Quality Assurance Project Plan
for the
Great Lakes Water Quality Surveys
Revised
March 2003
Prepared by
Great Lakes National Program Office
U.S. Environmental Protection Agency
77 West Jackson Boulevard
Chicago, Illinois 60604
Distribution List
Paul Bertram, GLNPO
Kenneth Klewin, GLNPO
David Rockwell, GLNPO
Ship Operations Contractor Lead
Louis Blume, GLNPO
Marvin Palmer, GLNPO
Glenn Warren, GLNPO
Chemical Hygiene Officer
Paul Horvatin, GLNPO
George Ison, GLNPO
GLAS Project Manager
Grantee Principal Investigator
-------
-------
Table of Contents
1.0 Project Background Page 1
2.0 Project Description Page 1
3.0 Project Organization Page 5
3.1 Survey Project Management Page 5
3.2 On-Ship Roles and Responsibilities Page 11
4.0 Data Quality Objectives Page 15
4.1 Project Quality Objectives Page 15
4.2 Measurement Quality Objectives Page 16
5.0 Special Training Requirements Page 22
6.0 Documentation and Records Page 23
6.1 Sample Identification and Labeling Page 23
6.2 Record keeping Page 25
7.0 Sampling Process Design Page 27
7.1 Site Selection Strategy Page 27
7.2 Depth Selection Criteria Page 28
7.3 Sampling Sequence/Frequency Strategy Page 31
8.0 Sampling Method Requirements Page 33
8.1 Sampling with the Rosette Page 35
8.2 Zooplankton Sampling Tows Page 36
8.3 Benthic Invertebrate Sampling Methods Page 37
8.4 Meteorological Methods Page 37
9.0 Sample Handling and Custody Requirements Page 37
9.1 Rosette Sample Handling Page 38
9.2 Zooplankton Sample Handling Page 39
9.3 Benthic Invertebrate Sample Handling Page 39
10.0 Analytical Methods Requirements Page 42
10.1 Nitrate/Nitrite Page 42
10.2 Total Phosphorus and Total Dissolved Phosphorus Page 42
10.3 Chloride and Silica Page 42
10.4 Particulate Organic Carbon Page 42
10.5 Particulate Nitrogen Page 43
10.6 Particulate Phosphorus Page 43
10.7 Dissolved Organic Carbon Page 43
10.8 Total Suspended Solids Page 44
10.9 Specific Conductance, Total Alkalinity, Turbidity, and pH Page 44
10.10 Dissolved Oxygen Page 44
10.11 Phytoplankton Page 45
10.12 Zooplankton Page 45
10.13 Chlorophyll a Page 45
-------
10.14 Benthic Invertebrates Page 46
10.15 Meteorology Measurements Page 46
11.0 Quality Control Requirements Page 46
12.0 Instrument/Equipment Testing, Inspection, and Maintenance Requirements
Page 57
13.0 Instrument Calibration and Frequency Page 58
14.0 Inspection/Acceptance Requirements for Supplies and Consumables Page 60
15.0 Requirements for Acquisition of Non-direct Measurement Data Page 60
16.0 Data Management Page 60
16.1 Field Data Management Page 60
16.2 GLENDA Data Management Page 61
17.0 Assessments and Response Actions Page 62
17.1 Surveillance Page 62
17.2 Peer Review Page 62
17.3 Quality System Audits Page 62
17.4 Readiness Reviews Page 62
17.5 Technical Systems Audit Page 63
17.6 Data Quality Audits Page 63
17.7 Data Quality Assessment Page 63
18.0 Reports to Management Page 66
19.0 Data Verification and Validation Page 66
20.0 References Page 66
Attachment a - monitoring station and depths Page A-l
-------
1.0 Project Background
EPA's Great Lakes National Program Office (GLNPO) conducts biannual surveys of water quality within
each of the Great Lakes. These surveys are conducted every spring and summer to assist the United
States in fulfilling its responsibilities under the Great Lakes Water Quality Agreement.1 Annex 11 of this
agreement requires the U.S. and Canada to conduct surveillance and monitoring activities to:
(1) Evaluate the degree to which jurisdictional control requirements are being met.
(2) Assess water quality trends to evaluate the effectiveness of existing control measures and
determine if new control measures are needed, evaluate enforcement and management strategies,
and identify the need for new research and technology development.
(3) Identify emerging problems in the Great Lakes Basin Ecosystem.
In response to these requirements, GLNPO has been conducting water quality surveys since 1983.
Initially, the surveys focused on chemical eutrophication and whole lake response to changes in
phosphorous loadings. Therefore, early monitoring efforts were directed at Lakes Michigan, Erie, and
Huron. Lake Superior was excluded from early efforts because it is not affected by eutrophication, and
Lake Ontario was not sampled because extensive annual monitoring of that lake was already being
implemented by the Canadian government. Lake Superior was added in 1992 to provide EPA with
information on a healthy lake and dovetail with environmental monitoring initiatives being pursued under
the Environmental Monitoring and Assessment Program (EMAP). Lake Ontario was added later to
supplement the Canadian data gathering activities. The scope of the survey parameters has similarly
broadened overtime and now includes several measures of biological and sediment quality in the lakes.
The Water Quality Surveys are unique in that all five lakes are sampled by one agency, using one vessel
and one principal laboratory for each parameter group. Thus inter-lake comparisons based on data
collected during the program are not complicated by differences in sampling procedures, interlaboratory
differences, or analytical techniques.
2.0 Project Description
GLNPO has primary responsibility within the US for conducting surveillance monitoring of the offshore
waters of the Great Lakes. The Water Quality Surveys generally consist of two surveys per year: a Spring
Survey and a Summer Survey.2 The monitoring effort is focused on whole lake responses to changes in
loadings of anthropogenic substances. Generally speaking, change in the water column is best detected in
areas of minimal variability and change in sediment quality is best detected in shallower areas where
physical and chemical impacts can occur within reasonably short time frames. As a result, sampling
activities in the surveys are largely restricted to the relatively homogeneous offshore waters of each lake
(for water column measurements) and to the shallow areas of these waters for sediment measurements.
To ensure that sampling activities are representative of lake conditions, samples are collected from
multiple sites within each lake basin. The number and locations of the sites needed to obtain a
representative sampling of each basin was statistically determined using historical data collected during
intensive surveys of each lake. Each basin consists of several routine monitoring stations and a 'master
station'. The master stations generally represent the deepest area of the basin and are often used to collect
supplementary data for other (non-survey) purposes.
'Great Lakes Water Quality Agreement of 1972, renewed in 1978, amended in 1987.
2 The Summer Survey is periodically canceled to allow for dry dock and repairs to the ship.
March 2003 Page 1
-------
Sampling activities also are conducted two times per year to capture information about the lakes during
different conditions. The spring surveys are designed to collect water quality information during
unstratified (isothermal) conditions of the lake, so the survey circuit is planned to move from warmest to
coolest waters to ensure that sampling at all sites is conducted before stratification begins. The summer
surveys are designed to monitor the quality of each lake during stratified conditions. As a result, the
number of depths sampled during the summer is slightly greater than the number of depths sampled
during the spring surveys.
Survey activities are conducted onboard EPA's R/VLake Guardian, a former offshore oil field supply
vessel built by Halter Marine, Moss Point, MS, in 1981. Ship dimensions are: 180' length, 40' beam, 11'
draft, and 850 displaced tonnage. Propulsion is twin screws enclosed in Kort nozzles and driven by 1200
hp Caterpillar diesel engines. Cruising speed is 11 knots; range is 6,000 miles. Sampling activities
onboard the Lake Guardian are primarily implemented through the use of:
A Rosette sampling device deployed from the ship to collect samples for a variety of nutrients,
physical parameters, and biological parameters
Tow nets used to collect zooplankton samples, and
A Ponar grab sampling device used to collect sediment samples for benthic organisms and
chemical measurements.
Most of the survey measurements are made on board the ship, either on the bridge or deck (e.g.,
meteorological measurements such as wind speed and direction, wave height and direction, air
temperature, etc.), by the conductivity/temperature/depth (CTD) probe attached to the Rosette, or in the
on-ship laboratory (e.g., turbidity, conductivity, pH, etc.). The remaining measurements are made by
GLNPO's Great Lakes Analytical Services (GLAS) contractor and a biology grantee in land-based
laboratory facilities. Table 2-1 provides a detailed list of all parameters monitored during the surveys.
The surveys are timed to occur at roughly the same time each year, with minimal variations allowed for
weather conditions (e.g., ice dispersal, storms, etc.) and other scheduling constraints such as the
availability of services while in port.
The Spring Survey generally lasts 40 days, with the first 25 days covering a circuit that runs from
Milwaukee, WI through Lakes Michigan, Huron, Erie, and Ontario, returning through Lake Erie
to Bay City, MI. Lake Superior is sampled last, after ice has sufficiently dispersed. The Lake
Superior sampling period, including transit, generally lasts approximately 15 days.
The Summer Survey generally takes 35 days beginning in the first week of August; it covers a
transit from Bay City, MI through Lakes Erie and Ontario, and returning through Lakes Erie,
Huron, and Michigan and then transiting to Lake Superior.
