
Temperature Requirement
No requirements
4° C or less
No Requirements
Reference
TO-1 1 A Compendium
Section 9.4.3
-------
Project: Model QAPP
Element No: 11
Revision No: 1.0
Date: 11/04/02
Page 9 of9
11.5.4 Permissible Holding Times
The permissible holding times for the sample are clearly detailed in the attached appendices..
These holding times are provided in Table 11-5.
Table 11-5 Holding Times
Item
PMio filter
temperature
VOC canister
DNPH Cartridge
Filter
Holding
Time
No limits
<30 days
<30 days
From:
Completion
of sample
period
Sample end
date/time
To:
Time of
analysis
Date of Post
Weigh
Reference
TO- 15 Compendium
Section 9.4.2.1
TO- 11 Compendium
Section 11.1.1
-------
Project: Model QAPP
Element No: 12
Revision No: 1.0
Date: 11/04/02
Page 1 of3
12.0 Sampling Custody
12.1 Sample Custody Procedure
Sample custody and handling is one of the most important aspects of a quality system. The
TCAPCD tracks the whereabouts of each sample at each stage throughout the data collection
operation. Entries on Chain of Custody (CoC) forms will be made by hand and the information
will be entered into the sampling tracking system, where an electronic record will be kept. This
section will address sample custody procedures at the following stages:
• Pre-sampling;
• Post-sampling;
Sample receipt;
Sample archive.
One of the most important values in the sample custody procedure is the unique sample ID
number. The ID is an alpha-numeric value. The alpha values identify the type of sample (V for
VOC, D for DNPH or M for metals), a field blank (FB), a lab blank (LB) or collocated (C). The
next two values (YY) represent the last two digits of the calendar year and the next 4 digits
represent a unique date (MM/DD). Therefore, for 1998 the first routine filter will be numbered
M980101 for a metals filter and the collocated sample will be MC980101. The field blank for
the same day would be label MFB980101. The filter ID will be generated by the laboratory
analyst at the time of preparation of the sample.
12.1.1 Pre-Sampling Custody
The District's pre-sampling SOPs define how the samples will be conditioned, weighed, placed
into the protective shipping container, sealed with tape, and stored or refrigerated. See Table 11-
5 for details on sample holding. The Inventory Sheets containing the ID, Sample Type,
Container ID, and the Pre-sampling date will be attached to the field shelf for use by the site
operator. Each sampling period, the site operators will select samples that they will use for the
field. The number selected will depend on the time in the field prior to returning to the laboratory
and the number of samplers to be serviced. The site operator will perform the following Pre-
sampling activities:
• Contact Mr. Arcemont or Ms. Killion for access to laboratory.
• Put on appropriate laboratory attire.
Enter the filter storage area.
-------
Project: Model QAPP
Element No: 12
Revision No: 1.0
Date: 11/04/02
Page 2 of3
• Review the Inventory Sheet and select the next set of samples on the sheet. Ensure the
seals are intact. Since the site operator can not check the ID he will have to use the
container ID value.
• Take the Chain of Custody Records for each site visited. Fill out the columns of the "Pre-
Sampling Selection," portion of the Chain of Custody Record for each sample.
• Initial the column "Site Operator" on the Inventory Sheets to signify selection of the filters.
• Pack samples in sample coolers for travel to the field.
Upon arrival at a site:
• Select the appropriate samples.
Once the samples are installed at the site, complete the remainder of the columns of the
"Pre-Sampling Selection" portion of the Chain of Custody Record.
12.1.2 Post Sampling Custody
The field sampling SOPs specify the techniques for properly collecting and handling the sample
filters. Upon visiting the site:
• Select the appropriate Chain of Custody Records. Ensure that the filter ID is correct.
Remove the sample. Please refer to the SOPs for explicit details on unloading samples.
Briefly examine and place it into the protective container per SOPs and seal with tape.
• Place the protective container(s) into the shipping/transport container with the appropriate
temperature control devices.
• Record "Post Sampling Filter Recovery Information" on the Filter Chain of Custody
Record.
-------
Project: Model QAPP
Element No: 12
Revision No: 1.0
Date: 11/04/02
Page 3 of3
12.1.3 Sample Receipt
The samples, whether transported by the site operator or next day air, will be received by either
Janet Hoppert or David Bush at the Shipping/Receiving Office. The Shipping/Receiving Office
will:
Receive shipping/transport containers;
Upon receipt, open the containers to find Filter Chain of Custody Record?, or collect the
originals from the site operator (if delivered by operator).
Fill out the "Filter Receipt" area of the Filter Chain of Custody Records. Check sample
container seals.
If the samples are delivered on a weekday, follow sequence 5; if the samples are delivered on
a weekend, follow sequence 6.
Check the "Sent to Laboratory" column of the Filter Chain of Custody Records and transport
the filters to the appropriate laboratory room. Upon delivery to the laboratory, complete the
"Filter Transfer" area of the Filter Chain of Custody Records.
Store the samples in the refrigerator and check the "archived" column of the Filter Chain of
Custody Records. On the Monday of the following week, deliver the archived filters to the
laboratory and complete the "Filter Transfer" area of the Filter Chain of Custody Records.
12.1.4 Sample Archive
Once the analysis laboratory receives the filter, they will use their raw data entry sheets to log
the samples back in from receiving and prepare them for post-sampling weighing activities.
These activities are included in the analytical SOPs. The laboratory technicians will take the
filters out of the protective containers or folders and examine them for integrity, which will be
marked on the data entry sheets. During all post-sampling activities, filter custody will be the
responsibility of Mr. Arcemont. The samples will be stored within the laboratory freezer. The
laboratory has restricted access to Ms. Killion and Mr. Arcemont.
Upon completion of post-sampling weighing activities, the Filter Archiving Form will be used
by the laboratory technicians to archive the filter. Each filter will be packaged according to the
SOPs and stored in a box uniquely identified by Site ID and box number. Samples will be
archived in the filter storage facility for one year past the date of collection..
-------
Project: Model QAPP
Element No: 13
Revision No: 1.0
Date: 11/04/02
Page 1 of4
13.0 Analytical Methods Requirements
13.1 Purpose/Background
The methods stated here provide for gravimetric, spectrophotometric and chromatographic
analyses of samples collected in the Toxa City network. The basic methods used by the agency
are based on the Toxic Organic and Inorganic Compendia1'2'3'. These are listed in the Reference
area of this section.
13.2 Preparation of Samples
The Toxa City NATTS is at one location within a network consist of 5 sites. The primary
samplers will operate on a 1 in 6 day schedule. The collocated samplers are on a 1 in 12 day
schedule. Therefore, the approximate number of routine samples that have to be prepared, used,
transported, and conditioned is 3-6 per week. In addition, field blanks and lab blanks must also
be prepared. See the attached SOPs for activities associated with preparing pre-sample batches.
Upon delivery of approved sample media for use in the Toxa City network, the receipt is
documented and the pre-sampling media stored in the conditioning room/laboratory. Storing
samples in the laboratory makes it easier to maximize the amount of time available for
conditioning. Upon receipt, samples will be labeled with the date of receipt, opened one at a
time and used completely before opening another case. In the case of canisters, each canister
will be cleaned according to the cleaning procedures in the SOP. DNPH cartridges will be stored
in a refrigerator until taken into the field. All PMi0 filters in a lot will be used before a case
containing another lot is opened. When more than one case is available, the "First In - First Out"
rule will apply. This means that the first case of filters received is the first case that will be used.
13.3 Analysis Methods
13.3.1 Analytical Equipment and Method
The instruments used for analysis are listed in Table 13.1.
-------
Project: Model QAPP
Element No: 13
Revision No: 1.0
Date: 11/04/02
Page 2 of4
Table 13.1 Instruments Used in the Toxa City Laboratory
Parameter
Metals
Aldehydes
VOCs
Instrument
AnTech 3000
ZanTechSOOl
AnTech 3 001
Method
Inductively Coupled Plasma
High Pressure Liquid Chromatography
Gas Chromatography
Range
0.01 to 50 ug/m3
0.01 to25ppbv
0.001 to lOOppbv
13.3.2 Environmental Control
The Toxa City filter weigh room facility is an environmentally controlled room with temperature
and humidity control. Temperature is controlled at a minimum from 20 - 30° C. Humidity is
controlled from 30 - 40% relative humidity. Temperature and relative humidity are measured
and recorded continuously during equilibration. The balance is located on a vibration free table
and is protected from or located out of the path of any sources of drafts. Filters are conditioned
before both the pre- and post-sampling weighing. Filters must be conditioned for at least 24
hours to allow their weights to stabilize before being weighed. The areas used for preparation of
the canister, and aldehyde cartridge samples are clean laboratory benches in the main part of the
lab
13.4 Internal QC and Corrective Action for Measurement System
A QC notebook or database (with disk backups) will be maintained which will contain QC data,
including the calibrations, maintenance information, routine internal QC checks of mass
reference standards and laboratory and field or lab filter blanks, and external QA audits. It is a
requirement that QC charts be maintained on each instrument and included in their maintenance
notebooks. These charts may allow the discovery of excess drift that could signal an instrument
malfunction.
At the beginning of each analysis day, after the analyst has completed zeroing and calibrating the
instruments and measuring the working standard, analyze the laboratory filter blanks established
for the current samples to be analyzed.
Corrective action measures in the system will be taken to ensure good quality data. There is the
potential for many types of sampling and measurement system corrective actions. Each of the
SOPs outline exact actions that will be taken if the analytical systems are out of control.
-------
Project: Model QAPP
Element No: 13
Revision No: 1.0
Date: 11/04/02
Page 3 of4
13.5 Sample Contamination Prevention, Preservation, and Holding
13.5.1 Sample Contamination Prevention
The analytical support component of the network has rigid requirements for preventing sample
contamination. To minimize contamination, the sample media clean-up and sample preparation
rooms are separate from the instrumentation rooms. In addition, Heating and Ventilation system
is check annually by certified technicians. Hoods are also checked annually. PMi0 filters are
equilibrated/conditioned and stored in the same room where they are weighed. Powder free
gloves are worn while handling filters and filters are only contacted with the use of smooth non-
serrated forceps. Upon determination of its pre-sampling weight, the filter is placed in its filter
holding jacket for storage.
For VOC analytical method, the best prevention of contamination is not opening the canister in
the laboratory. All post sampling canisters that enter the laboratory should be under pressure
between 12-14 psig. With positive pressure, there is less likely that the sample will be
contaminated. However, care must be taken when the canisters are under vacuum and stored in
the laboratory. If there is a slight leak in the canister cap or valve, then laboratory air can enter
into the canister and contaminate the run.
For DNPH cartridges, the best prevention is to not take the cartridges out of the sealed shipping
packet until they are loaded into the sampler in the field. TCAPCD purchases the cartridges
from a chemical supply house with the DNPH coating already applied. Upon receipt and log-in,
the cartridges are immediately stored in a refrigerator within the sealed package. The field
technicians remove the cartridges (still in the sealed Mylar package) from the refrigerator and
log-out the samples. The samples are then refrigerated at the field monitoring site. When the
technician loads the samples into the aldehyde sampler, the DNPH cartridges are removed from
their Mylar package and installed.
13.5.2 Temperature Preservation Requirements
The temperature requirements of the laboratory and field situations are detailed in IO and TO
methods. In the weigh room laboratory, the PM10 filters must be conditioned for a minimum of
24 hours prior to pre-weighing; although, a longer period of conditioning may be required. The
weigh room laboratory temperature must be maintained between 20 and 30° C, with no more
than a +/- 5° C change over the 24 period prior to weighing the filters. During transport from the
weigh room to the sample location, there are no specific requirements for temperature control;
however, the filters will be located in their protective container and excessive heat avoided.
-------
Project: Model QAPP
Element No: 13
Revision No: 1.0
Date: 11/04/02
Page 4 of4
The specifics of temperature preservation requirements for VOC, and DNPH cartridges are
clearly detailed in Section 11 and the TO and IO methods1'2'3'. These requirements pertain to
both sample media before collection and both the sample media and sample after a sample has
been collected. Additionally, during the sample collection there are requirements for
temperature control. These are listed in Table 11.4.
13.5.4 Permissible Holding Times
The permissible holding times for the sample are clearly detailed in the TO and IO
Compendia1'2'3'. See Table 11.5.
References
1. Compendium Method for the Determination of Inorganic Compounds in Air, United States
Environmental Protection Agency, June 1999, Section IO-2.1.
2. Compendium Method for the Determination of Toxic Organic Compounds in Air, United
States Environmental Protection Agency, Section TO-11 A, January 1999
3. Compendium Method for the Determination of Toxic Organic Communes in Air, United
States Environmental Protection Agency, Section TO-14A, January 1999
-------
Project: Model QAPP
Element No: 14
Revision No: 1.0
Date: 11/04/02
Page 1 of 7
14.0 Quality Control Requirements
Quality Control (QC) is the overall system of technical activities that measures the attributes and
performance of a process. In the case of the ATMP, QC activities are used to ensure that
measurement uncertainty, as discussed in Section 7, is maintained within acceptance criteria for
the attainment of the DQO. Figure 14.1 represents a number of QC activities that help to
evaluate and control data quality for the program. Many of the activities in this figure are
implemented by the Air Division and are discussed in the appropriate sections of this QAPP.
14.1 QC Procedures
Day-to-day quality control is implemented through the use of various check samples or
instruments that are used for comparison. The measurement quality objectives table in Section 7
contains a complete listing of these QC samples as well as other requirements for the program.
The procedures for implementing the compounds collected are included in the field and
analytical methods section (Sections 11 and 13 respectively). The following information
provides some additional descriptions of these QC activities, how they will be used in the
evaluation process, and what corrective actions will be taken when they do not meet acceptance
criteria.