Page 2
March 2003
-------
Table 2-1. List of Parameters Monitored in the Great Lakes Water Quality Surveys
Parameter Class
Parameter
Sampling Device/Measurement Location
Nutrients
Nitrate plus Nitrite Nitrogen (NO2/NO3)
Particulate Organic Carbon (POC)
Total Phosphorus
Total Dissolved Phosphorus (TDP)
Rosette/land-based lab
Particulate Nitrogen
Chloride
Particulate Phosphorus
Reactive Silica
Physical Measurements
Aesthetics
Observation/Bridge
Optical Transmittance
SeaBird CTD
Air temperature
Digital Thermometer/Bridge
Water T emperature
SeaBird
Turbidity
Rosette/Board
Wind Speed
Danforth/Bridge
Wind Direction
Danforth/Bridge
Specific Conductivity
Rosette/Board
Alkalinity
Rosette/Board
pH
Rosette/Board
Water Clarity
Secchi Disk/Deck
Total Suspended Solids
Rosette/land-based lab
Wave Height
Observation/Bridge
Dissolved Oxygen
Rosette/Board and SeaBird CTD probe
Site Location
Digital GPS/Bridge
Biological Indicators
Phytoplankton
Rosette/land-based lab
Zooplankton
Tow nets/land-based lab
Benthic Invertebrates
Ponar Grab/land-based lab
Chlorophyll a
Rosette/land-based lab
* Dissolved organic carbon (DOC) also is determined when hydrophobic organic compounds are included in special
studies.
In addition to the general Water Quality Monitoring Surveys described above, GLNPO conducts an
intensive dissolved oxygen (DO) survey of Lake Erie each summer. Details of this survey can be found
in Dissolved Oxygen and Temperature Profiles for the Central Basin of Lake Erie Quality Assurance
Project Plan, located in Appendix D of the manual, Sampling and Analytical Procedures for GLNPO's
Open Lake Survey of the Great Lakes. The United States has made a substantial (approximately $8
billion) investment in pollution controls to address severe oxygen impairment problems within Lake Erie,
and the purpose of GLNPO's annual DO surveys is to determine if these investments are resulting in
demonstrable improvements to the lake. To the extent possible, the Lake Erie DO Surveys are timed to
coincide with the summer Water Quality Surveys onboard the R/VLake Guardian. Similarly, GLNPO
attempts to coordinate its spring and summer survey activities with other organizations, such as EPA
Regions 2 and 5, who are involved in Great Lakes research activities. The goal of this coordination is to
maximize sampling activities onboard the R/V Lake Guardian.
This Quality Assurance Project Plan (QAPP) presents the data quality objectives (DQOs) and
measurement quality objectives (MQOs) established for the collection and analysis of environmental
samples collected during each Water Quality Survey. This QAPP also describes the roles and
responsibilities of the organizations and staff participating in the Surveys and the methods and procedures
that will be followed to ensure that Survey DQOs and MQOs are met. This QAPP does not address the
activities of other projects, such as the Lake Erie DO Survey, that are related to or being conducted
concurrently with the Water Quality Surveys. QA requirements and management practices for those
studies are described in separate, project-specific QA plans.
March 2003
Page 3
-------
This document was prepared in accordance with and contains each of the elements described in the most
recent version of EPA's requirements for QAPPs [1]. To improve clarity, the order of certain elements in
this QAPP has been modified slightly from the order presented in EPA QA/R-5. For example, the Project
Organization section (element "A4" in QA/R-5) comes after the Project Background and Project
Description (elements "A5" and "A6" in QA/R-5) in this QAPP.
In accordance with the guidance provided in QA/R-5, this QAPP is considered to be a dynamic document
that is subject to change as sample collection and analysis progresses. All changes to procedures
described in this QAPP will be reviewed by the GLNPO Quality Assurance Manager, the Environmental
Monitoring and Indicators Team Lead, and the appropriate technical leads (e.g., biology, dissolved
oxygen, limnology, board chemistry, information management, or bridge measurements), to determine if
the changes significantly impact the technical and quality objectives of the project. If changes are
deemed to be significant, the QAPP will be revised accordingly.
This QAPP is an appendix to the manual, Sampling and Analytical Procedures for GLNPO's Open Lake
Water Survey (WQS manual). The procedures and requirements discussed in this QAPP are further
described in the WQS manual. The manual is organized into six chapters. Chapter 1 describes the WQS
and includes a Standard Operating Procedure (SOP) for General Shipboard Scientific Operations. This
SOP presents information on: 1) roles and responsibilities, 2) the sequence of sampling events, and 3)
safety and training. Chapter 1 also includes an SOP for Electronic Field Information Recording.
Chapter 2 provides sampling and analytical procedures for nutrient parameters targeted in the survey.
Chapter 3 provides sampling and analytical procedures for physical parameters including meteorological
data and total suspended solids. Chapter 4 provides sampling and analytical procedures for biological
parameters. Chapter 5 provides analytical procedures for several chemical parameters that are analyzed
in the laboratory aboard the R/VLake Guardian ("board chemistry" parameters) including pH, specific
conductivity, total alkalinity, turbidity and dissolved oxygen by the Winkler method. Chapter 6 provides
sampling and analytical procedures for nutrients in sediments.
This manual also contains several appendices. Appendix A contains maps of the Great Lakes spring and
summer survey stations. Appendix B contains the quality assurance project plan (QAPP) for the Great
Lakes Water Quality Surveys. Appendix C contains a staff scheduling form that lists shift dates for
proposed and actual GLNPO staff participating in the survey. Appendix D contains a QAPP as well as
the sampling and analytical procedures used in GLNPO's intensive DO survey of Lake Erie's Central
Basin. Appendix E contains survey planning forms that must be submitted by researchers requesting use
of the R/VLake Guardian for sampling purposes. Appendix F contains the self-certification form that all
GLNPO staff and contractors/grantees participating in the survey must complete and provide to the
GLNPO QA Manager to certify their meeting pre-survey training requirements as specified on the form.
Appendix G contains the Medical History Questionnaire that is required to be completed and submitted
by all personnel that participate in GLNPO's WQS. Appendix H contains hard-copy field information
recording forms used to document the data generated during the surveys. Appendix I contains lists of
acceptable field remark, lab remark, and quality control sample identifier codes used in GLNPO's Great
Lakes Environmental Database (GLENDA). Appendix J contains a user's manual for the GLENDA
remote data entry tool. Appendix K includes a list of current GLENDA analyte names and codes for
environmental parameters being measured. Appendix L contains a list of roles and responsibilities for
the Chief Scientist, in addition to a list of priorities for each staffing period and a checklist that must
completed by the shift supervisor for each shift of 12 hours and provided to the QA Manager to verify
that they met their responsibilities involving ship operations and survey data management. Appendix M
contains a suggested revision sheet that survey participants should use to document issues, errors, and
suggested revisions and clarifications to the manual. Appendix N contains a data status tracking sheet to
track specific data sets throughout the data verification process and upload to the GLENDA database.
Appendix O contains a data discrepancy form to initiate a revision to the GLENDA database when errors
Page 4
March 2003
-------
are identified. Appendix P contains a summary of the sample depth's collected as part of the WQS for
all five lakes. Finally, Appendix Q contains the station location change form that must be completed and
approved by pertinent survey participants to finalize a change to a station location. Earlier revisions of
the manual contained: 1) an addendum to the WQS QAPP that provides the information needed to
perform the base monitoring program activities on alternate vessels when the R/VLake Guardian is not
available (March 2001, Appendix C) and 2) GLNPO's Standard Operating Procedures for Winter
Operations (March 2001, Appendix F). These documents are available to support specific projects
involving these special circumstances.
The WQS manual also is a dynamic document and in accordance with GLNPO's document and records
management policy, a control copy of the manual (and this QAPP therein) is maintained by GLNPO's
document control coordinator. The current version of the manual can be obtained from GLNPO's
Environmental Monitoring and Indicators Team Lead, the Quality Assurance Manager, and GLNPO's
document control coordinator within GLNPO's Policy Coordination and Communications Branch.
3.0 Project Organization
The Great Lakes National Program Office (GLNPO) is responsible for overall management of the Water
Quality Surveys. GLNPO also is responsible for providing oversight over all scientific activities while
onboard ship and for performing many of the sampling and analysis activities onboard ship. GLNPO is
supported by a grantee and several contractors, including:
A grantee organization that provides scientific and technical support in the field and in a land-
based laboratory facility for collection and biological analysis of samples.
A Great Lakes Analytical Services (GLAS) contractor that provides scientific and technical
support in the field and in a land-based facility for collection and chemical analysis of samples.
A Ship Operations contractor responsible for all aspect of ship operations.
A Chemical Hygiene contractor that provides a Chemical Hygiene Officer for all activities in
which scientific operations are being conducted onboard ship.
Section 3.1 describes the overall project management and lines of authority within GLNPO and during
the water quality surveys. It includes an organization chart illustrating the general organization of
GLNPO (Figure 3-1) and detailed descriptions of the roles and responsibilities of GLNPO staff that are
involved with management and implementation of the Water Quality Surveys. Section 3.2 describes roles
and responsibilities of survey staff while onboard the Lake Guardian. In this section, Table 3-1
illustrates the lines of authority and responsibilities of survey participants while onboard ship. Section 3.3
briefly describes the major roles and responsibilities that must be defined by all grantees and contractors
supporting the surveys.
3.1 Survey Project Management
GLNPO is organized into functional teams that specialize in specific areas of interest and expertise for
data gathering activities. Each team is comprised of a team lead and staff from GLNPO and where
appropriate, other organizations within EPA (such as Region 5, Region 2, etc). Five of the GLNPO teams
are directly involved in the Water Quality Surveys:
1) Environmental Monitoring and Indicators Team
2) Health and Safety team
3) Information Management and Data Integration Team
4) QA Team
5) Management Team
March 2003
Page 5
-------
Project management responsibilities within the Water Quality Surveys reflect the emphasis on these four
teams, all of whom ultimately report to the director of GLNPO. Specific roles and responsibilities within
each of these areas is described below.