14.1.1 Calibrations
Calibration is the comparison of a measurement standard or instrument with another standard or
instrument to report, or eliminate by adjustment, any variation (deviation) in the accuracy of the
item being compared. The purpose of calibration is to minimize bias.
Calibration activities for air toxics samplers follow a two step process:
1. Certifying the calibration standard and/or transfer standard against an authoritative
standard, and;
2. Comparing the calibration standard and or transfer standard against the routine
sampling/analytical instruments.
Calibration requirements for the critical field and laboratory equipment are found in the
respective SOPs.
-------
Project: Model QAPP
Element No: 14
Revision No: 1.0
Date: 12/3/02
Page 2 of7
14.1.2 Blanks
Blank samples are used to determine contamination arising from principally four sources: the
environment from which the sample was collected/analyzed, the reagents used in the analysis,
the apparatus used, and the operator/analyst performing the analysis. Three types of blanks will
be implemented in the air toxics program:
Lot blanks - shipments of 8 x 11 inch filters will be periodically sent from the vendor to
TCAPCD. Each shipment must be tested to determine the length of time it takes the filters to
stabilize. Upon arrival of each shipment, 3 lot blanks will be randomly selected for the shipment
and be subjected to the conditioning/pre-sampling weighing procedures. The blanks will be
measured every 24 hours for a minimum of one week to determine the length of time to maintain
a stable weight reading.
Field blanks - provides an estimate of total measurement system contamination. By comparing
information from laboratory blanks against the field blanks, one can assess contamination from
field activities. Details of the use of the field blanks can be found in field SOPs. Field blanks
will be utilized for the aldehydes and metals. Field blanks cannot be utilized with the VOC
canisters since they arrive in the field under vacuum.
Lab blanks -provides an estimate of contamination occurring at the weighing/analysis facility.
Details of the use of the lab blanks can be found in can be found in SOPs. Lab blanks will be
utilized for the aldehydes, metals and VOCs. Lab blanks for VOCs are generated by the canister
cleaning system.
Blank Evaluation
The laboratory will include 3 field and 3 lab blanks into session batch. A batch is defined in
section 14.2. The following statistics will be generated for data evaluation purposes:
Difference for a single check (d) - The difference, J, for each check is calculated using
Equation 1, where X represents the concentration produced from the original weight and Y
represents the concentration reported for the duplicate weight (PM10/metals only).
D = I Y-X I
Percent Difference for a Single Check (
-------
Project: Model QAPP
Element No: 14
Revision No: 1.0
Date: 12/3/02
Page 3 of7
Yi-X x 100
(Y.-+XO/2
Mean difference for batch (dz) - The mean difference dz for both field and lab blanks within
an analysis batch, is calculated using equation 3 where d} through dn represent individual
differences (calculated from equation below) and n represents the number of blanks in the batch
dz —
Corrective action- The acceptance criteria for field blanks are discussed in the individual
SOPs. Field and lab blanks differences are determined by equation 1. However the mean
difference based upon the number of blanks in each batch will be used for comparison against
the acceptance criteria. If the mean difference of either the field or laboratory blanks is greater
than the accepted values in Table 14.1 then these will be noted in the QA report. For PMi0 filter,
the laboratory balance will be checked for proper operation. If the blank means of either the
field or lab blanks are still out of the acceptance criteria, all samples within the analysis session
will be flagged with the appropriate flag) and efforts will be made to determine the source of
contamination. In theory, field blanks should contain more contamination than laboratory
blanks. Therefore, if the field blanks are outside of the criteria while the lab blanks are
acceptable, analysis can continue on the next batch of samples while field contamination sources
are investigated. If the mean difference of the laboratory blanks is greater than the acceptance
criteria, the laboratory will stop until the issue is satisfactorily resolved. The laboratory
technician will alert the Laboratory Branch Manager and QA Officer of the problem. The
problem and solution will be reported and appropriately filed under response and corrective
action reports that will be summarized in the QA report.
Lab and field blanks will be control charted (see Section 14.3). The percent difference
calculation (equation 2) is used for control charting purposes and can be used to determine status.
-------
Project: Model QAPP
Element No: 14
Revision No: 1.0
Date: 12/3/02
Page 4 of7
14.1.3 Precision Checks
Precision is the measure of mutual agreement among individual measurements of the same
property, usually under prescribed similar conditions. In order to meet the data quality
objectives for precision, the District must ensure the entire measurement process is within
statistical control. Precision measurements will be obtained using collocated monitoring.
Evaluation of Collocated Data- All collocated data will be reported to AQS. The following
algorithms will be used to evaluate collocated data. Collocated measurement pairs are selected
for use in the precision calculations only when both measurements are within the acceptance
criteria. Please see Table 14.1.
Percent Difference for a collocated (Check (
-------
Project: Model QAPP
Element No: 14
Revision No: 1.0
Date: 12/3/02
Page 5 of7
the issue and may be asked to help find a common solution. The problem and solution will be
reported and appropriately filed under response and corrective action. This information will also
be included in the annual QA report.
Table 14.1 Precision Acceptance Criteria
Parameter
Both samples did not run 24 hours +/- 10 min.
One or both filters are damaged or exhibit a pinhole or tear
One or both samplers has erratic flow pattern
The difference in the pressure of the VOC canisters is > 2 psig
One or both samples are not kept within the holding and storage temperature
requirements for any length of time
Decision
Do not accept
Do not accept
Do not accept
Do not accept
Do not accept
14.1.4 Accuracy Checks
Accuracy is defined as the degree of agreement between an observed value and an accepted
reference value and includes a combination of random error (precision). Three accuracy checks
are implemented in the air toxics monitoring program:
• Flow rate audits;
Balance checks, and
Laboratory audits.
Flow Rate Audits
The flow rate audit is made by measuring the field instrument's normal operating flow rate using
a certified flow rate transfer standard. The flow rate standard used for auditing will not be the
same flow rate standard used to calibrate the analyzer. However, both the calibration standard
and the audit standard may be referenced to the same primary flow rate or volume standard.
Report the audit (actual) flow rate and the corresponding flow rate indicated or assumed by the
sampler. The procedures used to calculate measurement uncertainty are described below.
Accuracy of a Single Sampler - Single Check (Quarterly) Basis (dj). The percentage
difference (4) for a single flow rate audit / is calculated using the following equation, where Xt
represents the audit standard flow rate (known) and Yt represents the indicated flow rate.
-------
Project: Model QAPP
Element No: 14
Revision No: 1.0
Date: 12/3/02
Page 6 of7
Yi-X x 100
"
Balance Checks- Balance checks are frequent checks of the balance working standards (100 and
200 mg standards) against the balance to ensure that the balance is within acceptance criteria
throughout the pre- and post-sampling weighing sessions. Toxa City will use ASTM class 1
weights for its primary and secondary (working) standards. Both working standards will be used
measured at the beginning and end of the sample batch. Balance check samples will be
controlled charted.
Balance Check Evaluation- The following algorithm will be used to evaluate the balance
checks
Difference for a single check (dy) - The difference, dy, for each check is calculated using the
following equation, where X represents the certified mass weight and /represents the reported
weight.
dv = Y - X
Corrective Action - The difference among the reported weight and the certified weight must be
< 5 mg. Since this is the first check before any pre-or post-sampling weighing, if the acceptance
criteria is not met, corrective action will be initiated. Corrective action may be as simple as
allowing the balance to perform internal calibrations or to sufficiently warm-up, which may
require checking the balance weights a number of times. If the acceptance criteria are still not
met, the laboratory technician will be required to verify the working standards to the primary
standards. Finally, if it is established that the balance does not meet acceptance criteria for both
the working and primary standards, and other trouble shooting techniques fail, the Libra Balance
Company service technician will be called to perform corrective action.
If the balance check fails acceptance criteria during a run, the 10 filters weighed prior to the
failure will be rerun. If the balance check continues to fail, trouble-shooting, as discussed above,
will be initiated. The values of the 10 samples weighed prior to the failure will be recorded and
flagged, but will be remain with the un-weighed samples in the batch to be reweighed when the
balance meets the acceptance criteria. The data acquisition system will flag any balance check
outside the acceptance criteria. The samples that were flagged will be un-flagged once the
balance comes into compliance with the QC procedure.
-------
Project: Model QAPP
Element No: 14
Revision No: 1.0
Date: 12/3/02
Page 7 of7
Accuracy of a Laboratory Audit - Single Check (Annual) Basis (
-------
Project: Model QAPP
Element No: 15
Revision No: 1.0
Date: 11/04/02
Page 1 of7
15.0 Instrument/Equipment Testing, Inspection, and Maintenance
15.1 Purpose/Background
The purpose of this element in the Toxa City QAPP is to discuss the procedures used to verify
that all instruments and equipment are maintained in sound operating condition and are capable
of operating at acceptable performance levels.
15.2 Testing
All samplers used in the Toxa City ATMP will be similar to the instruments described in the TO
and IO Compendia. Therefore, they are assumed to be of sufficient quality for the data
collection operation. Prior to field installation, Toxa City will assemble and run the samplers at
the laboratory facilities. The field operators will perform external and internal leak checks and
temperature, pressure and flow rate verification checks. If any of these checks are out of
specification, the field technicians will attempt to correct them. If the problem is beyond their
expertise, the division director will contact the vendor for guidance. If the vendor does not
provide sufficient support, then the instrument will be returned to the vendor. Once installed at
the site, the field operators will run the tests at least one more time. If the sampling instrument
meets the acceptance criteria, it will be assumed to be operating properly.
15.3 Inspection
Inspection of various equipment and components are provided here. Inspections are subdivided
into two sections: one pertaining to laboratory issues and one associated with field activities.
15.3.1 Inspection in Laboratory
There are several items that need routine inspection in the laboratory. Table 15-1 details the
items to inspect and how to appropriately document the inspection. All of the different areas of
the laboratory (PMW mass weight, Gas Chromatography/Mass Spec, and the ICP rooms) will be
maintained according to Table 15.1.
-------
Project: Model QAPP
Element No: 15
Revision No :1
Date: 12/3/02
Page 2 of7
Table 15.1 Inspections in the Laboratory
Item
Weighing
Room
Temperature
Weighing Room
Humidity
Weighing
Room
Cleanliness
GC/MC Room
Temperature
GC/MS
Cleanliness
ICP
Temperature
ICP
Cleanliness
HPLC Room
Temperature
HPLC
Cleanliness
Extract ion
Room
Inspection
Frequency
Daily
Daily
Monthly
Daily
Monthly
Daily
Monthly
Daily
Monthly
Weekly
Inspection
Parameter
20 - 30° C
30 - 40° RH
Use glove and
visually inspect
20 - 30° C
Use glove and
visually inspect
20 - 30° C
Use glove and
visually inspect
20 - 30° C
Use glove and
visually inspect
Use glove and
visually inspect
Action if Item Fails Inspection
1.) Check HVAC System
2.) Call service provider that
holds maintenance agreement
1.) Check HVAC System
2.) Call service provider that
holds maintenance agreement
Clean room
1.) Check HVAC System
2.) Call service provider that
holds maintenance agreement
Clean room and remove clutter
put canisters back into rack
1.) Check HVAC System
2.) Call service provider that
holds maintenance agreement
Clean room and remove clutter
store and clean vial. Discard old
filters
1.) Check HVAC System
2.) Call service provider that
holds maintenance agreement
Clean room and store cartridges
Thoroughly clean room and
remove all materials. Clean all
removal instrument and
autoclave
Documentation
Requirement
1.) Document in log book
2.) Notify Lab Manager
1.) Document in log book
2.) Notify Lab Manager
Document in Log Book
Document in Logbook
Document in Log Book
Document in Logbook
Document in Log Book
Document in Logbook
Document in Log Book
Document in Log Book
-------
Project: Model QAPP
Element No: 15
Revision No :1
Date: 12/3/02
Page 3 of7
15.3.2 Inspection of Field Items
There are several items to inspect in the field before and after a sample has been taken. The
attached appendices discuss in detail the items that need to be inspected. Please refer to the
attached SOPs.
15.4 Maintenance
There are many items that need maintenance attention in the network. This section describes the
laboratory and field items.
15.4.1 Laboratory Maintenance Items
The successful execution of a preventive maintenance program for the laboratory will go a long
way towards the success of the entire program. In the Toxa City network, laboratory preventive
maintenance is handled through the use of several contractors. The Smith and Jones HVAC
Company has a contract to take care of all preventive maintenance associated with the heating,
ventilation, and air conditioning system (HVAC). In addition to these contracts, the TCAPCD
also hires LabTech Inc. to perform the maintenance on the ICP, GC/MS and the Liquid
Chromatograph. The Smith and Jones HVAC Company can be paged for all emergencies
pertaining to the laboratory HVAC system. Preventive maintenance for the micro-balance is
performed by the Libra Balance Company service technician. Preventive maintenance for the all
analytical instruments is scheduled to occur at initial set-up and every 6 months thereafter. In the
event that there is a problem with the analytical instruments that cannot be resolved within the
Toxa City organization, the Libra Balance Company and LabTech Inc. service technician can be
paged. The District's service agreement with Libra Balance Company and LabTech Inc. calls
for service within 24 hours. The service technician will also have a working micro-balance in
his/her possession that will be loaned to Toxa City in the case that the District's micro-balance
can not be repaired on-site. In the event one of the other analytical instruments fail, the service
technicians for the vendors will visit the TCAPCD laboratory and ascertain the problem. The
parts will be shipped and replaced as soon as possible.
Service agreements with both the Smith and Jones HVAC Company, Libra Balance Company
and LabTech Inc. are expected to be renewed each year. In the event either companies service
agreement is not renewed, a new service provider will be selected and contract put in place. The
following tables details the maintenance items, how frequently they will be replaced, and who
will be responsible for performing the maintenance.