Page 6
March 2003
-------
TL - Technical Lead
Page 7
-------
3.1.1 Director of the Great Lakes National Program Office
The GLNPO Director, Gary Gulezian, is responsible for providing financial and staff resources
necessary to meet the objectives and implement the requirements described in this QAPP for the
Water Quality Surveys. The Director is responsible for establishing GLNPO quality policy and
resolving related issues which are identified through the Quality Management Team and survey
participants.
3.1.2 QA Manager
The GLNPO QA Manager, Louis Blume, is independent of the Water Quality Surveys project
and reports directly to the GLNPO Director. The QA Manager is responsible for assisting the
Chief of the Monitoring Indicators and Reporting Branch, the Environmental Monitoring Team
Lead, the Technical Leads, and the EPA Project Officer for the Ship Operations Contractor
(SOC) with the development and implementation of QAPPs that are consistent with GLNPO's
Quality Management Plan and quality system. The QA Manager also is responsible for:
Concurring with the Technical Leads on all methods and method modifications selected
for use in the surveys
Ensuring that all QA procedures described in this QAPP are followed, reporting any
deviations from this QAPP to the Team Leads, Technical Leads, and SOC Project Officer
in implementing corrective actions necessary to resolve these deviations
Conducting periodic Quality System Audits (QSAs) to verify that survey procedures are
capable of meeting survey objectives
Ensuring that all staff participating in the surveys complete a Self-Certification Form that
documents they have received all required training
Ensuring that the Shift supervisors submit a completed Shift Supervisor Checklist
(Appendix L), and using information from this checklist to identify roles and
responsibilities that require revision or clarification.
The QA Manager reports directly to the GLNPO Director.
3.1.3 Environmental Monitoring and Indicators Staff
The Chief of GLNPO's Monitoring Indicators and Reporting Branch is responsible for providing
overall direction concerning all aspects of the Water Quality Surveys. He is supported by an
Environmental Monitoring Team Lead and several Technical Leads. Roles and responsibilities
for each of these individuals are described below.
3.1.3.1 Monitoring Indicators and Reporting Branch Chief
The Monitoring Indicators and Reporting Branch Chief, Paul Horvatin, reports directly to
the GLNPO Director and is responsible for providing overall direction concerning the
surveys to the Environmental Monitoring Team Lead and Technical Leads. The Branch
Chief also is responsible for:
Implementing an overall QAPP applicable to all phases of the Water Quality
Surveys
Communicating survey objectives to the Environmental Monitoring Team Lead,
and Technical Leads
Ensuring EPA and contractor resources are available to implement the survey
design
Reviewing and approving all major work products associated with the surveys
Page 8
March 2003
-------
Participating in meetings with the Team Leader, Project Officers, other EPA
staff, and staff from other organizations and contractors concerning the surveys
Working with the GLNPO QA Manager to identify corrective actions necessary
to ensure that study objectives are met.
3.1.3.2 Environmental Monitoring Team Lead
The Monitoring Team Leader (Glenn Warren) reports directly to the GLNPO Director,
and is responsible for:
Developing and implementing the scientific aspects of this QAPP
Reporting to GLNPO's Management Team monthly regarding technical, QA,
and resource issues concerning the Water Quality Surveys
Daily oversight of EPA and contractor staff involved in all survey activities
related to sample collection, analysis, and data management for limnology (e.g,
nutrients and water chemistry) measurements
Providing as needed assistance to the Biology, Dissolved Oxygen, and Board
Chemistry Technical Leads with oversight of EPA, contractor, and grantee staff
involved in all water quality survey activities associated with sampling, analysis,
and data management activities in their functional areas
Reviewing and approving technical work products related to the Water Quality
Surveys
Participating in meetings with the various team leaders, Technical Leads, the
GLNPO QA Manager, and the GLNPO Director concerning Survey objectives,
schedules, and concerns.
3.1.3.3 GLNPO Technical Leads
The Water Quality Surveys are supported by several Technical Leads who provide
expertise and leadership in each of the survey's functional areas: board chemistry
(Marvin Palmer), biology (Marc Tuchman), dissolved oxygen (Paul Bertram), and
limnology (Glenn Warren).
Within GLNPO, the Technical Leads report to the persons indicated in Figure 3-1. For
the Water Quality Surveys, they also report to the Environmental Monitoring Team Lead.
Each Technical Lead is responsible for:
Developing and implementing the aspects of this QAPP that relate to their
functional area
Daily oversight of EPA, contractor, and grantee staff involved in all onboard
survey activities related to sample collection, analysis, and data management for
their functional area
Prior to each survey, randomly selecting site locations and depths for Rosette
field duplicates (Board Chemistry Technical Lead only)
Communicating survey objectives to all EPA, contractor, and grantee staff
involved in the collection and analysis of survey samples (for their functional
area)
Reviewing and approving major deliverables related to Survey sample collection
and analysis or data evaluation their functional area
Participating in meetings with the Monitoring Team Leader, other Technical
Leads, the GLNPO QA Manager, and the GLNPO Director concerning survey
objectives, schedules, and concerns
March 2003
Page 9
-------
Approving methodologies, including method modifications, to be used for
sampling and analysis of parameters in their functional area.
Completing the Chief Scientist Roles and Responsibilities (listed in Appendix L)
to verify that all project staff fulfilled the responsibilities described in this QAPP
and that he has verified the accuracy of data collected and recorded during the
cruise.
3.1.4 Ship Operations and Safety
Overall responsibility for safety operations within GLNPO lies with Louis Blume, the GLNPO
Safety Officer. Safety management responsibilities within the Water Quality Surveys is
delegated to George Ison, who also serves as the manager of Ship Operations during the Surveys.
This approach ensures that decisions related to ship operations do not adversely affect ship safety
procedures during the surveys (and vice versa). Specific responsibilities related of the Ship
Operations and Safety Manager with respect to the Water Quality Surveys are described in
Section 3.1.4.1. A second member of GLNPO's safety team is responsible for oversight of the
contractor-provided Chemical Hygiene Officer (CHO). Because GLNPO does not have expertise
in the area of chemical hygiene, this role is filled by a Safety Team member from Region 5.
Section 3.1.4.2 describes the roles and responsibilities of the CHO Manager.
3.1.4.1 Ship Operations and Safety Manager
George Ison is a member of GLNPO's Safety Team, and reports directly to the
Monitoring and Indicators Branch Chief. Under the Water Quality Surveys, Mr. Ison is
responsible for:
Oversight of the Ship Operations Contractor
• Scheduling all R/VLake Guardian activities
Developing and implementing procedures and training programs to ensure the
safety of all operations staff and scientists participating in the surveys (and in
other R/V Lake Guardian projects)
Ensuring that all designated ' 1st responders' have received required training
and/or re-certification as EMT's. (These 1st responders include the Chief
Scientists, Shift Supervisor, First Mate, and Second Mate).
Developing and implementing procedures to ensure that survey participants are
sufficiently healthy to participate in survey activities and that appropriate medical
precautions are taken when necessary
Assisting the Team Leaders and Data Managers in considering ship operations
and safety requirements when designing their survey objectives and plans
Ensuring avenues of communication to and from the ship. This is primarily
accomplished through electronic mail, cell phones, and facsimile. Responsible
for making available an emergency phone number via satellite during cruises to
allow for rapid communication in cases of emergencies.
3.1.4.2 Chemical Hygiene Officer (CHO) Manager
The Region 5 Safety Manager, Jim Finn, is responsible for oversight and management of
survey activities performed by the CHO contractor. The CHO is provided by the Public
Health Services. For the purposes of the Water Quality Surveys, the Region 5 Safety
Manager reports to the Ship Safety Manager and to the GLNPO Safety Team Lead.
Specific responsibilities include:
Participation in the GLNPO Safety Team
Page 10
March 2003
-------
Ensuring that the CHO contractor has sufficient funds and tasking to deliver a
qualified CHO for each survey
Oversight and technical direction of the CHO
Responding to technical questions and concerns raised by the CHO
Acts as contact person to locate reagents onboard when survey staff need them.
3.1.5 Information Management Team
All data management activities are overseen by GLNPO's Information Management Team Lead,
Pranus Pranckevicius, who reports directly to the director of GLNPO. He is responsible for
ensuring that the resources, systems, and staff necessary to store, retrieve, and manipulate data
gathered during GLNPO studies. He also is responsible for identifying and staying abreast of
emerging technologies to ensure that GLNPO data management functions are up to date.
The Information Management Team Lead is supported by Ken Klewin, manager of the Great
Lakes Environmental Database (GLENDA). GLNPO will begin capturing all survey data in
GLENDA beginning with the 2000 surveys. As the manager of the GLENDA database, Dr.
Klewin is responsible for:
Ensuring the database is available to store survey data in a timely fashion
Working with the Environmental Monitoring Team Lead, the Technical Leads, and the
QA Manager to ensure that any modifications needed to ensure compatibility of
GLENDA with survey requirements are implemented
Developing and implementing systems for upload of survey data into GLENDA
Developing and implementing appropriate database security and management procedures
Implementing effective data retrieval procedures to facilitate use of data by GLNPO staff,
contractors, grantees, and members of the scientific community
Oversight of any contractor support required to modify the GLENDA database structure.