-------
Project: Model QAPP
Element No: 15
Revision No :1
Date: 12/3/02
Page 4 of7
Table 15.2 Preventive Maintenance in Weigh Room Laboratories
Item
Multi- point Micro-balance maintenance
Calibration
Comparison of NIST Standards to laboratory working
and primary standards
Verify Humidity and Temperature sensors
HEPA filter replacement
HVAC system preventive maintenance
Computer Back-up
Computer Virus Check
Computer system preventive maintenance (clean out
old files, compress hardrive, inspect)
Maintenance Frequency
6 Months
6 Months
Monthly
Monthly
Yearly
Weekly
Weekly
Yearly
Responsible Party
Libra Balance Company
Libra Balance Company
Balance Analyst
Balance Analyst
Smith and Jones HVAC
Lab Analyst
Lab Analyst
PC support personnel
Table 15.3 Preventive Maintenance in VOC Laboratories
Item
Multi- point maintenance calibration
Comparison of NIST Standards to laboratory
working and primary standards
Filament Replacement
Carrier gas scrubber replaced
MS Quadruples or ion source cleaned
RF Generator Replaced
Test lines for pressure integrity
Replace Traps
Computer Back-up
Computer Virus Check
Computer system preventive maintenance
(clean out old files, compress hardrive,
inspect)
Maintenance Frequency
6 Months or after initial setup, after
maintenance or repair, after column is replaced
Weekly
As necessary
When trap color indicates
Every 3 months
As needed
Annually
as needed
Weekly
Weekly
Yearly
Responsible Party
Lab Analyst
Lab Analyst
Lab Analyst
Lab Analyst
Lab Analyst
Lab Analyst
Lab Analyst
Lab Analyst
Lab Analyst
Lab Analyst
PC support
personnel
-------
Project: Model QAPP
Element No: 15
Revision No :1
Date: 12/3/02
Page 5 of7
Table 15.4 Preventive Maintenance in Liquid Chromatography Laboratory
Item
Multi- point maintenance calibration
Comparison of NIST Standards to laboratory working
and primary standards
Replace Chromatography Column
Replace delivery system motor
Change Column guard
Replace Teflon delivery tubing
Test Acetonitrile used for sample extraction
Computer Back-up
Computer Virus Check
Computer system preventive maintenance (clean out
old files, compress hardrive, inspect)
Maintenance Frequency
6 Months
6 Months
As needed
2 years
As needed
Yearly
Monthly
Weekly
Weekly
Yearly
Responsible Party
LabTech Inc.
Lab Analyst
Lab Analyst
LabTech Inc.
Lab Analyst
Lab Analyst
Lab Analyst
Lab Analyst
Lab Analyst
PC support personnel
Table 15.5 Preventive Maintenance in Inductively Coui
Item
Instrument Tuning
Torch and Spray chambers cleaned
Multi- point maintenance calibration
Comparison of NIST Standards to laboratory working
and primary standards
Clean Oven
Plasma Generator
Heat Generator
Computer Back-up
Computer Virus Check
Computer system preventive maintenance (clean out
old files, compress hardrive, inspect)
3led Plasma Laboratories
Maintenance Frequency
Initial Setup
3 months
6 Months
Monthly
Monthly
Monthly
Yearly
Weekly
Weekly
Yearly
Responsible Party
LabTech Inc.
Lab Analyst.
LabTech Inc.
Lab Analyst
Lab Analyst
Lab Analyst
LabTech Inc
Balance Analyst
Balance Analyst
PC support personnel
-------
Project: Model QAPP
Element No: 15
Revision No :1
Date: 12/3/02
Page 6 of7
15.4.2 Field Maintenance Items
There are many items associated with appropriate preventive maintenance of a successful field
program. Please see Table 15.6 details the appropriate maintenance checks of the samplers and
their frequency.
Table 15.6 Preventive Maintenance on Field Instruments
Instrument
PMjo sampler
Aethalometer
VOC Sampler
Item
Motor Brush replacement
Clean inside of sampler
Replace Motor
Replace Motor
Replace Motor gaskets
Filter screen inspected for impacted
deposits or bits of filter
Check connecting tube and power
lines for holes, crimps or cracks
Replace quartz tape
Clean Pmfine inlet
Replace pump
Replace sample lines
Clean flow controller
Maintenance
Frequency
3 Months
6 Months
Annually
Annually
When motor is
replaced
Annually
6 months
3 Months
3 Months
As needed
Annually
Annually
Responsible Party
Field Technician
Field Technician
Field Technician
Field Technician
Field Technician
Field Technician
Field Technician
Field Technician
Field Technician
Field Technician
Senior Field Technician
Senior Field Technician
-------
Project: Model QAPP
Element No: 15
Revision No :1
Date: 12/3/02
Page 7 of7
Aldehyde Sampler
Replace 1/8" connectors
Cartridge connectors
Replace Motor Brushes
Fan motor replacement
Clean inside of sampler
Annually
Annually
Annually
2 years
6 Months
Field Technician
Field Technician
Field Technician
Field Technician
Field Technician
-------
Project: Model QAPP
Element No: 16
Revision No :1
Date: 11/04/02
Page 1 of8
16.0 Instrument Calibration and Frequency
16.1 Instrumentation Requiring Calibration
16.1.1 Analysis of Instruments - Laboratory
The laboratory support for Toxa City includes calibration. As indicated in Section 13, the
instruments are calibrated using NIST traceable standards (if available) once a year under a
service agreement. For the Libra 101, the service technician performs routine maintenance and
makes any balance response adjustments that the calibration shows to be necessary. During the
visit by the service technician, both the in-house primary and secondary (working) standards are
checked against the service technician's standards to ensure acceptability. All of these actions are
documented in the service technician's report, a copy of which is provided to the laboratory
manager, which after review, is appropriately filed.
The laboratory also maintains a set of standards for each of the laboratory systems. Please see
Table 16.1. Below are brief statements on how these Calibrations are performed.
• For the Libra 101, the technician uses 3 Class A weights to verify that the balance is weighing
within the tolerance limits. Once this is performed, the balance is tarred. Filters are weighed
in batches of 10 samples. After a sample batch has been weighed, the technician re-weighs
on filter (duplicate weight) and re-tares the balance. At the end of the day (or end of the
weighing session) the technician reweighs the 3 Class A weights. Any difference in weight
is noted.
• For the Gas Chromatographs, the NIST Traceable cylinder is attached to a mass flow control
calibration unit. The concentrations of benzene, propane and methylene chloride are blended
down to a value which will be in the higher 80% of the range of compounds found in ambient
concentrations. This usually is ~ 20 ppbv. The Gas Chromatographs is allowed to reach
operating conditions. The gas from the mass flow controller is injected into the system and
the carrier helium is allowed to flow. Once the calibration gas is allowed to enter, two peaks
should appear. The mass flow controller is then adjusted to allow the gas concentration to be
~ 40%. This process is then repeated with a concentration of 20% of range of compounds.
Zero air is then generated and a baseline is determined. The system is now ready to accept
ambient concentrations. After the day's batches are run, a single point (80%) is injected into
the GC.
• After the Inductively Coupled Plasma unit is allowed to come to operating conditions, a
standard solution of metals is injected into the ICP. The responses are noted. Distilled ion-
free water is then injected into the ICP. This allows the system to reach a baseline.
• For the Liquid Chromatographs, the procedure is the same, with the exception of the
compounds injected. 2,4 Dinitro-phenylhydrazine is dissolved in ultra-pure Acetonitrile.
-------
Project: Model QAPP
Element No: 16
Revision No :1
Date: 11/04/02
Page 2 of8
These become the standard solutions. After the LCs have come to operating conditions,
ultra-pure Acetonitrile is injected. This allows the system to reach a baseline. Different
concentration at 80% of the normal ambient concentrations of DNPH in Acetonitrile are
injected into the LC. Response peaks are observed and recorded. This procedure is repeated
at the end of the analysis batch run.
Table 16.1 Lab Instruments Standards
Manufacturer
Libra 101
(filter weights
AntechSOOO
(metals)
ZanTechSOOl
(Aldehydes)
Antech 3001
(VOCs)
Instrument
Balance
Inductively Coupled
Plasma
Liquid
Chromatographs
Gas Chromatography
Type of
Standard
Class A Weights
High Purity
Reagents - High
Purity grade
Standards
High Purity 2,4
Dinitro
phenylhydrazine
crystals
dissolved in
Acetonitrile
Compressed Gas
Cylinder
Frequency
1 every 10 samples
Before and after
each batch run
Before and after
each batch run
Before and after
each batch run
NIST Traceability
Class A Weights
99.99% pure ultra
high grade
Standard solutions
Reagent grade
available from
Chemical vendor
Benzene,
Methylene
Chloride are
NIST Traceable
through vendor
16.1.2 Flow Rate - Laboratory
Laboratory technicians perform the comparison of the flow rate transfer standard to a NIST-
traceable primary flow rate standard and once every year sends the primary standard to NIST for
recertification. The laboratory and field personnel chose an automatic dry-piston flow meter for
field calibrations and flow rate verifications of the flow rates of the network samplers. This type
of device has the advantage of providing volumetric flow rate values directly, without requiring
conversion from mass flow measurements, temperature, pressure, or water vapor corrections. In
addition, the manual bubble flowmeter will be used in the lab as a primary standard and as a
backup to the dry-piston flowmeter, where the absence of wind and relatively low humidity will
have less negative effect on flowmeter performance.
-------
Project: Model QAPP
Element No: 16
Revision No :1
Date: 11/04/02
Page 3 of8
Upon initial receipt of any new, repaired, or replaced air toxics sampler, a field technician will
perform multi-point flow rate calibration verification on the sampler flow rate to determine if
initial performance is acceptable. Once sampler flow rate is accepted, the lab performs the
calibration and verifications at the frequency specified in Section 14, as well as directly
performing or arranging to have another party perform the tests needed to recertify the
organizations standards.
16.1.3 Sampler Temperature, Pressure, Time Sensors - Laboratory
The lab arranges support for the field calibration of temperature and pressure sensors by
acquiring the necessary equipment and consumables, preparing and lab testing the temperature
comparison apparatus. A stationary mercury manometer in the laboratory is used as a primary
standard to calibrate the two electronic aneroid barometers that go out in the field as transfer
standards.
16.1.4 Field
The following calibrations are performed in the field:
• calibration of volumetric flow rate meter of each samplers against the working standard;
• calibration of sampler temperature and pressure sensors against the working temperature
standard (VOC and Aldehyde Samplers only);
• calibration of the min/max thermometers, normally located in the coolers in which DNPH
cartridges are transported to and from the sampler in the field, against the laboratory-checked
working standard thermometer;
• Check the background using the fine mesh screen on the Aethalometer.
16.2 Calibration Method
16.2.1 Laboratory - Gravimetric (Mass) Calibration
The calibration and QC (verification) checks of the microbalance are addressed in Sections
16.1.1 and 13.3 of this QAPP. For the following 3 reasons, the multipoint calibration for this
method will be zero, 100 and 200mg: 1) the required sample collection filters weigh between 100
and 200 mg; 2) the anticipated range of sample loadings for the 24 hour sample period is rarely
going to be more than a few 100 mgs; and 3) the lowest, commercially available check weights
that are certified according to nationally accepted standards are only in the single milligram
range. Since the critical weight is not the absolute unloaded or loaded filter weight, but the
difference between the two, the lack of microgram standard check weights is not considered
-------
Project: Model QAPP
Element No: 16
Revision No :1
Date: 11/04/02
Page 4 of8
cause for concern about data quality, as long as proper weighing procedure precautions are taken
for controlling contamination, or other sources of mass variation in the procedure.
16.2.2 Laboratory/ Field - Flow Calibration.
The Air Monitoring and Laboratory Branch Managers conduct spot checks of lab and field
notebooks to ensure that the lab and field personnel are following the SOPs, including the
QA/QC checks, acceptance criteria and frequencies.
Method Summary: After equilibrating the calibration device to the ambient conditions, connect
the flow calibration device on the sampler down tube or filter holding device. If the sampler has
not been calibrated before, or if the previous calibration was not acceptable, perform a leak
check according to the manufacturer's operational instruction manual, which is incorporated into
Toxa City ATMP SOPs.
Otherwise, place the sampler in calibration or "run" mode and perform a one-point
calibration/verification or a one-point flow rate verification. The field staff will only perform a
leak check after calibration or verification of are outside of the acceptance criteria.
Following the calibration or verification, turn off the sampler pump, remove the filter, cartridge,
and remove the flow calibration device, (and flow adaptor device if applicable), and replace the
sampler inlet or hood. If the flow rate is determined to be outside of the required target flow rate,
attempt to determine possible causes by minor diagnostic and trouble shooting techniques (e.g.,
leak checks), including those listed in the manufacturer's operating instruction manual.
16.2.3 Sampler Pressure Calibration Procedure
General: According to ASTM Standard D 3631 (ASTM 1977), a barometer can be calibrated by
comparing it with a secondary standard traceable to a NIST primary standard.
Precautionary Note: Protect all barometers from violent mechanical shock and sudden changes
in pressure. A barometer subjected to either of these events must be recalibrated. Maintain the
vertical and horizontal temperature gradients across the instruments at less than 0.1* C/m. Locate
the instrument so as to avoid direct sunlight, drafts, and vibration.
A Fortin mercury type of barometer is used in the laboratory to calibrate and verify the
aneroid barometer used in the field to verify the barometric sensors of samplers. Details are
provided in the appropriate SOP.
-------
Project: Model QAPP
Element No: 16
Revision No :1
Date: 11/04/02
Page 5 of8
16.3 Calibration Standard Materials and Apparatus
Table 16.2 presents a summary of the specific standard materials and apparatus used in
calibrating measurement systems.
Table 16.2 Standard Materials and/or Apparatus for Air Toxics Calibration
Parameter
M-Material
A=Apparatus
MassM
Temperature
M+A
M+A
Pressure
M+A
A
Flow Rate
A
A
A
Std. Material
Class A wgts
Hg
NA
Hg
NA
NA
Std. Apparatus
NA
Thermometer
Thermistor
Fortin
Aneroid
Piston Meter
Bubble Meter
High Volume Flow
Mfr. Name
ScalesTech. Inc.