3.2 On-Ship Roles and Responsibilities
Due to the scientific and navigational nature of the surveys, dual lines of leadership and authority exist
onboard ship. Ultimate responsibility and authority for all scientific and technical operations lies with
GLNPO's Chief Scientist. However, the ship Captain has ultimate responsibility for all maritime and
safety operations onboard ship, and the Captain has the authority to halt all scientific and technical
operations when s/he considers it necessary to ensure the safety of all passengers and crew. In addition,
contractors and grantees participating in the survey are responsible for reporting to their own management
as well as to EPA scientists who provide them with site-specific instruction regarding sample collection,
handling, and/or analysis. This instruction differs from technical direction in that it does not increase the
level of effort or cost of existing tasking and focuses on the minor technical details (such as when to drop
the winch) rather than significant instructions warranting formal contract management oversight.
All personnel that participate in GLNPO's WQS are required to complete and submit a Medical History
Questionnaire for safety reasons. The Captain maintains these questionnaires in medical history
envelopes for all survey participants in an onboard safe. This questionnaire is located in Appendix G.
GLNPO's Technical Leads (Figure 3-1) and other GLNPO staff serve as Chief Scientists onboard the
vessel. A single Chief Scientists is assigned for each cruise staffing change by the Monitoring Team Lead
through the Staffing Scheduling Form (Appendix C). Table 3-1 summarizes the roles and responsibilities
of key technical staff involved in Water Quality Survey activities onboard the R/VLake Guardian.
March 2003
Page 11
-------
3.3 Contractor and Grantee Staff
All contractors and grantees supporting the Water Quality Surveys must have approved Quality
Management Plans (QMPs) in place and operating prior to and throughout each survey. These approved
QMPs shall clearly identify and define roles, responsibilities, authority, and lines of communication for
all project staff involved in the management, implementation, QA, documentation, and survey activities.
This includes, but is not limited to, a Project Manager responsible for oversight of all project activities
performed by the contractor or grantee, a QA Coordinator or QA Officer that is independent of the
project, a data management coordinator responsible for ensuring that all required hardcopy and electronic
documentation is collected, stored, managed and reported in a manner that meets or exceeds the
requirements of this QAPP, and all technical staff required to implement the scientific and technical
activities (on ship and off ship) needed to fulfill the contractor's (or grantee's) requirements with respect
to the Water Quality Surveys. In addition, all contractors and grantees collecting environmental data in
support of GLNPO's water quality survey also have a QMP.
Page 12
March 2003
-------
Table 3-1. Roles, Responsibilities, Authority, and Lines of Communication Onboard the
R/VLake Guardian During Water Quality Surveys
Organization
Role
Responsibility
Authority
Lines of Communication
EPA
Chief Scientist
• All scientific and technical operations
onboard ship during the surveys
• Ensuring all requirements of this QAPP
are followed throughout the survey
• Ensuring that all EPA, contractor, and
grantee staff adhere to applicable SOPs
and collect and document all required
data at all stations
• Ensuring adherence to all ship
requirements (safety, etc) during surveys
• Ensuring the performance of and/or
participating in technical systems audits
• Ensuring transport of samples to the
contract laboratory
• Calls Science Meeting after ship sails
• Complete
authority for
all scientific
and technical
operations
• Communicates with Captain
or senior officer on bridge
• Communicates with EPA,
contractor, and grantee staff
• Reports to Monitoring Team
Lead or WQI Branch Chief
when in dock
Shift
Supervisor
• Back up to the Chief Scientist (see
above responsibilities)
• All scientific and technical operations
when Chief Scientist is off duty
• Complete
authority for
all scientific
and technical
operations
when Chief
Scientist is off
duty
• May be over-
ruled by Chief
Scientist
• Communicates with Captain
or senior officer on bridge
• Communicates direction to
all EPA, contractor, and
grantee staff during his/her
shift
• Reports to Chief Scientist
while onboard ship
Ship
Operations
Contractor
Captain
• Ultimate responsibility of ship safety
• Ensuring ship arrives at designated sites
on schedule established prior to cruise
• Oversight of all ship contractor staff
• Ensuring all survey participants have
practiced ship safety drill
• Collecting and documenting all bridge
data as described in this QAPP and
applicable SOPs
• Maintains medical history envelope for
survey participants in an onboard safe
• Complete
authority to
halt scientific
and technical
operations
when
necessary to
protect safety
of crew and
passengers
• Communicates direction to
all ship operations
contractor staff
• Communicates with Chief
Scientist for site-specific
instructions
• Contractually reports to Ship
Operations Project Officer
and Contracting Officer
First Mate (or
mate of the
watch)
• Back up to Captain when off duty (see
above responsibilities)
• Same as
above when
Captain is off
duty
• Reports to Captain
• Same as above when
Captain is off duty
Marine
Technicians
• Assist GLNPO scientists with sample
collection activities
• Assist GLNPO scientist with "board
chemistry" analyses in wet chem lab
• Understanding and implementing all
QA/QC requirements in this QAPP
• None
• Supervised by Deck Officer
• Receive site-specific
instruction from Chief
Scientist/Shift Supervisor
Abie/Ordinary
Seaman
• Operate ship's mechanical sampling
equipment, such as winches
• None
• Supervised by Deck Officer
• Receive site-specific
instruction from Chief
Scientist/Shift Supervisor
March 2003
Page 13
-------
Organization
Role
Responsibility
Authority
Lines of Communication
Grantee
Biologist(s)
• All on board sampling, sample handling,
labeling, and sample preservation
activities related to zooplankton tows
and benthos grabs in accordance with
applicable SOPs
• All on board sample handling, labeling,
and preservation activities related to
phytoplankton and chlorophyll a in
accordance with applicable SOPs
• Assistance with Rosette sampling when
requested by GLNPO staff
• Understanding and implementing all
QA/QC requirements described in this
QAPP
• Informing Chief Scientist/Shift
Supervisor of technical or QA issues
• Informing technical staff of changes in
required procedures
• Senior
biologist has
authority over
second shift
biologist
• Receives site-specific
instruction from Chief
Scientist/Shift Supervisor
• Communicates through the
Biology Technical Lead and
Grants Officer
GLAS
Contractor
Chemist(s)
• Assisting with Rosette sampling when
requested by GLNPO staff
• Adhering to all applicable SOPs to
conduct all on board sample handling,
preservation, and filtration activities
related to nutrient and other chemical or
physical analyses performed in a land-
based lab by the GLAS contractor
• Understanding and implementing all
QA/QC requirements described in this
QAPP.
• Informing Chief Scientist/Shift
Supervisor of technical or QA issues
• Informing technical staff of changes in
required procedures
• Senior
chemist has
authority over
second shift
chemist
• Receives site-specific
instruction from Chief
Scientist/Shift Supervisor
• Communicates through the
Limnology Technical Lead
and Project Officer.
Public Health
Service/
Region 5
Chemical
Hygiene Officer
• Responsible for all aspects of chemical
and chemical waste management
• Calls Safety Meeting after ship sails
• Manages onboard storage and control of
chemicals
• Disposes of laboratory chemical wastes
• Verifies all onboard staff have turned in
sealed medical questionnaires
• Verifies all onboard staff have taken the
laboratory safety course
• Serves as back-up to Captain in
ensuring all participants try on survival
suits
• Acts as contact person to locate
reagents onboard when survey staff
need them.
• Require
changes in
chemical
handling,
management
or disposal
operations
• Require
changes in
food
management
and handling
operations
• Reports to Jim Finn of
Region 5 through Public
Health Service
• Takes survey-specific
direction from Ship
Operations Manager and
Chief Scientist/Shift
Supervisor
Page 14
March 2003
-------
4.0 Data Quality Objectives
4.1 Project Quality Objectives
As was noted in Section 1 of this QAPP, the Great Lakes Water Quality Surveys are intended to fulfill
EPA's surveillance and monitoring obligations under Annex 11 of the Great Lakes Water Quality
Agreement. In establishing this surveillance and monitoring program, the U.S. and Canadian
governments stated their primary objectives in pursing these activities were to evaluate the effectiveness
of historical pollution control/pollution reduction strategies in the Great Lakes, identify emerging
problems, and identify the need for new or revised strategies and further research. Given these objectives,
GLNPO has designed the Water Quality Surveys to:
• Focus on key physical, chemical, and biological indicators of lake health
• Evaluate the health of each lake under different conditions (stratified and unstratified)
• Allow for real-time detection of significant changes in water quality, as indicated by significant
changes in one or more parameters
• Provide data that can be compared from year to year
• Yield data that are of sufficient quality to support decisions regarding the need for further study or new
pollution control strategies
For the purpose of the surveys, GLNPO considers a significant change to be a 20% change from historical
measurements made for a particular parameter in a particular lake during a particular season (e.g., a 22%
increase of total phosphorus concentrations in Lake Huron over historical summer values). GLNPO also
considers a reasonable potential for detecting such changes as an 80% likelihood that the change will be
detected. Therefore, the data quality objective (DQO) for the Surveys is to "collect measurements that
will yield an 80% chance of detecting a change of 20% or more within a particular lake and season, at the
90% confidence level". The key to meeting this objective is to minimize uncertainty. Major sources of
uncertainty during the surveys include:
(1) Sampling from a selected set of representative sampling stations instead of the entire lake
(2) Variability in the sample collection process, and
(3) Variability in the sample analysis process.
The issue of sample representativeness is addressed by the survey design, which is detailed in Section 7
of this QAPP. Briefly, the Survey design focuses on collecting samples from the same set of sampling
stations each year. These stations are primarily located in 1) open lake waters (at least 13 kilometers from
shore at least 30 meter depth) for water column sampling because historical data has shown that spatial
variation in open water is small compared to variation nearshore and 2) shallower, nearshore waters for
sediment sampling because historical data has shown that sediment changes in the open lake occur too
slowly for detection on a year to year basis. Individual station locations for the survey were determined
by EPA statisticians, working in conjunction with GLNPO scientists, who statistically evaluated historical
data collected from intensive surveys of each lake to identify the optimum number of samples and sample
locations needed to yield a representative sample of each lake basin. Figure 4-1 indicates the station
locations selected in each lake.
Variability in the sample collection process will be controlled through the use of trained sampling teams,
standardized, sample collection protocols, collection of field QC samples, and direct oversight of all
activities by GLNPO's Chief Scientists. Details of these quality assurance techniques are given in
Sections 5, 8, and 11 of this QAPP.
March 2003
Page 15
-------
Figure 4-1. Sampling Stations for the Great Lakes Water Quality Monitoring Surveys
Similarly, variability in the sample analysis process will be controlled through the use of trained
technicians and analysts, the use of detailed standard operating procedures (SOPs) for all field and
laboratory activities, and the use of standard QC measures and pre-defined QC acceptance criteria with
each set of analyses. Details of these QA techniques are given in Sections 5, 10, and 11 of this QAPP.
4.2 Measurement Quality Objectives
The physical, chemical, and biological parameters selected for inclusion in the study are listed in Table 2-
1. All of the nutrient and biological parameters are considered to be key indicators of lake health. Most
of the physical parameters are collected as supplementary data that are useful in interpreting the nutrient
and biological survey results. As a result, measurement quality objectives (MQOs) for the physical
parameters are not as stringent as those for the nutrient and biological measurements; the QC
requirements described in this QAPP reflect this philosophy.
The following subsections and Section 11 of this QAPP provide details on how EPA's standard data
quality indicators (precision, bias, accuracy, sensitivity, comparability, and completeness) will be
monitored and controlled in this study. Representativeness is addressed above in Section 4.1.
4.2.1 Precision
Precision is the degree of agreement among replicate measurements of the same property, under
prescribed similar conditions [2], It can be expressed either as a range, a standard deviation, or as a
percentage of the mean of the measurements (e.g., relative range or relative standard deviation). GLNPO
Page 16
March 2003
-------
has established several approaches to measuring and evaluating precision during the surveys. The
approaches vary according to monitoring parameter, expected variability, and importance of the measure
to survey objectives.
Ideally, precision is measured by collecting two environmental samples at the same time (in the case of
two sample collection devices) or sequentially (one device) and in the same place under the identical
circumstances (same personnel, procedures, etc). Each sample is preserved and numbered separately and
sent to laboratory as separate samples (coded FDn in the GLENDA database). Analysis of such
duplicates allows data users to evaluate the precision of the entire data collection effort, including
sampling, sample transport, and sample analysis. For the purposes of the Water Quality Surveys, the dual
and sequential approaches are considered to be equivalent because historical experience indicates that
minor time or spatial differences are not discernable in open lake measurements. In the Water Quality
Surveys, field duplicates will be required for all total and dissolved grab parameters collected with the
Rosette sampling device. These include:
N02 + N03 • Reactive silica • pH
Total P • Alkalinity • Chlorophyll a
TDP • Turbidity • Secchi Disk
Chloride • Specific Conductivity
In some cases, it is not feasible to collect true field duplicates or replicates; instead two or more
subsamples can be split from the original sample by sampling personnel in the field. Each subsample is
preserved and numbered separately, and the aliquots are sent to the laboratory as different samples.
Although such splits do not contain the full component of measurement uncertainty (they contain
uncertainty associated with field handling, laboratory preparation, and laboratory analysis procedures but
do not include sample collection uncertainty), they are considered acceptable tools for measuring overall
precision of those parameters in the Water Quality Surveys that are constrained by sample collection
devices (such as limited bottles on the Rosette). These parameters include the particulate parameters
which would require that a duplicate sample aliquot be processed through a separate field filter.
Specifically, these parameters are:
• POC • Particulate P
• Particulate N • TSS
Samples processed in this way during the Water Quality Surveys are identified as LDn in the GLENDA
Database.
Note: Field duplicates are collected on grab samples only (no field duplicates or lab duplicates are
collected on the integrated composite samples analyzed for phytoplankton and particulate nutrients).
Due to population variability among the zooplankton and benthic organisms, GLNPO will collect three
zooplankton tows at each master station and three ponar grab samples at all stations for benthic
organisms. Results for these three replicate field samples (i.e., numeric results such as biovolume) will be
averaged and the standard deviation reported. Currently, no MQO has been established for the relative
standard deviations among these field replicate measurements; GLNPO will establish one when sufficient
historical data becomes available.
In other cases, the cost and value of collecting precision measures is not justified by the survey objectives.
These cases include all physical measures that are collected to provide supplementary information that is
useful in interpreting survey results but not critical to assessing lake health.
March 2003
Page 17
-------
Measures of bias are collected for several of these parameters as shows in Table 11-4. Measures that fall
in this category include:
• Aesthetics
• Optical transmittance
• Air and water temperature
• Water clarity
• Wave height
• Wave direction
• Wind speed and wind direction
For those parameters in which precision measures are required, GLNPO is compiling and evaluating
historical data generated from the Great Lakes. GLNPO will use these data to generate statistically
derived MQOs that reflect precision levels that realistically can be achieved with the sampling and
analysis methods specified in this QAPP. In the meantime, GLNPO's interim MQO for survey precision
is that chemistry and physical results from 90% of these duplicate or split pairs agree within +/- 50% for
values greater than 5x the method detection limit and that 90% of these duplicate or split pairs agree
within +/- 100% for values less than 5x the method detection limit. No interim MQOs have been
established for the biological parameters.
In addition to the use of field duplicates and field splits to assess overall measurement precision,
laboratory staff will employ several types of laboratory QC measures (e.g., laboratory reference samples,
laboratory performance check samples, laboratory duplicates, etc.) that provide information about the
precision associated with various components of the analysis process. These QC elements and associated
requirements are described in greater detail in Section 11 (Quality Control Requirements) of this QAPP.
It should be noted that survey DQOs are based on overall data quality, and failure to meet any single
laboratory precision measure does not automatically imply the data are unacceptable for use in this study.
Instead, laboratory QC measures are used to monitor and control precision in real time so that overall
precision goals are met. Details regarding the data quality assessment process governing use of survey
data are given in Sections 16, 17, and 19.
Bias is the systematic distortion of a measurement process that causes errors in one direction [2]. Bias in
the Water Quality Surveys is measured through analysis of analytical standards of known concentration,
comparison among multiple instruments, and the participation in intercomparison studies when resources
allow. Control of bias also is assured through regular factory calibration of instrumentation. Bias
objectives and goals are currently limited to analytical chemistry measurements because 1) no source of
standard reference materials or intercomparison study samples exists to allow for determination of
biological measurement bias, 2) the cost of collecting bias measures on the physical measurements is not
justified by survey objectives. When intercomparison studies are conducted, GLNPO's MQO for bias is
that GLNPO results fall within two standard deviations of the mean study results.
At a less formal level, GLNPO's Technical Leads routinely make scientific assessments of data quality as
part of their routine data evaluation and analysis activities. Because lake chemistry changes slowly and
the Technical Leads are experts in recognizing measurement anomalies within each lake, their
interpretation of potential measurement bias can be quite valuable and will be used as a QA tool within
the Water Quality Surveys.
In addition to the use of intercomparison studies and best professional judgement to assess overall
measurement bias, laboratory staff will employ several types of laboratory QC measures (e.g., instrument
calibration standards, method blanks, etc.) that provide information about the bias associated with various
components of the analysis process. These QC elements and associated requirements are described in
greater detail in Section 11 (Quality Control Requirements) of this QAPP. It should be noted that survey
DQOs are based on overall data quality, and failure to meet any single laboratory bias measure does not
4.2.2 Bias
Page 18
March 2003
-------
automatically imply the data are unacceptable for use in this study. Instead, laboratory QC measures are
used to monitor and control bias in real time so that overall bias goals are met. Details regarding the data
quality assessment process governing use of Survey data are given in Sections 16, 17, and 19.
4.2.3 Accuracy
Accuracy is a measure of the closeness of an individual measurement or the average of a number
measurements to the true value. Accuracy includes a combination of random error (precision) and
systematic error (bias) components that result from sampling and analysis operations. Accuracy is best
determined by routinely analyzing a reference material of known concentrations or by reanalyzing a
sample spiked with a known amount of pollutant [2]. Because accuracy is a function of precision and
bias, GLNPO will control survey accuracy through the precision and bias strategies described in sections
4.2.1 and 4.2.2.
4.2.4 Sensitivity
Measurement sensitivity is defined as the minimum concentration of an analyte above which a data user
can be reasonably confident that the analyte was reliably detected and quantified. Above the detection
limit is the level at which reliable quantitative measurements can be made. This level is generically
termed the "quantification limit" or "quantification level" and it usually represents a multiplier (typically
between 3 and 5) of added certainty above the detection level.
Measurement sensitivity includes many components of variability, including variability of the
measurement device (instrument), variability of the sample preparation process, and variability introduced
during sampling. As a result, a myriad of detection limit concepts exist, including instrument detection
limits (IDLs), method detection limits (MDLs), and system detection limits or SDLs. Although the
details and emphasis of the various sensitivity concepts vary, nearly all definitions of detection limits are
based on statistical analyses of laboratory data. To determine an MDL using EPA's 40 CFRpart 136 B
approach, for example, at least seven replicate samples with a concentration of the pollutant of interest
near the estimated MDL are analyzed. The standard deviation among the analyses is determined and
multiplied by 3.14. The result of this calculation becomes the MDL. The factor of 3.14 is based on at-
test with six degrees of freedom and provides a 99 percent confidence that the analyte can be detected at
this concentration. Similarly, a common method for determining the IDL is to inject several blank
replicates or low concentration standard replicates into an instrument and calculate a limit based on the
standard deviation of the responses. System detection limits are often determined by determining the
standard deviation of replicate field blank measurements.