Hot Water Inc.
True Temp.
You Better...
Aviators Choice
Flowtech Inc.
SaapTech. Inc
Top Hat Inc..
Model #
111
5500
8910
22
7-11
F199
LG88
TP-1
Frequency
of Calibration
NA
NA
Annually
NA
Quarterly
Annually
NA
Annually
Flow Rate
The flow rate standard apparatus used for flow-rate calibration (field- NIST-traceable, piston-
type volumetric flow rate meter; laboratory -NIST-traceable manual soap bubble flow meter and
time monitor) has its own certification and is traceable to other standards for volume or flow rate
which are themselves NIST-traceable. A calibration relationship for the flow-rate standard, such
as an equation, curve, or family of curves, is established by the manufacturer (and verified if
needed) that is accurate to within 2% over the expected range of ambient temperatures and
pressures at which the flow-rate standard is used. The flow rate standard will be recalibrated and
recertified at least annually.
The actual frequency with which this recertification process must be completed depends on the
type of flow rate standard- some are much more likely to be stable than others. The Division
will maintain a control chart (a running plot of the difference or percent difference between the
flow-rate standard and the NIST-traceable primary flow-rate or volume standard) for all
comparisons. In addition to providing excellent documentation of the certification of the
standard, a control chart also gives a good indication of the stability of the standard. If the two
standard-deviation control limits are close together, the chart indicates that the standard is very
stable and could be certified less frequently. The minimum recertification frequency is 1 year.
-------
Project: Model QAPP
Element No: 16
Revision No :1
Date: 11/04/02
Page 6 of8
On the other hand, if the limits are wide, the chart would indicate a less stable standard that will
be recertified more often.
The High Volume sampler flow rate device is a Top Hat Inc., TP-1, which is certified to a NIST
traceable Roots meter. The High Volume orifice is sent to the State's certification laboratory on
an annual basis to verify its flow rate.
Temperature
The operations manuals associated with the TCAPCD samplers identify types of temperature
standards recommended for calibration and provide a detailed calibration procedure for each type
that is specifically designed for the particular sampler.
The EPA Quality Assurance Handbook, Volume IV (EPA 1995), Section 4.3.5.1, provides
information on calibration equipment and methods for assessing response characteristics of
temperature sensors.
The temperature standard used for temperature calibration will have its own certification and be
traceable to a NIST primary standard. A calibration relationship to the temperature standard (an
equation or a curve) will be established that is accurate to within 2% over the expected range of
ambient temperatures at which the temperature standard is to be used. The temperature standard
must be re-verified and recertified at least annually. The actual frequency of recertification
depends on the type of temperature standard; some are much more stable than others. The
Division will use NIST-traceable mercury in glass thermometer, for laboratory calibration and
certification of the field thermistor.
The temperature sensor standards chosen by the lab and field staff and managers are both based
on standard materials contained in standardized apparatus; each has been standardized
(compared in a strictly controlled procedure) against temperature standards the manufacturers
obtained from NIST.
The TCAPCD laboratory standards are 2 NIST-traceable mercury-in-glass thermometers from
the Hot Water Inc\ each with its own certificate summarizing the company's NIST traceability
protocol and documenting the technician's signature, comparison date, identification of the NIST
standard used, and the mean and standard deviation of the comparison results. There are 2
thermometers with overlapping ranges that span the complete range of typically measured
summer to winter lab and field temperature values.
There are two TCAPCD field temperature standards True Temp.8910 ® thermistor probes and
one digital readout module with RS232C jack and cable connector available for linkage to a data
-------
Project: Model QAPP
Element No: 16
Revision No :1
Date: 11/04/02
Page 7 of8
logger or portable computer. The two probes have different optimum ranges, one including the
full range of temperatures ever recorded in the summer and the other including the full range of
temperatures ever recorded in the winter by the National Weather Service at the Toxa City sites.
Each probe came with a certificate of NIST-traceability with the same kind of information as the
thermometer certificates contained.
Pressure
The Fortin mercurial type of barometer works on fundamental principles of length and mass and
is therefore more accurate but more difficult to read and correct than other types. By comparison,
the precision aneroid barometer is an evacuated capsule with a flexible bellows coupled through
mechanical, electrical, or optical linkage to an indicator. It is potentially less accurate than the
Fortin type but can be transported with less risk to the reliability of its measurements and
presents no damage from mercury spills. The Fortin type of barometer is best employed as a
higher quality laboratory standard which is used to adjust and certify an aneroid barometer in the
laboratory. The Toxa City pressure standard is a You Better Believe It® Model 22 Fortin-type
mercury barometer. The field working standard is an Aviator's Choice® 7-11 aneroid barometer
with digital readout.
16.4 Calibration Frequency
See Table 16-1 for a summary of Primary and Working Standards QC checks that includes
frequency and acceptance criteria and references for calibration and verification tests. All of
these events, as well as sampler and calibration equipment maintenance will be documented in
field data records and notebooks and annotated with the flags. Laboratory and field activities
associated with equipment used by the respective technical staff will be kept in record notebooks
as well. The records will normally be controlled by the Branch Managers, and located in the labs
or field sites when in use or at the manager's offices when being reviewed or used for data
validation.
References
1. ASTM. 1977. Standard test methods for measuring surface atmospheric pressure. American Society for
Testing and Materials. Philadelphia, PA. Standard D 3631-84.
2. ASTM. 1995. Standard test methods for measuring surface atmospheric pressure. American Society for
Testing and Materials. Publication number ASTM D3631-95.
3. EPA. 1995. Quality Assurance Handbook for Air Pollution Measurement Systems Volume IV:
Meteorological Measurements. U.S. Environmental Protection Agency. Document No. EPA/600/R-
94/038d. Revised March.
-------
Project: Model QAPP
Element No: 16
Revision No :1
Date: 11/04/02
Page 8 of8
4. NIST. 1976. Liquid-in-glass thermometry. National Institute of Standards and Technology. NBS
Monograph 150. January .NIST. 1986. Thermometer calibration: a model for state calibration laboratories.
National Institute of Standards and Technology. NBS Monograph 174. January.
5. NIST. 1988. Liquid-in-glass thermometer calibration service. National Institute of Standards and
Technology. Special publication 250-23. September.
6. NIST. 1989. The calibration of thermocouples and thermocouple materials. National Institute of Standards
and Technology. Special publication 250-35. April 1989
-------
Project: Model QAPP
Element No: 17
Revision No :1
Date: 11/04/02
Page 1 of3
17.0 Inspection/Acceptance of Supplies and Consumables
17.1 Purpose
The purpose of this element is to establish and document a system for inspecting and accepting
all supplies and consumables that may directly or indirectly affect the quality of the Program.
The Toxa City Air Toxics Monitoring Network relies on various supplies and consumables that
are critical to its operation. By having documented inspection and acceptance criteria and
consistency of the supplies can be assured. This section details the supplies/consumables, their
acceptance criteria, and the required documentation for tracking this process.
17.2 Critical Supplies and Consumables
Table 17.1 details the various components for the laboratory and field operations.
Table 17.1 Critical Field Supplies and Consumables
Area
PMjo Sampler
PM10 Sampler
PM10 Sampler
VOC Sampler
VOC Sampler
Aldehyde
Sampler
Aldehyde
Sampler
Aldehyde
Sampler
Aldehyde
Sampler
Aethalometer
Aethalometer
Item
8x 11" Quartz filters
High Volume Motor
Motor Brushes
Stainless Steel tubing
Mass Flow Controller
DNPH cartridges
Fuses
Mass Flow Controller
Motor
Quartz tape
Pump
Description
Quartz filter
20 amp. Blower motor
Carbon Brush
Elements
Clean SS tubing
0- 50 cc/min.
DNPH coated plastic
Cartridges
hi sampler
0-100 cc/min
0-200 cc/min
Quartz Tape
Pump
Vendor
FilterTech Inc.
XYZ Company
XYZ Company
Steeltech
Flowtech Inc.
CartTech Inc.
FuseTech Inc.
Flowtech Inc.
Flowtech Inc.
Magee Scientific
Magee Scientific
Model Number
NA
X300
X301
X3301
FL100
D100
F100
F1101
-------
Project: Model QAPP
Element No: 17
Revision No :1
Date: 11/04/02
Page 2 of3
Table 17.2 Critical Laboratory Supplies and Consumables
Area
Weigh Room
Weigh Room
Weigh Room
All
All
Liquid
Chromatography
Liquid
Chromatography
GC/MS
GC/MS
GC/MS
GC/MS
GC/MS
GC/MS
GC/MS
GC/MS
ICP
ICP
ICP
All Instruments
All Instruments
All Instruments
Item
Staticide
Forceps
Air Filters
Powder Free Antistatic
Gloves
Low- lint wipes
Teflon tubing
Chromatographs
Column
Chromatographs
Column
FID Detector
Helium
Hydrogen Gas
Zero Air
Liquid Nitrogen
Silica Gel
cryogenic traps
Argon Coolant
Deionized H20
Photo multiplier Tube
Reagent Grade Solvents
Reagent Grade Solvents
Various sizes of ferrules,
tubing and connectors
Description
Anti-static solution
non-serrated/Teflon
Coated
High Efficiency
Vinyl, Class M4.5
4.5" x 8.5"
Cleaning Wipes
1/8" PTFE tubing
36" column
48" column
High Detection
Carrier Gas
Flame Gas
Calibration Gas
200 gallons tank
Canister
stainless steel
Coolant Flow
Post Flush
Analytical element
See SOPs
See SOPs
See SOPs
Vendor
WeighTech
WeighTech
Purchase Local
Fisher Scientific @
Kimwipes@
TubeTech Inc
ZanTech Inc.
ZanTech Inc.
ZanTech Inc.
CylinderTech
CylinderTech
CylinderTech
All Gases Inc.
Zantech Inc
CylinderTech
CylinderTech
Various Vendors
ZanTech Inc.
Various Vendors
Various Vendors
Various Vendors
Model Number
W1024
WWW
11-393-85A
34155
T108
CW01
C1004
D1001
HI 002 3
HI 002 2
HI 0024
HI 0021
SI 00 2 2
HI 002 3
A10022
FT 10045
-------
Project: Model QAPP
Element No: 17
Revision No :1
Date: 11/04/02
Page 3 of3
17.3 Acceptance Criteria
Acceptance criteria must be consistent with overall project technical and quality criteria. It is
the air monitoring branch chief and the field technician's responsibility to update the criteria for
acceptance of consumables. As requirements change, so do the acceptance criteria. Knowledge
of field and laboratory equipment and experience are the best guides to acceptance criteria.
Other acceptance criteria such as observation of damage due to shipping can only be performed
once the equipment has arrived on site.
17.4 Tracking and Quality Verification of Supplies and Consumables
Tracking and quality verification of supplies and consumables have two main components. The
first is the need of the end user of the supply or consumable to have an item of the required
quality. The second need is for the purchasing District to accurately track goods received so that
payment or credit of invoices can be approved. In order to address these two issues, the
following procedures outline the proper tracking and documentation procedures to follow:
• Receiving personnel will perform a rudimentary inspection of the packages as they are
received from the courier or shipping company. Note any obvious problems with a receiving
shipment such as crushed box or wet cardboard.
• The package will be opened, inspected and contents compared against the packing slip.
• If there is a problem with the equipment/supply, note it on the packing list, notify the branch
chief of the receiving area and immediately call the vendor.
• If the equipment/supplies appear to be complete and in good condition, sign and date the
packing list and send to accounts payable so that payment canbe made in a timely manner.
• Notify appropriate personnel that equipment/supplies are available. For items such as the
filters, it is critical to notify the laboratory manager of the weigh room so sufficient time for
processing of the filters can be allowed.
• Stock equipment/supplies in appropriate pre-determined area.
• For supplies, consumables, and equipment used throughout the program, document when
these items are changed out. A sign-in/sign-out sheet is placed outside of the stockroom. All
personnel must sign-out for any consumables removed or added to the stock room. A lab
technician then enters this data into the equipment tracking database.
-------
Project: Model QAPP
Element No: 18
Revision No :1
Date: 11/04/02
Page 1 of3
18.0 Data Acquisition Requirements
This section addresses data not obtained by direct measurement from the Air Toxics Monitoring
Program. This includes both outside data and historical monitoring data. Non-monitoring data
and historical monitoring data are used by the Program in a variety of ways. Use of information
that fails to meet the necessary DQOs for the ATMP leads to erroneous trend reports and
regulatory decision errors. The policies and procedures described in this section apply both to
data acquired through the TCAPCD ATMP and to information previously acquired and/or
acquired from outside sources.
18.1 Acquisition of Non-Direct Measurement Data
The ATMP relies on data that are generated through field and laboratory operations; however,
other significant data are obtained from sources outside the TCAPCD or from historical records.
This section lists this data and addresses quality issues related to the ATMP.
Chemical and Physical Properties Data
Physical and chemical properties data and conversion constants are often required in the
processing of raw data into reporting units. This type of information that has not already been
specified in the monitoring regulations will be obtained from nationally and internationally
recognized sources. Other data sources may be used with approval of the Air Division QA
Officer.
National Institute of Standards and Technology;
• ISO, IUPAC, ANSI, and other widely-recognized national and international standards
organizations;
• U.S. EPA;
The current edition of certain standard handbooks may be used without prior approval of the
Toxa City QA Officer. Two that are relevant to the fine particulate monitoring program are
CRC Press' Handbook of Chemistry and Physics, and Merck Manual.