Because the data collected in the Water Quality Surveys will be used to support decisions regarding areas
of further study and will not be used for regulatory compliance or enforcement purposes, GLNPO's
objectives for determining and monitoring measurement sensitivity are based on: 1) the need to use
methods capable of detecting changes in water quality and 2) using the most pragmatic approach to
collecting this information. To ensure that methods used in the study are capable of detecting change, the
methods must be sensitive enough to detect the target parameters at the concentrations currently found in
the lakes. Where feasible, method sensitivity should be at least 3-5 times lower than the concentrations
historically found in the lakes so that results reported in the surveys can considered to be both
quantitatively and qualitatively accurate. For parameters that are likely to be influenced by background
contamination, method sensitivity should be at least lOx lower than historical lake concentrations so that
actual results reported can be reliably distinguished from any effects of contamination. Table 4-1 lists
sensitivity goals for each chemistry parameter. These goals reflect historical data collected in the lakes.
March 2003
Page 19
-------
Presently, only one parameter (total dissolved phosphorous) is barely meeting the sensitivity objectives
listed in Table 4-1. Although solutions, such as the use of alternate analysis methods, do exist, the costs
of such solutions may outweigh the benefits. Therefore, GLNPO is currently evaluating these costs and
benefits and will determine if changes to the measurement procedures are needed prior to future surveys.
For each chemical method, Method detection limit (MDL) studies should be performed according to 40
CFR, part 136, Appendix B (if applicable), or by another procedure that is pre-approved by the WQS QM
Technical Lead or the GLNPO QM. MDL studies must be conducted once per year and each time a
significant change is made to the analytical SOP, including changes in instrumentation.
As noted above, GLNPO's objective for measuring sensitivity is to use the most pragmatic means
available given study objectives and already known contributions to measurement uncertainty. Ideally,
for example, sensitivity is defined through the use of a system detection limit concept, because this
concept takes into account the sensitivity of the entire measurement process (field, laboratory, and
instrument). In practice, however, it is difficult and costly to collect and analyze replicate field blank
samples. In addition, the effects of field procedures on measurement sensitivity are usually small
compared to the effects of analysis method and instrument variability. The exception to this rule is for
ubiquitous pollutants, such as metals, that can contaminate samples at extremely low concentrations in the
field. None of the parameters monitored in the surveys fall into this category; therefore, determination of
system detection limits is not necessary for the survey, although they are determined for some analytes.
Similarly, the meteorological data collected from the Danforth Marine Indicator are ancillary to the
primary survey objectives. Therefore, a formal determination and verification of instrument or method
sensitivity by Survey staff is not necessary; instead, GLNPO relies on the biannual calibration and
certification of the instrument by its manufacturer.
Page 20
March 2003
-------
Table 4-1
Sensitivity Objectives for Water Quality Survey Parameters
Parameter
Sensitivity Objective
no2/no3
0.03 mg N/L
Particulate Nitrogen
.03 mg N/L
Particulate Organic Carbon
5.0 |jg C/L
Dissolved Organic Carbon
not specified
Total Phosphorus
1 ug/L
Total Dissolved Phosphorus
1 ug/L
Particulate Phosphorus
1 ug/L
Chloride
0.03 mg Cl/L
Reactive Silica
0.010 mg Si/L
Specific Conductance
not specified
Alkalinity
not specified
PH
not applicable
Total Suspended Solids
not specified
Dissolved Oxygen
0.1 mg/L
4.2.5 Representativeness
Representativeness is a measure of the degree to which data accurately and precisely represent a
characteristic of a population parameter at a sampling point or for an environmental condition. It is a
qualitative term that is evaluated to determine if data appropriately reflect the media and phenomenon
being measured or studied. [2] As indicated in Section 4.1, the sampling stations selected for the Water
Quality Surveys were chosen so that the target populations are represented by valid sub-samples. As
resources allow, GLNPO will periodically assess results of its survey activities over time to verify that the
statistical power originally anticipated for the study is still being met.
4.2.6 Completeness
Ideally, GLNPO and contractor/grantee scientists will be able to collect a complete suite of data from all
stations in all surveys. In practice, however, a lower level of completeness can still lead to a valid study.
If a significant lack of completeness is obtained in a particular survey (cruise), causes of the problem(s)
will be investigated and lessons learned will be incorporated into planning of future surveys. When
samples are not collected at a station, the reason will be documented by the Chief Scientist. The remarks
column on the data forms (located in Appendix H) can be used for this purpose.
4.2.7 Comparability
Comparability expresses the confidence that two data sets can contribute to a common analysis and
interpolation. Because the surveys involve the collection and analysis of numerous samples during
different seasons and years, data will be generated from samples collected by different sampling teams
and analyzed by multiple laboratory staff and organizations. To ensure comparability among these data
sets, GLNPO and its team of contractors and grantees will:
• Employ standard operating procedures for sample collection (or require that their contractors and
grantees employ such SOPs)
• Employ or require use of detailed analytical methods/SOPs specifying each step of the laboratory
process
March 2003
Page 21
-------
• Use one method/SOP for all analyses of a given pollutant, unless significant improvements to
measurement techniques are identified. In such cases, the Technical Lead and QAM will evaluate the
potential impacts of method changes before approving their use.
• Offer an annual training session on the SOPs used onboard the vessel
• Ensure that the Chief Scientist oversees all onboard sampling and analytical activities
• Specify method detection limits and QC acceptance criteria that must be met throughout the surveys
• Specify data reporting formats and units that must be used by all survey participants
• Use a standardized data quality assessment process
5.0 Special Training Requirements
All EPA, contractor, and grantee staff participating in the surveys must be fully trained in onboard safety
requirements and any scientific functions for which they are responsible. Survey scientists involved in
sample or data collection also must be trained to use the equipment and procedures specified in this
QAPP. In addition to training received prior to boarding the ship, all new staff responsible for performing
scientific activities onboard ship are provided with hands on training and supervision by senior scientists
during the first few days of each survey. Further, the Chief Scientist conducts a science meeting at the
start of a survey and with each new crew change. Logistics, roles and responsibilities, and special
objectives of the survey are presented and any questions addressed.
Training responsibilities vary according to specialty. GLNPO is responsible for providing all safety
training and for training survey scientists in the use of the Rosette sampling device, the collection of
onboard measurements using the SeaBird, and the use of specific instruments and techniques for
measurement of other physical parameters. The biology grantee is responsible for training its staff in the
collection and analysis of biological parameters of interest. Similarly, the GLAS contractor is responsible
for training its staff in the collection and analysis of nutrient parameters of interest. In addition, the
GLAS contract and the grant agreement specify minimum skills (i.e., education and experience) required
of the analysts performing work under these vehicles.
Specific training requirements will include some or all of the following courses, depending on the specific
responsibilities of the survey participant(s).
• 24-hour Laboratory Safety Course/4-hour Laboratory Safety Refresher Course (required for all
government, contractor, and grantee survey scientists working in the ship laboratory)
• First Aid, CPR, EMT training for all designated 'first responders' (Chief Scientist, Shift Supervisor,
First Mate, and Second Mate)
• Fire Fighting (required for the contractor-provided ship operating personnel)
• Powered Industrial Trucks/fork lifts (required for the contractor provided ship operating personnel)
• GLNPO Chemical Hygiene Plan
• Safety Orientation Video (required for every individual that participants in cruises, regardless of
responsibility)
• Boat Handling and Seamanship (required of the contractor-provided ship operating personnel)
Additional information concerning safety training procedures and requirements can be found in GLNPO's
Health, Safety, and Environmental Compliance Manual (May 1997 or as amended).
All staff involved in the Water Quality Surveys are responsible for ensuring that they have received the
required training and that their training is up-to-date. Each individual participating in the survey also is
responsible for completing and submitting to the QA Manager a Self-Certification form (Appendix F of
the WQS manual) that certifies s/he has completed all required training. The QA Manager is responsible
Page 22
March 2003
-------
for ensuring that all survey participants have completed and submitted this form prior to the start of the
cruise.
6.0 Documentation and Records
Formal chain of custody procedures are not required for the Water Quality Surveys because survey data
are not used for enforcement or compliance monitoring activities. Instead, GLNPO employs several QA
strategies to ensure that samples are accurately documented and tracked throughout the sample collection,
handling, transport, analysis, and reporting processes. These include the use of centralized sample
identification numbers that are pre-assigned by GLNPO before each survey and strict Record keeping
requirements employed by all staff throughout the surveys. These requirements are described in Sections
6.1 and 6.2 respectively.
In addition, a series of forms are used to document specific processes employed during implementation of
the surveys. Appendix C contains a staff scheduling form that lists shift dates for proposed and actual
GLNPO staff participating in the survey. Appendix E contains survey planning forms that must be
submitted by researchers requesting use of the Lake Guardian for sampling purposes. Appendix F
contains the self-certification form that all GLNPO staff and contractors/grantees participating in the
survey must complete and provide to the GLNPO QA Manager to certify their meeting pre-survey
training requirements as specified on the form. Appendix G contains the Medical History Questionnaire
that is required to be completed and submitted by all personnel that participate in GLNPO's WQS.