Sampler Operation and Manufacturers' Literature
Another important source of information needed for sampler operation is manufacturers'
literature. Operations manuals and users' manuals frequently provide numerical information and
equations pertaining to specific equipment. TCAPCD personnel are cautioned that such
information is sometimes in error, and appropriate cross-checks will be made to verify the
reasonableness of information contained in manuals. Whenever possible, the field operators will
compare physical and chemical constants in the operators manuals to those given in the sources
-------
Project: Model QAPP
Element No: 18
Revision No :1
Date: 11/04/02
Page 2 of3
listed above. If discrepancies are found, determine the correct value by contacting the
manufacturer. The following types of errors are commonly found in such manuals:
• Insufficient precision;
• outdated values for physical constants;
• Typographical errors;
• incorrectly specified units;
• inconsistent values within a manual, and
• use of different reference conditions than those called for in EPA regulations.
Geographic Location
Another type of data that will commonly be used in conjunction with the Monitoring Program is
geographic information. For the current sites, the District will locate these sites using global
positioning systems (GPS) that meet EPA Locational Data Policy of 25 meters accuracy. USGS
maps were used as the primary means for locating and siting stations in the existing network.
External Monitoring Data Bases
It is the policy of the TCAPCD that no data obtained from the Internet, computer bulletin boards,
or data bases from outside organizations shall be used in creating reportable data or published
reports without approval of the Air Division Director. This policy is intended to ensure the use
of high quality data in Toxa City publications.
Data from the EPA -AIRS data base may be used in published reports with appropriate caution.
Care must be taken in reviewing/using any data that contain flags or data qualifiers. If data is
flagged, such data shall not be utilized unless it is clear that the data still meets critical QA/QC
requirements. It is impossible to assure that a data base such as AIRS is completely free from
errors including outliers and biases, so caution and skepticism is called for in comparing Toxa
City data from other reporting agencies as reported in AIRS. Users should review available
QA/QC information to assure that the external data are comparable with Toxa City
measurements and that the original data generator had an acceptable QA program in place.
Speciated Particulate Data
The TCAPCD has been routinely monitoring PM2 5 mass sampler at TC5 since the 1999. In
mid-2000, the EPA notified the TCAPCD that it should operate a PM 2.s speciation sampler at
the TC5 site as well.
-------
Project: Model QAPP
Element No: 18
Revision No :1
Date: 11/04/02
Page 3 of3
PM2.s Speciation data has collected at TC5 and is one of the reasons that TC5 was selected as the
NATTS.
U.S. Weather Service Data
Meteorological information is gathered from the U.S. Weather Service station at the Toxa City
International Airport. Parameters include: temperature, relative humidity, barometric pressure,
rainfall, wind speed, wind direction, cloud type/layers, percentage cloud cover and visibility
range. Historically, these data have not been used to calculate pollutant concentration values for
any of the Toxa City monitoring sites, which each have the required meteorological sensors.
However, NWS data are often included in summary reports. No changes to the way in which
these data are collected are anticipated due to the addition of the air toxics data to the Toxa City
Air Pollution Control District.
-------
Project: Model QAPP
Element No: 19
Revision No :1
Date: 11/04/02
Page 1 of 10
19.0 Data Management
19.1 Background and Overview
This section describes the data management operations pertaining to measurements for the air
toxics stations operated by TCAPCD. This includes an overview of the mathematical operations
and analyses performed on raw ("as-collected") data. These operations include data recording,
validation, transformation, transmittal, reduction, analysis, management, storage, and retrieval.
Data processing for air toxics data are summarized in Figure 19-1. Data processing steps are
integrated, to the extent possible, into the existing data processing system used for The TCAPCD
air toxics network. The data base resides on a machine running the Windows™ NT Server
operating system, which is also the main file server for the Air Quality Division. This machine is
shown in the upper left of Figure 19-1.
The sample tracking and chain of custody information are entered into the Laboratory
Information Management System (LEVIS) at three main stages as shown in Figure 19-1.
Managers are able to obtain reports on status of samples, location of specific samples, etc., using
the LIMS. All users must be authorized by the Manager, Air Quality Division, and receive a
password necessary to log on to the LIMS. Different privileges are given each authorized user
depending on that person's need. The following privilege levels are defined:
• Data Entry Privilege - The individual may see and modify only data within LEVIS, he or she
has personally entered. After a data set has been "committed" to the system by the data entry
operator, all further changes will generate entries in the system audit trail;
• Reporting Privilege - This without additional privileges;
• Data Administration Privilege - Data Administrators for the LEVIS are allowed to change data
as a result of QA screening and related reasons. All operations resulting in changes to data
values are logged to the audit trail. The Data Administrator is responsible for performing the
following tasks on a regular basis;
• Merging/correcting the duplicate data entry files;
• Running verification/validation routines, correcting data as necessary and generating
summary data reports for management;
• Uploading verified/validated data to EPA -AQS.
-------
Project: Model QAPP
Element No: 19
Revision No :1
Date: 11/04/02
Page 2 of 10
Presampling ID — ^- Sample ^
No. assigned Collection
Transfer to Lab
1
T ^^
-"- LIMS
fe™* ' '• ,.'-- .; -
m»
1
Data Manager
Review
I
Transfer to AQS
1
"£
Sample Login
i
Sample
Analysis/Data
Calculation
t
Lab Supervisor
QA Review
t
Data Entry
t
Laboratory Meta
data Storage
Short Term
Storage
t
Laboratory Meta
data Storage
Long Term
Storage
^ Short Term
Storage
J
Long Term
Stoarge
^^ Data Transfer
Figure 19.1 Data Management and Sample Flow Diagram
-------
Project: Model QAPP
Element No: 19
Revision No :1
Date: 11/04/02
Page 3 of 10
19.2 Data Recording
Data entry, validation, and verification functions are all integrated in the LEVIS. Bench sheet
data are entered into the LIMS by laboratory personnel. Procedures for filling out the laboratory
sheets and subsequent data entry are provided in SOPs listed in Table 19.1 and included in the
SOPs.
19.3 QA Review
QA review is the data validation portion that is a combination of checking that data processing
operations have been carried out correctly and of monitoring the quality of the field operations.
Data validation can identify problems in either of these areas. Once problems are identified, the
data can be corrected or invalidated, and corrective actions can be taken for field or laboratory
operations. Numerical data stored in the LEVIS are never internally overwritten by condition
flags. Flags denoting error conditions or QA status are saved as separate fields in the data base,
so that it is possible to recover the original data.
The following functions are incorporated into the LIMS ensure quality of data entry and data
processing operations:
• Duplicate Key Entry - the following data are subjected to duplicate entry by different
operators: filter weight reports, field data sheets, chain of custody sheets. The results of
duplicate key entry are compared and errors are corrected at biweekly intervals. The method
for entering the data is given in the SOPs. Procedures for reconciling the duplicate entries are
given in SOPs.
• Range Checks - almost all monitored parameters have simple range checks programmed in.
For example, valid times must be between 00:00 and 23:59, summer temperatures must be
between 10 and 50 degrees Celsius, etc. The data entry operator is notified immediately when
an entry is out of range. The operator has the option of correcting the entry or overriding the
range limit. The specific values used for range checks may vary depending on season and
other factors. Since these range limits for data input are not regulatory requirements, the Air
Division QA Officer may adjust them from time to time to better meet quality goals.
• Completeness Checks - When the data are processed certain completeness criteria must be
met. For example, each sample must have a start time, an end time, an average flow rate,
dates weighed or analyzed and operator and technician names. The data entry operator will
be notified if an incomplete record has been entered before the record can be closed.
• Internal Consistency and Other Reasonableness Checks - Several other internal
consistency checks are built into the LEVIS. For example, the end time of a sample must be
greater than the start time. Computed filter volume (integrated flow) must be approximately
equal to the exposure time multiplied by the nominal flow. Additional consistency and other
checks will be implemented as the result of problems encountered during data screening..
-------
Project: Model QAPP
Element No: 19
Revision No :1
Date: 11/04/02
Page 4 of 10
• Data Retention - Raw data sheets are retained on file in the Air Quality Division office for a
minimum of five years, and are readily available for audits and data verification activities.
After five years, hardcopy records and computer backup media are cataloged and boxed for
storage at the Toxa City Services Warehouse. Physical samples such as filters shall be
discarded with appropriate attention to proper disposal of potentially hazardous materials.
• Statistical Data Checks - Errors found during statistical screening will be traced back to
original data entry files and to the raw data sheets, if necessary. These checks shall be run on
a monthly schedule and prior to any data submission to AIRS. Data validation is the process
by which raw data are screened and assessed before it can be included in the main data base
(i.e., the LIMS).
• Sample Batch Data Validation-, which is discussed in Section 23, associate flags that are
generated by QC values outside of acceptance criteria, with a sample batch. Batches
containing too many flags would be rerun and or invalidated.
Table 19.1 summarizes the validation checks applicable to the data.
Table 19.1 Validation Check Summaries
Type of Data Check
Data Parity and Transmission Protocol Checks
Duplicate Key Entry
Date and Time Consistency
Completeness of Required Fields
Range Checking
Statistical Outlier Checking
Manual Inspection of Charts and Reports
Field and Lab Blank Checks
Electronic
Transmission
and Storage
•
Manual
Checks
•
•
•
•
•
Automated
Checks
•
•
•
•
The objective of the TCAPCD will be to optimize the performance of its monitoring equipment.
Initially, the results of collocated operations will be control charted (see Section 14). From these
charts, control limits will be established to flag potential problems.
-------
Project: Model QAPP
Element No: 19
Revision No :1
Date: 11/04/02
Page 5 of 10
19.4 Data Transformation
Calculations for transforming raw data from measured units to final concentrations are
relatively straightforward.
19.5 Data Transmittal
Data transmittal occurs when data are transferred from one person or location to another or when
data are copied from one form to another. Some examples of data transmittal are copying raw
data from a notebook onto a data entry form for keying into a computer file and electronic
transfer of data over a telephone or computer network. Table 19-3 summarizes data transfer
operations.
Table 19.2 Data Transfer Operations
Description of Data
Transfer
Keying Data into The
LIMS
Electronic data transfer
Filter Receiving and
Chain-of-Custody
Calibration and Audit
Data
AIRS data summaries
Originator
Laboratory Technician
(hand- written data form)
(between computers or
over network)
Shipping and Receiving
Clerk
Auditor or field
supervisor
Air Quality Supervisor
Recipient
Data Processing
Personnel
~
The LIMS computer
(shipping clerk enters
data at a local terminal)
Air Quality Field
Supervisor
AQS (U.S. EPA)
QA Measures Applied
Double Key Entry
Parity Checking;
transmission protocols
Sample numbers are
verified automatically;
reports indicate missing
filters and/or incorrect
data entries
Entries are checked by
Air Quality Supervisor
and QA Officer
Entries are checked by
Air Quality Supervisor
and QA Officer
The TCAPCD will report all ambient air quality data and information specified by the AQS
Users Guide, coded in the AQS format. Such air quality data and information will be fully
screened and validated and will be submitted directly to the AQS via electronic transmission, in
the format of the AQS, and in accordance with the quarterly schedule. The specific quarterly
reporting periods and due dates are shown in the Table 19.3.
-------
Project: Model QAPP
Element No: 19
Revision No :1
Date: 11/04/02
Page 6 of 10
Table 19.3 Data Reporting Schedule
Reporting Period
January 1- March 31
April 1-June 30
July 1- September 30
October 1 -December 31
Due Date
June 30
September 30
December 3 1
March 3 1
19.6 Data Reduction
Data reduction processes involve aggregating and summarizing results so that they can be
understood and interpreted in different ways. Since air toxics have no regulatory requirements,
such as those with the NAAQS, monitoring regulations are not required to be reported regularly
to U.S. EPA. Examples of data summaries include:
• average concentration for a station or set of stations for a specific time period;
accuracy, and precision statistics;
data completeness reports based on numbers of valid samples collected during a specified
period.
The Audit Trail is another important concept associated with data transformations and
reductions. An audit trail is a data structure that provides documentation for changes made to a
data set during processing. Typical reasons for data changes that would be recorded include the
following:
• corrections of data input due to human error;
• application of revised calibration factors;
• addition of new or supplementary data;
• flagging of data as invalid or suspect;
• logging of the date and times when automated data validation programs are run.
The audit trail is implemented as a separate table a relational data base. Audit trail records will
include the following fields:
operator's identity (ID code);
date and time of the change;
-------
Project: Model QAPP
Element No: 19
Revision No :1
Date: 11/04/02
Page 7 of 10
• table and field names for the changed data item;
• reason for the change;
• full identifying information for the item changed (date, time, site location, parameter, etc.);
• value of the item before and after the change.
When routine data screening programs are run, the following additional data are recorded in the
audit trail:
• version number of the screening program;
• values of screening limits (e.g., upper and lower acceptance limits for each parameter);
• numerical value of each data item flagged and the flag applied.
The audit trail is produced automatically and can only document changes; there is no "undo"
capability for reversing changes after they have been made. Available reports based on the audit
trail include:
• log of routine data validation, screening, and reporting program runs;
• report of data changes by station for a specified time period;
• report of data changes for a specified purpose;
• report of data changes made by a specified person.
Because of storage requirements, the System Administrator must periodically move old audit
trail records to backup media. Audit trail information will not be moved to backup media until
after the data are reported to AIRS. All backups will be retained so that any audit trail
information can be retrieved for at least three years.
19.7 Data Summary
The TCAPCD is currently implementing the data summary and analysis program The following
specific summary statistics will be tracked and reported for the network:
Single sampler accuracy (based on audit flow checks and laboratory audits);
Single sampler precision (based on collocated data);
• Network-wide bias and precision;
• Data completeness.
-------
Project: Model QAPP
Element No: 19
Revision No :1
Date: 11/04/02
Page 8 of 10
Equations used for these reports are given in the Table 19.4.
Table 19.4 Report Equations
Criterion
Accuracy of Single Sampler Flow
- Single Check (d;) X; is reference
flow; Y; is measured flow
Percent Difference for a Single
Check (d;) - X; and Y; are
concentrations from the primary
and duplicate samplers,
respectively.
Upper 95% Confidence Limit
Lower 95% Confidence Limit
Completeness
Equation
d = Yi-X x 100
x,
4 = Yi-X x 100
(Yi+Xj)/2
Umit=di+J.96*Si/»2
Limit =d, _1.96*S, /• 2
Completeness = Nvaiid * 100
^ nheorectical
19.8 Data Tracking
The LEVIS contains the necessary input functions and reports necessary to track and account for
the whereabouts of filters and the status of data processing operations for specific data.
Information about filter location is updated at distributed data entry terminals at the points of
significant operations. The following input locations are used to track sample location and
status:
• Laboratory (initial receipt)
• Sample receipt (by lot);
• Pre-sampling processing or weighing (individual filter or cartridge number first enters the
system);
• Canister number (VOC only);
• Filter packaged for the laboratory (filter numbers in each package are recorded);
• Shipping (package numbers are entered for both sending and receiving);
-------
Project: Model QAPP
Element No: 19
Revision No :1
Date: 11/04/02
Page 9 of 10
• Laboratory(receipt from field)
• Package receipt (package is opened and filter numbers are logged in);
• Filter post-sampling weighing;
• Filter archival.
In most cases the tracking data base and the monitoring data base are updated simultaneously.
For example, when the filter is pre-weighed, the weight is entered into the monitoring data base
and the filter number and status are entered into the tracking data base. For the VOC system, the
sample handling is different. The VOC canisters are reused many times before they are retired
from field use. Each canister has its own unique code that designates the can number. When the
canister is sent into the field, a canister number becomes a portion of the tracking code. This
allows the sample that was in the canister to be tracked.
The Air Division Branch Chief or designee is responsible for tracking sample status at least
twice per week and following up on anomalies such as excessive holding time in the laboratory
before analysis.
19.9 Data Storage and Retrieval
Data archival policies for the data are shown in Table 19.5.
Table 19.5 Data Archive Policies
Data Type
Weighing records; chain
of custody forms
Laboratory Notebooks
Field Notebooks
Data Base (excluding
Audit Trail records)
Trail record
PMjo Quartz filters
VOC canisters
DNPH cartridge
Medium
Hardcopy
Hardcopy
Hardcopy
Electronic (on-
line)
Hardcopy and
electronic
reports
Filters
metal can
plastic
cartridge
Location
Laboratory
Laboratory
Air Quality
Division
Air Quality
Division
Air Quality
Division
Laboratory
Laboratory
Laboratory
Retention Time
3 years
3 years
3 years
indefinite (may be
moved to backup media
after 5 years)
3 years
1 year
reused after cleaning
6 months
Final Disposition
Discarded
N/A
Discarded
Backup tapes retained
indefinitely
N/A
Discarded
Recycled
Discarded
-------
Project: Model QAPP
Element No: 19
Revision No :1
Date: 11/04/02
Page 10 of 10
The data reside on an Local Access Network on the TCAPCD server. This computer has the
following specifications:
Storage: 18 GB (SCSI RAID 0 array);
Backup: DAT (3 GB per tape) - incremental backups daily; full backups biweekly;
Network: Windows NT, 100 Mbps Ethernet network (currently 23 Windows 95 and NT
workstations on site; additional workstations via 28.8 kbps dial-in modem);
Security: Password protection on all workstations and dial-in lines; Additional password
protection applied by application software.
Security of data in the data base is ensured by the following controls:
Password protection on the data base that defines three levels of access to the data;
Regular password changes (quarterly for continuing personnel; passwords for personnel leaving
the Air Division will be canceled immediately);
Independent password protection on all dial-in lines;
Logging of all incoming communication sessions, including the originating telephone number, the
user's ID, and connect times;
Storage of media including backup tapes in locked, restricted access areas.
-------
Project: Model QAPP
Element No: 20
Re vision No: 1
Date: 11/04/02
Page 1 of8
20.0 Assessment and Response Actions
An assessment is defined as an evaluation process used to measure the performance or
effectiveness of the quality system or the establishment of the monitoring network and sites
and various measurement phases of the data operation..
The results of quality assurance assessments indicate whether the control efforts are adequate
or need to be improved. Documentation of all quality assurance and quality control efforts
implemented during the data collection, analysis, and reporting phases is important to data
users, who can then consider the impact of these control efforts on the data quality (see
Section 21). Both qualitative and quantitative assessments of the effectiveness of these
control efforts will identify those areas most likely to impact the data quality and to what
extent. In order to ensure the adequate performance of the quality system, the TCAPCD in
conjunction with the State, EPA Regional office will perform the following assessments.
20.1 Assessment Activities and Project Planning
20.1.1 Management Systems Review
An MSR is a qualitative assessment of a data collection operation or organization to establish
whether the prevailing quality management structure, policies, practices, and procedures are
adequate. MSRs conducted every three years by the QA Division. The MSR will use
appropriate regulations, and the QAPP to determine the adequate operation of the air
program and its related quality system. The quality assurance activities of all criteria
pollutants including air toxics will be part of the MSR. The QA Office Director's staff will
report its findings to the appropriate Divisions within 30 days of completion of the MSR.
The report will be appropriately filed. Follow-up and progress on corrective action(s) will
be determined during regularly scheduled division directors meetings
-------
Project: Model QAPP
Element No: 20
Re vision No: 1
Date: 11/04/02
Page 2 of8
20.1.2 Network Reviews
Conformance with network requirements of the monitoring network is accomplished through
annual reviews. The network review is used to determine how well a particular air
monitoring network is achieving its required air monitoring objective, and how it should be
modified to continue to meet its objective. The network review will be accomplished every
3 years. Since the states are also required to perform these reviews, the District will
coordinate its activity with the State in order to perform the activity at the same time (if
possible). The Air Monitoring Branch will be responsible for conducting the network review.
The following criteria will be considered during the review:
• date of last review;
• areas where attainment/nonattainment re-designations are taking place or are likely to
take place;
• results of special studies, saturation sampling, point source oriented ambient monitoring,
etc.;
• proposed network modifications since the last network review.
In addition, pollutant-specific priorities may be considered in areas that models may show
persons to be at risk.
Prior to the implementation of the network review, significant data and information
pertaining to the review will be compiled and evaluated. Such information might include the
following:
network files (including updated site information and site photographs);
air quality summaries for the past five years for the monitors in the network;
air toxics emissions trends reports for major metropolitan area;
• emission information, such as emission density maps for the region in which the
monitor is located and emission maps showing the major sources of emissions;
• National Weather Service summaries for monitoring network area.
Upon receiving the information it will be checked to ensure it is the most current.
Discrepancies will be noted on the checklist and resolved during the review. Files and/or
photographs that need to be updated will also be identified. The following categories will
emphasized during network reviews:
-------
Project: Model QAPP
Element No: 20
Re vision No: 1
Date: 11/04/02
Page 3 of8
Adequacy of the network will be determined by using the following information:
maps of historical monitoring data;
maps of emission densities;
dispersion modeling;
special studies/saturation sampling;
best professional judgement;
• SIP requirements;
• GIS updates.
The number of samplers operating can be determined from the AMP220 report in AIRS. The
number of monitors required, based on concentration levels and population, can be
determined from the AMP450 report and the latest census population data.
Location of Monitors- Adequacy of the location of monitors can only be determined on the
basis of stated objectives. Maps, graphical overlays, and GIS-based information will be
helpful in visualizing or assessing the adequacy of monitor locations. Plots of potential
emissions and/or historical monitoring data versus monitor locations will also be used.
During the network review, the stated objective for each monitoring location or site (see
section 10) will be "reconfirmed" and the spatial scale "re-verified" and then compared to
each location to determine whether these objectives can still be attained at the present
location.
Probe Siting Requirements- The on-site visit will consist of the physical measurements
and observations to determine the best locations. Prior to the site visit, the reviewer will
obtain and review the following::
• most recent hard copy of site description (including any photographs);
• data on the seasons with the greatest potential for high concentrations for specified
pollutants;
• predominant wind direction by season.
A checklist similar to the checklist used by the EPA Regional offices during their scheduled
network reviews will be used. This checklist can be found in the SLAMS/NAMS/PAMS
Network Review Guidance which is intended to assist the reviewers In addition to the items
on the checklist, the reviewer will also perform the following tasks:
• ensure that the inlet is clean;
• record findings in field notebook and/or checklist;
-------
Project: Model QAPP
Element No: 20
Re vision No: 1
Date: 11/04/02
Page 4 of8
• take photographs/videotape in the 8 directions;
• document site conditions, with additional photographs/videotape.
Other Discussion Topics - In addition to the items included in the checklists, other subjects
for discussion as part of the network review and overall adequacy of the monitoring program
will include:
• installation of new monitors;
• relocation of existing monitors;
• siting criteria problems and suggested solutions;
problems with data submittals and data completeness;
maintenance and replacement of existing monitors and related equipment;
quality assurance problems;
air quality studies and special monitoring programs;
other issues;
A report of the network review will be written within two months of the review and
appropriately filed.
20.1.3 Technical Systems Audits
A ISA is a thorough and systematic on-site qualitative audit, where facilities, equipment,
personnel, training, procedures, and record keeping are examined for conformance to the
QAPP. TSAs of the network will be accomplished every three years and will stagger the
required ISA conducted by the State or EPA Regional QA Office. The QA Office will
implement the TSA either as a team or as an individual auditor. The QA Office will perform
three TSA activities that can be accomplished separately or combined:
• Field - handling, sampling, shipping.;
1 • • 'Laboratory - Pre-sampling , shipping, receiving, post-sampling weighing, analysis,
archiving, and associated QA/QC;
• Data management - Information collection, flagging, data editing, security, upload.
Key personnel to be interviewed during the audit are those individuals with responsibilities
for: planning, field operations, laboratory operations, QA/QC, data management, and
reporting. To increase uniformity of the TSA, an audit checklist will be developed and used.
This checklist is based on the EPA R-5 guidance.
The audit team will prepare a brief written summary of findings, organized into the following
areas: planning, field operations, laboratory operations, quality assurance/quality control,
-------
Project: Model QAPP
Element No: 20
Re vision No: 1
Date: 11/04/02
Page 5 of8
data management, and reporting. Problems with specific areas will be discussed and an
attempt made to rank them in order of their potential impact on data quality.
The audit finding form has been designed such that one is filled out for each major deficiency
that requires formal corrective action. The finding should include items like: systems
impacted, estimated time period of deficiency, site(s) affected, and reason of action. The
finding form will inform the Division about serious problems that may compromise the
quality of the data and therefore require specific corrective actions. They are initiated by the
Audit Team, and discussed at the debriefing. During the debriefing, if the audited group is in
agreement with the finding, the form is signed by the group's branch manager or his designee
during the exit interview. If a disagreement occurs, the Audit Team will record the opinions
of the group audited and set a time at some later date to address the finding at issue.
Post-Audit Activities- The major post-audit activity is the preparation of the systems audit
report. The report will include:
•
audit title and number and any other identifying information;
audit team leaders, audit team participants and audited participants;
background information about the project, purpose of the audit, dates of the audit;
particular measurement phase or parameters that were audited, and a brief description of
the audit process;
summary and conclusions of the audit and corrective action requires;
attachments or appendices that include all audit evaluations and audit finding forms.
To prepare the report, the audit team will meet and compare observations with collected
documents and results of interviews and discussions with key personnel. Expected QA Project
Plan implementation is compared with observed accomplishments and deficiencies and the audit
findings are reviewed in detail. Within thirty (30) calendar days of the completion of the audit, the
audit report will be prepared and submitted. The systems audit report will be submitted to the
appropriate branch managers and appropriately filed.
If the branch has written comments or questions concerning the audit report, the Audit Team will
review and incorporate them as appropriate, and subsequently prepare and resubmit a report in
final form within thirty (30) days of receipt of the written comments. The report will include an
agreed-upon schedule for corrective action implementation.
Follow-up and Corrective Action Requirements- The QA Office and the audited organization
may work together to solve required corrective actions. As part of corrective action and follow-
up, an audit finding response letter will be generated by the audited organization. The audit
finding response letter will address what actions are being implemented to correct the finding of
-------
Project: Model QAPP
Element No: 20
Revision No: 1
Date: 12/3/02
Page 6 of 8
the TSA. The audit response letter will be completed by the audited organization within 30 days
of acceptance of the audit report.
20.1.4 Performance Audit
A Performance Audit is a field operations audit that ascertains whether the samplers are operating
within the specified limits as stated in the SOPs and QAPP. The Performance Audit is performed
every year in conjunction with the field TSA. The audit consists of challenging the samplers to
operate using independent NIST-traceable orifices or other flow devices. Once the audit has been
performed, the flow rate is calculated and compared against the flow rates as specified in the
QAPP or SOPs. If the flowrates are not within these ranges, then the field operations technician
is notified and corrective action ensues. Once the field technicians have remedied the situation, a
post audit confirms the adjustment or maintenance. The audit results are then written in a detailed
report.
20.1.5 Data Quality Assessments
A data quality assessment (DQA) is the statistical analysis of environmental data to determine
whether the quality of data is adequate to support the decision which are based on the DQOs.
Data are appropriate if the level of uncertainty in a decision based on the data is acceptable. The
DQA process is described in detail in Guidance for the Data Quality Assessment Process, EPA
QA/G-9 and is summarized below.
• Review the data quality objectives (DQOs) and sampling design of the program: review the
DQO. Define statistical hypothesis, tolerance limits, and/or confidence intervals.
• Conduct preliminary data review. Review Precision &Accuracy (P&A) and other available
QA reports, calculate summary statistics, plots and graphs. Look for patterns, relationships,
or anomalies.