Appendix L contains a list of roles and responsibilities for the Chief Scientist, in addition to a list of
priorities for each staffing period and a checklist that must completed by the shift supervisor for each shift
of 12 hours and provided to the QA Manager to verify that they met their responsibilities involving ship
operations and survey data management. Appendix M contains a suggested revision sheet that survey
participants should use to document issues, errors, and suggested revisions and clarifications to the
manual. Appendix N contains a data status tracking sheet to track specific data sets throughout the data
verification process and upload to the GLENDA database. Appendix O contains a data discrepancy form
to initiate a revision to the GLENDA database when errors are identified. Finally, Appendix Q contains
the station location change form that must be completed and approved by pertinent survey participants to
finalize a change to a station location.
6.1 Sample Identification and Labeling
Sample definition and numbering: Prior to each survey, the Technical Lead for Board Chemistry will
create and assign unique sample numbers for each sample to be collected during the survey. This
approach ensures that there is no confusion among scientists and staff during data gathering, data
reporting, or data evaluation concerning sample identity. For the purposes of the Water Quality Surveys,
a sample is defined as a discrete volume of water or sediment collected with a particular type of device at
a particular site and depth on a given day. A sample may then be aliquotted into several fractions and
bottles for analysis of different parameters. Each of these bottles is given the same sample number. Field
duplicates and blanks are treated as separate samples and, therefore, assigned different sample numbers.
Pre-labeled bottles and containers: Prior to each survey, the Board Chemistry Technical Lead creates
computer-generated sample labels that are pre-affixed to each cubitainer and bottle that will be used to
store samples in the field. Copies of these labels also will be affixed to the cubitainer or bottle cap to
serve as a back up means of identification should the labels come off during sample processing. To
further ensure that samples are properly identified and handled in the field, each sample label will be
color coded to indicate the preservation and filtration requirements needed for each sample bottle (i.e.,
yellow labels are used for total nutrient aliquots requiring preservation with sulfuric acid, orange labels
March 2003
Page 23
-------
are used for total dissolved nutrients which are filtered and preserved with sulfuric acid, and white labels
are used for unpreserved sample aliquots).
Prior to arrival at a sampling station, the labels for the associated samples will be segregated and applied
to the sampling bottles. When sample bottling and preservation are completed, a record of the numbers
on the labels used will be made on analysis request sheets. The analysis request sheet will be used to
track samples through the processing and analysis.
Label information: Each sample label will contain the following information:
• A 9-digit sample number (coded as described • Survey data
below) that is unique to the sampling date, • Preservation used
location, depth and sampling device • Parameter to be measured
• Lake
• Station number
Sample numbering scheme: Each sample number is coded in the following format to provide summary
level information about the sample:
Year Sample Lake Series Sample Sample
Device Type Number
Valid codes for the sample numbers are as follows.
Year: two digit years such as "00", "01", "02"
Division: G for GLNPO
Lake: A = Michigan; B = Huron; C = Erie; D = Connecting Channels; E = Ontario;
S = Superior
Series: Number sequence
Sample Type: S = Primary; I = Integrated; D = Duplicate; R = Field Blank; C = Duplicated
Analysis; X = Spike; and B = Laboratory Blank.
In addition, samples that are collected for production of the integrated sample only (i.e., they are not
otherwise collected for determination of study parameters but solely for production of the integrated
sample), will be identified with an a, b, or c at the end of the sample number.
Station Identification: To avoid confusion regarding station identification between lakes (i.e., to
distinguish between Lake Erie Station 61 and Lake Huron Station 61), GLNPO assigns a "Station ID
Number" to each station visited. The Station ID number consists of the first two letters of the lake name
followed by a space and the 2-digit station number (a leading zero is added to single digit station IDs).
For example, Lake Erie Station 61 is assigned the Station ID "ER 61" and Lake Huron Station 61 is "HU
61". Historically, a different numbering system was used for Lake Superior, but to improve consistency,
sampling stations for Lake Superior are now numbered SU 01 - SU 19.
Visit Identification Numbers: GLNPO also assigns unique "Visit ID" numbers to differentiate sampling
activities conducted at a single site on different dates (or visits). The Visit ID number consists of the first
letter of the lake name followed by the 3-digit station ID (leading zeros are added for one or two digit
station IDs) followed by the first letter of the month the survey began and the last two digits of the survey
Page 24
March 2003
-------
year. In order to give each month a unique identifier, Y is for May, U is for June, L is for July and G is
for August. For example, if Lake Erie Station 61 is visited in the spring of 2000, and the spring surveys
begin in April, the Visit ID assigned to this station on the spring cruise is "E061A00".
Time definitions: While surveys are in progress, a single 'ship time' is used throughout the survey
regardless of the time zones crossed during transit. Therefore, all survey participants are required to
synchronize their watches with 'ship time' when they board ship. Because the world standard for time
keeping is Greenwich Mean Time (GMT), all electronic measurements made on the bridge are captured in
GMT. To facilitate conversion between GMT and ship time, the "mate of the watch" also records the
'correction factor' (e.g., 5 hours) for converting GMT to the time zone standard (e.g., eastern daylight
time, central standard time, etc.) used at the ship location. The correction factor is entered into GLENDA
along with the actual time measurements recorded by scientists and bridge staff when the data are
uploaded to the database. Once in GLENDA, data users can specify the time standard they need when
retrieving data.
6.2 Record keeping
The Chief Scientist has primary responsibility for assuring that all data gathered in the survey is
documented through a combination of manual and electronic procedures as described below.
Documentation includes raw instrument level printouts, summary bench sheets, hard-copy field
information recording forms, and electronic records generated on board ship and in the laboratories.
All shipboard-generated strip charts, bench records, and computer printouts are kept in a folder, indexed
by station, until the remaining samples are transferred to the land-based laboratory. All raw data are
assembled and indexed by parameter, by lake, and by survey leg. Analog charts and digital conversion
printouts will be stapled together. Each parameter will be placed in manilla folder and transferred to the
GLNPO Chief Scientist for Board Chemistry upon conclusion of the survey. Several parameters are
recorded on the ship's log, such as vessel position and weather conditions. This information is described
in LG 300, SOP for Meteorological Data Aboard the R/VLake Guardian. These parameters are then
recorded on the GLENDA Station Information Field Recording Form at each station visit and entered into
the GLENDA database.
Hard-copy Field Information Recording Forms (located in Appendix H of the WQS manual) are used to
record data as it is being collected during the survey. GLNPO has developed 16 hard-copy data forms for
capturing all required data for the survey as well as supplementary data, such as weather. These 16 forms
are:
1)
Survey Information
2)
Station Information
3)
Rosette Sampling Data
4)
Ponar Grab Sampling Data
5)
Zooplankton Net Flowmeter Calibration
6)
Zooplankton Sampling and Secchi Disk Data
7)
Chlorophyll a Preparation
8)
Phytoplankton Preservation
9)
Nutrients Preparation
10)
POC, PN, PP Preparation
11)
TSS Preparation
12)
Preparation of Quality Assurance Samples
13)
Calibration Data of Board Chemistry Instruments plus Shiftwise Standardization
14)
Control Standards Data of Board Chemistry Parameters
15)
Board Chemistry Data
March 2003
Page 25
-------
16) Dissolved Oxygen Data (Winkler)
The data reporting forms listed above are intended to capture information about discrete measurements.
In addition, GLNPO collects extensive electronic data that is gathered continuously on the bridge at all
times while at sea and continuously by the SeaBird when the probe has been dropped. Bridge data
concerning temperature, barometric pressure, ship heading, depth, apparent wind angle and speed, GPS
position, Greenwich Mean Time, Greenwich Mean Day, course over ground, waypoint position (station
location), and bearing and range to waypoint are averaged every 15 minutes and stored on disk. One disk
per lake per survey is used to store all such electronic data from the bridge. SeaBird measurements are
averaged every half meter and stored using the SEASAVE software program which facilitate conversion
of the data to graphic or tabular format.
During the surveys, data are entered into an onboard data entry system, the GLENDA remote data entry
tool, designed to capture survey data into the database on a daily basis. A User guide for the tool is
located in Appendix J of the WQS manual. This system, available on the R/VLake Guardian and at
GLNPO headquarters, is designed to include real-time data entry checks to prevent analysts or technical
staff from entering 'nonsensical' values. The tool is based on a windows interface to the GLENDA
database and is intended to allow the easy input of field data from prescribed field information recording
forms mentioned above. The items from the recording forms are copied to corresponding locations on the
windows input sheets. If the tool is not functioning properly, the data can be entered into Excel electronic
files, which are a soft-copy version of the recording forms in Appendix H of the manual, as a back-up to
the tool. If this back-up option is used, the data are entered into the Remote Data Entry Tool from these
electronic files at GLNPO headquarters or onboard the ship if the Remote Data Entry Tool begins to
function during the survey leg. Once the data are in the Remote Data Entry Tool they are uploaded to
GLENDA as unverified data. A Data Flow Diagram is provided in Figure 1, of SOP LG101 of the WQS
manual, and presents the data management process for data collection and data verification for the WQS.
The Chief Scientist is responsible for transferring the original field information recording forms to the QA
Team at GLNPO (Marvin Palmer or his designee) for internal quality control checks. The QA Team
conducts checks of the recording forms against the data that has been uploaded to GLENDA and
addresses any data discrepancies. After completing these checks and resolving all discrepancies, the QA
Team transfers the forms to the Environmental Monitoring and Indicators Team for storage in the
designated file cabinet at GLNPO. The pertinent Technical Lead conducts a final review of the data and
notifies the Database Manager of approval. The Database Manager then finalizes the dataset as Version 1
and makes the data available. For each dataset, the status of this data management process is tracked by
the QA Team on the Data Status Tracking Sheet (Appendix N of the WQS manual).