• Select the statistical test: select the best test for analysis based on the preliminary review, and
identify underlying assumptions about the data for that test.
• Verify test assumptions: decide whether the underlying assumptions made by the selected test
hold true for the data and the consequences.
• Perform the statistical test: perform test and document inferences. Evaluate the performance
for future use.
-------
Project: Model QAPP
Element No: 20
Revision No: 1
Date: 12/3/02
Page 7 of 8
Measurement uncertainty will be estimated for both automated and manual methods.
Terminology associated with measurement uncertainty are found within 40 CFR Part 58
Appendix A and includes: (a) Precision - a measurement of mutual agreement among individual
measurements of the same property usually under prescribed similar conditions, expressed
generally in terms of the standard deviation; (b) Accuracy- the degree of agreement between an
observed value and an accepted reference value, accuracy includes a combination of random error
(precision) and systematic error (bias) components which are due to sampling and analytical
operations; (c) Bias- the systematic or persistent distortion of a measurement process which
causes errors in one direction. Estimates of the data quality will be calculated on the basis of
single monitors and aggregated to all monitors.
20.1.6 Performance Evaluations
The PE is an assessment tool for the laboratory operations. The EPA's Contract laboratory for the
UATMP creates "blind" samples and sends them periodically to the District's laboratory. Upon
receipt, the laboratory logs in the samples and performs the normal handling routines as any other
sample. The PE is analyzed in accordance with the SOPs and QAPP. The results are then sent to
the Laboratory Branch Manager for final review. Then the results are reported to the EPA's
Contract Laboratory Director. The Contract laboratory writes up a PE report and sends a copy of
the results to the Laboratory Branch Manager and the EPA QA Office. Any results outside of the
EPA's acceptance criteria are then noted in the PE report. The TCAPCD has 120 days to address
any deficiencies noted in the PE Report.
-------
20.2 External Assessment Schedule
Project: Model QAPP
Element No: 20
Revision No: 1
Date: 12/3/02
Page 8 of 8
Table 20.1 Assessment Summary
Agency
EPA - NAREL
ERG
OAQPS-EMAD
Regional Offices
Regional Offices
Type of Assessment
TSA and PEs, round robin
inter-laboratory samples
PEs
MSRs, TSAs
Network Reviews
TSAs and IPAs
Agency Assessed
ERG
S/L/T agencies
ERG, NAREL,
EPA Regional and
S/L/T agencies
S/L/T agencies
S/L/T agencies
Frequency
Annually
Annually
As needed by
EMAD
determination
Once every 5
years
Annually *
-------
Project: Model QAPP
ElementNo:21
Re vision No: 1.0
Date: 11/04/02
Page 1 of 2
21.0 Reports to Management
This section describes the quality-related reports and communications to management necessary
to support air toxics network operations and the associated data acquisition, validation,
assessment, and reporting.
Important benefits of regular QA reports to management include the opportunity to alert the
management of data quality problems, to propose viable solutions to problems, and to procure
necessary additional resources. Management should not rely entirely upon the MSRs for their
assessment of the data. The MSR only occur once every three years. Quality assessment,
including the evaluation of the technical systems, the measurement of performance, and the
assessment of data, is conducted to help insure that measurement results meet program objectives
and to insure that necessary corrective actions are taken early, when they will be most effective.
Effective communication among all personnel is an integral part of a quality system. Regular,
planned quality reporting provides a means for tracking the following:
adherence to scheduled delivery of data and reports,
• documentation of deviations from approved QA and test plans, and the impact of these
deviations on data quality;
• analysis of the potential uncertainties in decisions based on the data.
21.1 Frequency, Content, and Distribution of Reports
Required reports to management for monitoring in general are discussed in various sections of 40
CFR Parts 53 and 58. Guidance for management report format and content are provided in
guidance developed by EPA's Quality Assurance Division (QAD) and the Office of Air Quality
Planning and Standards. These reports are described in the following subsections.
21.1.1 QA Annual Report
Periodic assessments of air toxics data are required to be reported to EPA (40 CFR 58 Appendix
A, Section 1.4, revised July 18, 1997). The Toxa City Air Pollution Control Air Division's QA
Annual Report is issued to meet this requirement. This document describes the quality
objectives for measurement data and how those objectives have been met.
-------
Project: Model QAPP
ElementNo:21
Re vision No: 1.0
Date: 11/04/02
Page 2 of 2
The QA Annual Report will include Quality information for each air toxic monitored in the
network. Each section includes the following topics:
• program overview and update;
• quality objectives for measurement data;
• data quality assessment.
For reporting air toxics measurement uncertainties, the QA Annual Report contains the following
summary information:
• Flow Rate Audits;
• Collocated Samplers Audits using estimation of Precision;
Laboratory audits which include "round-robin" cylinders that are shared among many
laboratories;
NPAP audits.
21.1.2 Technical System Audit Reports
The TCAPCD performs Technical System Audits of the monitoring system (section 20). These
reports will be filed and made available to EPA personnel during their technical systems audits.
External systems audits are conducted at least annually by the EPA Regional Office as required
by 40 CFR Part 58. Further instructions are available from the EPA Regional QA Coordinator or
the Systems Audit QA Coordinator, Office of Air Quality Planning and Standards, Emissions
Monitoring and Analysis Division (MD-14), U.S. Environmental Protection Agency, Research
Triangle Park, NC 27711.
21.1.3 Response/Corrective Action Reports
The Response/Corrective Action Report procedure will be followed whenever a problem is
found such as a safety defect, an operational problem, or a failure to comply with procedures. A
Response/Corrective Action Report is one of the most important ongoing reports to management
because it documents primary QA activities and provides valuable records of QA activities.
-------
Project: Model QAPP
Element No: 22
Re vision No: 1.0
Date: 11/04/02
Page 1 of 4
22.0 Data Review
22.1 Data Review Design
The primary purpose of this section is to describe the data validation procedures which are used
by the TCAPCD to process ambient air toxics data. Data validation refers to those activities
performed after the fact, that is, after the data have been collected. The difference between data
validation and quality control techniques is that the quality control techniques attempt to
minimize the amount of bad data being collected, while data validation seeks to prevent any bad
data from getting through the data collection and storage systems.
It is preferable that data review be performed as soon as possible after the data collection, so that
the questionable data can be checked by recalling information on unusual events and on
meteorological conditions which can aid in the validation. Also, timely corrective actions should
be taken when indicated to minimize further generation of questionable data. The data review
group will attempt to review the data within 1 month after the end of the month of sampling.
This will also help with getting the data loaded onto AQS in a timely manner, as described in
Section 19.5.
Personnel performing data review should:
* Be familiar with typical diurnal concentration variations (e.g., the time daily maximum
concentrations occur and the interrelationship of pollutants.) For example, benzene, toluene
and xylene concentrations usually increase and decrease together, due to these being
attributed to mobile sources, whereas, metals are usually attributable to manufacturing
process, and may have a longer temporal cycle.
Be familiar with the type of instrument malfunctions which cause characteristic trace
irregularities.
Recognize that cyclical or repetitive variations (at the same time each day or at periodic
intervals during the day) may be caused by excessive line voltage or temperature variations.
Nearby source activity can also cause erroneous or non-representative measurements.
* Recognize that flow traces showing little or no activity often indicate flow problems, or
sample line leaks.
-------
Project: Model QAPP
Element No: 22
Re vision No: 1.0
Date: 11/04/02
Page 2 of 4
There is a wide variety of information with which to validate air toxics data. Among them are the
following, along with their uses:
Multi-point Calibration Forms - the multipoint forms should be used to establish proper
initial calibration and can be used to show changes in the calibration;
Span Control Charts - these charts will be the most valuable tool in spotting data that is out
of control limits;
Site and Instrument Logs - because all station activities are noted in one or both of these
logs, one can obtain a good picture of station operations by reading these logs
Data From Other Air Quality Stations - data from other air quality stations nearby can be
compared between two stations to help the identification of invalid data.
Blanks, Replicates and Spikes - these QC indicators can be used to ascertain whether sample
handling or analysis is causing bias in the data set.
• Monthly Summary Reports - The Monthly Summary Reports are outputs from the
Analytical Laboratory LIMS units. These reports are "canned" reports provided by the
computer vendor who writes the interface software. These reports provide the following
information: Completeness report, Initial Calibration Report from the Analytical
Instruments, Laboratory Control Sample Recoveries, Field or Laboratory Surrogate
Recoveries, Spike Recoveries, Laboratory Duplicate Results and Serial Dilution Results.
22.2 Data Review Testing
Recently, the TCAPCD has received a copy of the newly developed program VOCDat. This
program was developed by EPA-OAQPS for PAMS data validation. However, the TCAPCD will
apply this to the Organic Toxics data by using the following VOCDat tests:
22.2.1 Data Identification Checks
Data with improper identification codes are useless. Three equally important identification fields
which must be correct are time, location, parameter and sampler ID.
22.2.2. Unusual Event Review
Extrinsic events (e.g., construction activity, dust storms, unusual traffic volume, and traffic jams)
can explain unusual data. This information could also be used to explain why no data are
reported for a specified time interval, or it could be the basis for deleting data from a file for
specific analytical purposes.
-------
Project: Model QAPP
Element No: 22
Re vision No: 1.0
Date: 11/04/02
Page 3 of 4
22.2.3. Relationship Checks
Toxics data sets contain many physically or chemically related parameters. These relations can
be routinely checked to ensure that the measured values on an individual parameter do not exceed
the corresponding measured values of an aggregate parameter which includes the individual
parameter. For example, benzene, toluene and xylene are mobile source driven. The relative
concentrations are within +/- 10 ppbv, if these values are recorded at the same time and location.
Data sets in which individual parameter values exceed the corresponding aggregate values are
flagged for further investigation. Minor exceptions to allow for measurement system noise may
be permitted in cases where the individual value is a large percentage of the aggregate value.
22.2.4. Review of Spikes, Blanks and Replicates -
An additional check of the data set is to verify that the spikes, blanks and replicate samples have
been reviewed. Generally, recovery of spikes in samples should be greater than 80%. Blanks
should not be more than 3 times the MDL for any compound. The difference in concentration of
replicates should be within +/- 10%. If any of these are outside of this boundary, then the
reviewer should notify the air monitoring branch supervisor for direction. The air branch
supervisor will discuss these results with the lab branch supervisor and the QA officer. The three
will decide whether any of these results can or will invalidate a single run or batch.
22.3 Procedures
These tests check values in a data set which appear atypical when compared to the whole data set.
Common anomalies of this type include unusually high or low values (outliers) and large
differences in adjacent values. These tests will not detect errors which alter all values of the data
set by either an additive or multiplicative factor (e.g., an error in the use of the scale). The
following tests for internal consistency are used:
* Data Plots
* Ratio Test
* Students "t-test"
22.3.1. Tests for Historical and Temporal Consistency
These tests check the consistency of the data set with respect to similar data recorded in the past.
In particular these procedures will detect changes where each item is increased by a constant or by
a multiplicative factor. Gross limit checks are useful in detecting data values that are either
highly unlikely or considered impossible. The use of upper and lower 95% confidence limits is
very useful in identifying outliers.
-------
Project: Model QAPP
Element No: 22
Re vision No: 1.0
Date: 11/04/02
Page 4 of 4
22.3.2 Pattern and Successive Difference Tests
These tests check data for pollutant behavior which has never or very rarely occurred in the past.
Values representing pollutant behavior outside of these predetermined limits are then flagged for
further investigation. Pattern tests place upper limits on:
The individual concentration value (maximum-hour test),
• The difference in adjacent concentration values (adjacent hour test),
• The difference or percentage difference between a value and both of its adjacent values
(spike test), and
• The average of three or more consecutive values (consecutive value test)
22.3.3 Parameter Relationship Tests
Parameter relationship tests can be divided into deterministic tests involving the theoretical
relationships between parameters (e.g., ratios between benzene and toluene) or empirical tests
which determine whether or not a parameter is behaving normally in relation to the observed
behavior of one or more other parameters. Determining the "normal" behavior of related
parameters requires the detailed review of historical data.
-------
Project: Model QAPP
Element No: 23
Re vision No: 1.0
Date: 11/04/02
Page 1 of 3
23.0 Data Validation, Verification and Analysis
Many of the processes for verifying and validating the measurement phases of the data collection
operation have been discussed in Section 22. If these processes, as written in the QAPP, are
followed, and the sites are representative of the boundary conditions for which they were selected,
one would expect to achieve the DQOs. However, exceptional field events may occur, and field
and laboratory activities may negatively impact the integrity of samples. In addition, it is expected
that some of the QC checks will fail to meet the acceptance criteria. This section will outline
how the District will take the data to a higher level of analysis. This will be accomplished by
performing software tests, plotting and other methods of analysis.
23.1 Process for Validating and Verifying Data
23.1.1 Verification of Samples
After a sample batch is completed, a thorough review of the data will be conducted for
completeness and data entry accuracy. All raw data that is hand entered on data sheets will be
double keyed as discussed in Section 19, into the LIMS. For the chromatographic data, the data
will be transferred from a Level 1 to a Level 2 status. The entries are compared to reduce the
possibility of entry and transcription errors. Once the data is entered into the LIMS, the system
will review the data for routine data outliers and data outside of acceptance criteria. These data
will be flagged appropriately. All flagged data will be "re-verified" that the values are entered
correctly. The data qualifiers or flags can be found in the SOPs.
23.1.2 Validation
Validation of measurement data will require two stages: one at the Level I and the Level II.
Records of all invalid samples will be filed for 5 years. Information will include a brief summary
of why the sample was invalidated along with the associated flags. This record will be available
on the LIMS since all samples that were analyzed will be recorded. At least one flag will be
associated with an invalid sample, that being the "INV" flag signifying invalid, or the "NAR" flag
when no analysis result is reported, or "BDL" which means below the detection limit. Additional
flags will usually be associated with the NAR, INV or BDL flags that help describe the reason for
these flags, as well as free form notes from the field operator or laboratory technician.
-------
Validation of Measurement Values
Project: Model QAPP
Element No: 23
Re vision No: 1.0
Date: 11/04/02
Page 2 of 3
Certain criteria based upon field operator and laboratory technician judgement have been
developed that will be used to invalidate a sample or measurement. The flags listed in table 22-1
will be used to determine if individual samples, or samples from a particular instrument will be
invalidated. In all cases the sample will be returned to the laboratory for further examination.
When the laboratory technician reviews the field sheet and chain-of-custody forms he/she will
look for flag values. Filters that have flags related to obvious contamination (CON), filter
damage (DAM), field accidents (FAC) will be immediately examined. Upon concurrence of the
laboratory technician and laboratory branch manager, these samples will be invalidated. The flag
"NAR" for no analysis result will be placed in the flag area associated with this sample, along
with the other associated flags.
Other flags listed may be used alone or in combination to invalidate samples. Since the possible
flag combinations are overwhelming and can not be anticipated, the air division will review these
flags and determine if single values or values from a site for a particular time period will be
invalidated. The division will keep a record of the combination of flags that resulted in
invalidating a sample or set of samples. As mentioned above, all data invalidation will be
documented. Table 23.1 contains criteria that can be used to invalidate single samples based on
single flags.
Table 23.1 Single Flag Invalidation Criteria for Single Samples
Requirement
Contamination
Filter Damage
Event
Laboratory
Accident
Below Detection
Limit
Field Accident
Flag
CON
DAM
EVT
LAC
BDL
FAC
Comment
Concurrence with lab technician and branch manager
Concurrence with lab technician and branch manager
Exceptional , known field event expected to have affected
sample . Concurrence with lab technician and branch
manager
Concurrence with lab technician and branch manager
Value is below the Minimum Detection Limit of the
analytical system
Concurrence with lab technician and branch manager
-------
Project: Model QAPP
Element No: 23
Re vision No: 1.0
Date: 11/04/02
Page 3 of 3
23.2 Data Analysis
Data analysis refers to the process of attempting to make sense of the data that are collected. By
examining the list in Table 5-1, there are a large number of parameters to analyze. However,
many of these have similar characteristics: Volatile Organics, Semi-Volatile Organics and
paniculate metals. One would assume that there physical and chemical properties could group
them together.
23.2.1 Analytical tests
The District will employ several software programs towards analyzing the data. These are listed
below with a short explanation of each.
Spreadsheet - The District will perform a rudimentary analysis on the data sets using EXCEL™
spreadsheets. Spreadsheets allow the user to input data and statistically analyze, plot and graph
linear data. This type of analysis will allow the user to see if there are any variations in the data
sets. In addition, various statistical tests such as tests for linearity, slope, intercept or correlation
coefficient can be generated between two strings of data. Box and Whisker, Scatter and other
plots can be employed. Time series plots can help identify the following trends:
• Large jumps or dips in concentrations
• periodicity of peaks within a month or quarter
• Expected or un-expected relationships among species
VOCDat- As stated in Section 22, the EPA has placed resources into creating software that can
analyze data. One such program is VOCDat. This software program was originally written for
input of PAMS data. VOCDat is a Windows-based program that provides a graphical platform
from which to display collected VOC data; to perform quality control tasks on the data; and for
exploratory data analysis. This program will enable the TCAPCD to rapidly validate and release
their air toxics VOC data to AQS. VOCDat displays the concentrations of the VOC data using
scatter, fingerprint, and time series plots. Customizable screening criteria may be applied to the
data and the quality control codes may be changed for individual data points as well as for the
entire sample on all plots. VOCDat can allow a user to find out what percentage a particular
compound is of the total. This test allows the user the ability to see if the data exceeds the 3
sigma rule for outliers. For more details, please see Section 22.2.
-------
Project: Model QAPP
Element No: 23
Re vision No: 1.0
Date: 11/04/02
Page 4 of 3
Wind Rose Plots - Recently the TCAPCD has purchased a wind rose program that will except
pollutant data. The wind direction, wind speed and pollutant data will be input into the program
and wind rose which show the relative direction and speed of pollutants (transport) will be
graphically displayed.
GIS - GIS program that allows the user the ability to overlay concentration data on geographic
data. By creating "views", the user can overlay temporally changing data into a spatial analysis
too. Plots of concentrations of data can be temporal/spatially displaced.
-------
Project: Model QAPP
Element No: 24
Revision No: 1.0
Date: 11/04/02
Page 1 of5
24.0 Reconciliation with Data Quality Objectives
24.1 Reconciling Results with DQOs
The DQOs for the air toxics monitoring network were developed in Section 7. This is stated
below.
1. Detect a percent difference change between successive three-year average
concentration levels that are greater than or equal to 15 percent
In addition, for the rest of the air toxics systems in the network:
2. Determine the highest concentrations expected to occur in the area covered by the
network, i.e., to verify the spatial and temporal characteristics of HAPs within the
city.
This section of the QAPP will outline the assessment procedures that Toxa City will follow to
determine whether the monitors and laboratory analyses are producing data that comply with the
stated goals. This section will then clearly state what action will be taken as a result of the
assessment process. Such an assessment is termed a Data Quality Assessment (DQA) and is
thoroughly described in EPA QA/G-9: Guidance for Data Quality Assessment1.
For the stated DQO, the assessment process must follow statistical routines. The following five
steps will discuss how this will be achieved. Please note that OAQPS has notified the District
that it will perform DQAs of the data from a national perspective. Therefore, the District will
allow OAQPS to perform these assessments on behalf of the District. The DQAs that will be
performed by the District will pertain to the data collected at all five sites and the DQAs will
pertain to answering the second statement.
24.2 Five Steps of DQA Process
As described in EPA QA/G-9, the DQA process is comprised of five steps. The steps are
detailed below.
24.2.1 Review DQOs and Sampling Design
Section 7 of this QAPP contains the details for the development of the DQOs, including defining
the objectives of the air toxics monitoring network, and developing limits on the decision errors .
-------
Project: Model QAPP
Element No: 24
Revision No: 1.0
Date: 11/04/02
Page 2 of5
Section 10 of this QAPP contains the details for the sampling design, including the rationale for
the design, the design assumptions, and the sampling locations and frequency. If any deviations
from the sampling design have occurred, these will be indicated and their potential effect
carefully considered throughout the entire DQA. The State Agency performs annual network
reviews. The TCAPCD will request that the State Agency review the network siting and
maintenance.
24.2.2 Conduct Preliminary Data Review
A preliminary data review will be performed to uncover potential limitations to using the data, to
reveal outliers, and generally to explore the basic structure of the data. The first step is to review
the quality assurance reports. The second step is to calculate basic summary statistics, generate
graphical presentations of the data, and review these summary statistics and graphs.
Review Quality Assurance Reports.- Toxa City will review all relevant quality assurance
reports, internal and external, that describe the data collection and reporting process. Particular
attention will be directed to looking for anomalies in recorded data, missing values, and any
deviations from standard operating procedures. This is a qualitative review. However, any
concerns will be further investigated in the next two steps.
24.2.3 Select the Statistical Test
Toxa City will generate summary statistics for each of its primary and QA samplers. The
summary statistics will be calculated at the annual, and a three-year levels and will include only
valid samples. The following statistical test will be performed:
• Test to examine distribution of the data
• Simple annual and 3-year averages of all pollutants for examination of trends
• Examination of precision of the data as described in Table 19.6
• Seasonal averages to determine any seasonal variability
Particular attention will be given to the impact on the statistics caused by the observations noted
in the quality assurance review. In fact, Toxa City may evaluate the influence of a potential
outlier by evaluating the change in the summary statistics resulting from exclusion of the outlier.
Toxa City will generate some graphics to present the results from the summary statistics and to
show the spatial continuity over Toxa City. Maps will be created for the annual and three-year
-------
Project: Model QAPP
Element No: 24
Revision No: 1.0
Date: 11/04/02
Page 3 of5
means, maxima, and inter-quartile (i.e., seasonal) ranges for a total of 6 maps. The maps will
help uncover potential outliers and will help in the network design review. Additionally, basic
histograms will be generated for each of the primary and QA samplers and for the percent
difference at the collocated sites. The histograms will be useful in identifying anomalies and
evaluating the normality assumption in the measurement errors. GIS spatial analysis will also be
performed to see if meteorology and topography have any influence on the concentrations.
24.2.4. Verify Assumptions of Statistical Test. There are no NAAQS to compare with air
toxics. Therefore, verification of the data must be done against estimated values, such as
models. However, before this can occur, the distribution, tests for trends, tests for outliers must
be examined.
Normal distribution for measurement error- Assuming that measurement errors are normally
distributed is common in environmental monitoring. Toxa City has not investigated the
sensitivity of he statistical test to violation of this assumption; although, small departures from
normality generally do not create serious problems. Toxa City will evaluate the reasonableness
of the normality assumption by reviewing a normal probability plot and employing the
Coefficient of Variance Test. If the plot or statistics indicate possible violations of normality,
Toxa City may need to determine the sensitivity of the DQOs to departures in normality.
Trends Analysis- It is recommended that a simple linear regression test be performed to observe
the temporal variations in the data sets. Air toxics data can be roughly divided into two
categories: Point and area sources. In terms of area sources, of which many of these may be
mobile sources, one would assume that mobile related toxics would vary with the diurnal
variations of traffic in urban and suburban environment. The linear regression test would
provide information on whether certain compounds are tied to mobile sources. For instance,
benzene is identified as major mobile HAP. If a linear regression is performed against a
compound whose source is unknown, then a small correlation coefficient would provide
information on its possible source. In addition to the linear regression test, it is recommended
that annual and 3-year average trend plots be generated. These plots can give a long-term
temporal information. It will also allow the TCAPCD the justification to decrease the network if
trends illustrate that the values are also decreasing.
Measurement precisions- For each sampling system, TCAPCD will review the 95% confidence
limits as determined in Table 19.2. If any exceed 10%, Toxa City may need to determine the
sensitivity of the DQOs to larger levels of measurement imprecision. Before describing the
algorithm, first some ground work. Toxa City's strategy for accomplishing this will be to use all
available quarters of data as the basis for projecting where the precision estimates will be at the
end of the three-year monitoring period.
-------
Project: Model QAPP
Element No: 24
Revision No: 1.0
Date: 11/04/02
Page 4 of 5
Toxa City will develop confidence intervals for the precision estimates. This will be
accomplished using a re-sampling technique. The protocols for creating the confidence intervals
are described in the following equation. OAQPS has notified all of the State and Local agencies
involved in the NATTS that it will calculate network/program bias using round robin samples.
Please see Reference 1 .
Precision Algorithm
For each measurement pair, calculate the coefficient of variation according to Equation 20 from
Section 14 and repeated below:
Summarize the 95% confidence Limits to the quarterly level, according to where the number of
collocated pairs in quarter.
di = Y;-X; x 100
Upper 95% Confidence Limits: Limit =d, _1.96*S, /• 2
Upper 95% Confidence Limits: Limit =d, +1.96*8, /• 2
24.2.5 Draw Conclusions from the Data.
If the sampling design and the statistical test bear out, it can be assumed that the network design
and the uncertainty of the data are acceptable. This conclusion can then be written in the Annual
Report to management. Management may then decide whether to perform risk assessments,
allow the State and EPA to analyze the data or work closely with the nearby university to
determine whether this data can be used to assess conclusion from health effects studies.
24.26 Action Plan Based on Conclusions from DQA
A thorough DQA process will be completed during the summer of each year. For this section,
Toxa City will assume that the assumptions used for developing the DQOs have been met. If
this is not the case, Toxa City must first revisit the impact on the precision limits determined by
the DQO process. At some point in time, it may be necessary to reduce the network. This
would happen under the following scenario.
-------
Project: Model QAPP
Element No: 24
Revision No: 1.0
Date: 11/04/02
Page 5 of5
• The data at a particular location shows values that are very low or at the detection limit. If
this occurs it will be the District's option to re-locate the sampler or remove it from service.
• Vandalism or loss of right of way
References
1. Quality Management Plan for the National Air Toxics Trends Assessment Monitoring
Program, EPA document, 454/R-02-0006, November 2002
2. Guidance for the Data Quality Assessment Process EPA QA/G-9 U.S. Environmental
Protection Agency, QAD EPA/600/R-96/084, July 1996.
3. U.S. EPA (1997b) Revised Requirements for Designation of Reference and Equivalent
Methods for Air toxics and Ambient Air Quality Surveillance for Particulate Matter-Final
Rule. 40 CFR Parts 53 and 58. Federal Register, 62(138):38763-38854. July 18,1997.
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1. REPORT NO. 2.
EPA-454/R-02-007
4. TITLE AND SUBTITLE
Quality Assurance Guidance Document -
Model Quality Assurance Project Plan for the
National Air Toxics Trends Stations
7. AUTHOR(s): Dennis Mikel
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, NC 2771 1
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Air Quality Planning and Standards
Emission, Monitoring and Analysis Division
U.S. Environmental Protection Agency
Research Triangle Park, NC 2771 1
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE 12/02
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
1 1 . CONTRACT/GRANT NO.
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The Quality Assurance Document for the National Air Toxics Trends Stations was
written for air pollution control agencies to use as a model for the quality assurance
project plans that the agencies need to fulfill their QA requirements. This model
QAPP incorporates all of the elements needed to satisfy the R-5 EPA Requirements for
Quality Assurance Project Plans.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
18. DISTRIBUTION STATEMENT
Release Unlimited
b. IDENTIFIERS/OPEN ENDED
TERMS
Air Pollution control
19. SECURITY CLASS (Report)
Unclassified
20. SECURITY CLASS (Page)
Unclassified
c. COSATT Field/Group
21. NO. OF PAGES 1 42
22. PRICE
------- |