The Chief Scientist must assure that a copy of the hard-copy field information recording forms is made
and placed in the onboard designated file cabinet at the end of each survey leg. The Chief Scientist
assures that an electronic copy of all soft-copy field information recording forms is made and transferred
to the database manager at GLNPO (Ken Klewin) for storage. The Chief Scientist also must assure that
SeaBird data files are processed at the end of each survey leg and returned to the database manager for
storage. These items are included on the list of Chief Scientist Roles and Responsibilities (Appendix L of
the manual). For the bridge data, The Officer in Charge shall be responsible for reviewing the previous
watch entries in the Ships's Log, GLENDA Station Information Field Recording Form and the GLENDA
database to assure correctness of the data entered. Errors noted shall be corrected immediately.
A process also has been developed for correcting errors identified in verified data in GLENDA and is
illustrated in Figure 2 of SOP LG101 in the WQS manual. Once a data error is identified, a Data
Discrepancy Form (Appendix O of the WQS manual) is completed and submitted to the Quality
Assurance Manager and Database Manager. Data are revised in GLENDA and the revisions are verified
as correct by the QA Team. The pertinent Technical Lead also conducts a final review and notifies the
Page 26
March 2003
-------
Database Manager of approval. The Database Manager then finalizes the dataset as the subsequent
version and makes the data available.
Results generated in land-based laboratories will be reported on laboratory-specific data forms (i.e., the
laboratories are allowed to design and submit their own data reporting format that complies with
GLNPO's data entry standard and is subject to GLNPO approval).
Additional information concerning Record keeping procedures can be found in the Water Quality Survey
SOP LG101, Electronic Field Information Recording.
7.0 Sampling Process Design
The sampling design reflects the need to collect data from sites, depths, and times that are representative
of lake conditions. Section 7.1 summarizes the design strategy for selecting representative stations,
Section 7.2 summarizes the design strategy for selecting representative sampling depths at each station,
and Section 7.3 summarizes the strategy for selecting representative sampling frequencies at each station
and depth.
7.1 Site Selection Strategy
All sampling stations selected in the survey were chosen by scientists and statisticians using 1) historical
data from intensive lake studies (which indicates that spatial variation in open lake waters (more than 13
kilometers from shore and greater than 30 meter depths) is small compared to nearshore spatial variation,
and 2) the Water Quality Survey objective of detecting a 20% change from year to year.
Using this approach, stations for water column sampling were selected from homogenous areas of each
lake where routine variability in lake conditions would not likely mask actual changes. In contrast,
locations of benthos stations were selected to provide coverage of both the offshore benthic communities
at a variety of depths, and selected nearshore sites of specific interest. [3, 4, 5, 6] Over time, actual
station locations have been modified to reflect new data, information, and priorities. The water column
and benthic monitoring stations currently used during the surveys are shown in Appendix A of the Water
Quality Survey procedures manual, titled, Sampling and Analytical Procedures for GLNPO's Open Lake
Water Quality Survey of the Great Lakes, March 2003 (or as amended). These stations also are described
in Attachment A of this document, Tables A-l through A-18.
In addition to the routine monitoring stations, "master stations" have been identified in each major lake
basin. The routine monitoring stations are referred to as non-master stations. Originally, two master
stations were selected for each basin, with one master station being sampled on the way out and the other
being sampled on the return trip. The intent was to provide a form of replication to ensure that the data
collected during each survey would not be biased by transient effects. Most of the current master stations
represent the deepest point within its lake basin. These stations are sampled at more frequent depths to
provide GLNPO with a better means of characterizing vertical conditions during each survey. Master
stations identified for each lake basin are shown in Table 7-1.
In the event that a station location needs to be changed (e.g., when the biology team searches for a
depositional zone and would like to designate the coordinates of the new station location) certain steps
should be followed to assure that the proposed change is approved, all participants are notified, and that
all documentation are updated. Further, these steps will ensure that all documents, publications, and
databases use the same finalized (current) station coordinates. The master list of station location
coordinates will be posted on GLNPO's website, as well as in GLNPO's shared "Base Monitoring
Program" directory.
March 2003
Page 27
-------
To change a station location, the form GLNPO's Base Monitoring Program Station Location Change
Form, located in Appendix Q, needs to be completed by filling in the following fields: station ID, old
coordinates, new coordinates, new depth, the reason for the station location change, and the person's name
who is requesting the change. The completed form should be submitted to the appropriate technical lead
(Glenn Warren for open water stations and Marc Tuchman for benthos stations). Once the technical lead
approves the change, he needs to sign and date the form.
The completed form needs to be initialed by GLNPO's Monitoring Team Leader to give final approval to
the station location change, and then initialed by the Captain verifying that the ship's GPS unit has been
updated. The form must then go to the GLENDA database manager who will initial it, file it, and update
the master station location list. Before a new version of the SOP manual is released, these changes must
be incorporated into the WQS QAPP and the Lake maps updated.
For consistency, new station location coordinates should be rounded to the nearest whole second of arc,
decimal degrees should be limited to five significant digits, and decimal minutes should be limited to
three significant digits.
able 7-1. Master Station Locations for the Water Quality Surveys
Lake
Basin
Master Stations
Michigan
southern lake
Station 18
central lake
Station 27
northern lake
Station 41
Huron
northern lake
Stations 45 and 54*
central lake
Not applicable
southern lake
Station 15
Erie
western lake
Station 91
central lake
Station 78
eastern lake
Station 15
Ontario
western lake
Station 33
eastern lake
Station 55
Superior
western lake
Station 17
central lake
Station 08
eastern lake
Station 01
* Two master stations in northern Lake Huron are an artifact of the historical
two station approach
7.2 Depth Selection Criteria
Sampling depths vary according to the condition of the lakes (stratified versus unstratified conditions) and
the parameter of interest (for example, benthic organisms are only found on the lake bottom while
phytoplankton congregate at the upper depths of the lake). Further, these strategies vary depending if
samples are being collected from a master or non-master station.
Page 28
March 2003
-------
In general, the spring survey strategy is to collect samples from the surface, mid-depth, near bottom and
bottom depths of each lake to gather information about the lake during unstratified, or isothermal,
conditions. In general, the summer survey strategy is to collect samples from the surface, mid-epilimnion,
lower epilimnion (master stations only), mid-hypolimnion (non-master stations only and Lake Erie
central basin master station), thermocline (master stations only), upper hypolimnion (master stations
only), deep chlorophyll layer (if present), near bottom, and bottom depths of each lake to gather
information necessary to characterize the lake during stratified conditions. During spring and summer
surveys for master stations, samples also are collected at discrete depths throughout the water column. In
addition, integrated composite samples are collected for determination of several parameters of interest.
These depths, and the parameters sampled at each depth are detailed in Table 7-2.
March 2003
Page 29
-------
Table 7-2. Sampling depths and parameters measured in each season
Depth Description
Season(s) Sampled
Parameters Monitored
Surface (SRF, 1 meter below
surface)
spring and summer
Mid-depth (varies according to
site)
spring (non-master stations only)
Mid-epilimnion (MEP, depths
vary)
summer
Lower epilimnion (LEP, depths
vary)
summer (master stations only, excluding
Lake Erie Central and Western basins)
Thermocline (TRM, depths vary)
summer (master stations only, excluding
Lake Erie Central and Western basins)
Upper hypolimnion (UHY, depths
vary)
summer (master stations only, excluding
Lake Erie Central and Western basins)
Nitrate + Nitrite, Total
Phosphorus, Total Dissolved
Phosphorus, Chloride, Reactive
Silica, Turbidity, Specific
Conductance, pH, Dissolved
Oxygen, and Chlorophyll a
Mid-hypolimnion (MHY, depths
vary)
summer (non-master stations only and
Lake Erie Central basin master station)
Discrete depths including 5 m, 10
m, 20 m, 30 m, 40 m, 50 m, 100
m and 200 m (as station depth
allows)
spring (master stations only) and
summer (master stations only, excluding
5 m, 10 m, 20 m and 30 m)
10 meters from the bottom (B-10)
spring (excluding Lake Erie Central and
Western basins) and summer (for Lake
Erie, only Eastern basin master station)
2 meters from the bottom (B-2) in
all lakes excluding Lake Erie
spring (at all master and non-master
stations only if inverse stratification is
present) and summer (at all master and
non-master stations only if nepheloid
layer is present)
1 meter from the bottom (B-1)
only in Lake Erie
spring and summer
Deep Chlorophyll Layer, if present
(DCL, depths vary)
summer
Phytoplankton, Chlorophyll a,
Total Suspended Solids,
Particulate Organic Carbon,
Particulate Total Nitrogen,
Particulate Total Phosphorus,
Chloride Silica Total
Integrated sample of the upper
lake (SRF = 1 m, 5 m, 10 m, and
20 meters)3
spring
Integrated sample of the upper
epilimnion (SRF = 1 m, 5 m, 10 m
and LEP)b
summer
Phosphorus, Total Dissolved
Phosphorus, and Nitrate + Nitrite
Bottom sediments
spring and summer
Benthos in spring
Benthos, Grain Size, Total
Phosphorus0, Total Organic
Carbon0 and Total Nitrogen0 in
summer
From 20 meter depth to surface
spring and summer
Zooplankton
From B-2 to surface or from 100
m to surface whichever is less
spring and summer
a For an unstratified water column, the integrated sample is prepared by taking equal volumes of water from SRF (1
m), 5 m, 10 m and 20 meters unless the depth is less than 20 meters. If the total depth is between 15 and 22 meters,
Page 30
March 2003
-------
the 20 meter sample is replaced by the bottom sample (B-1 or B-2). If the total depth is less than 15 meters, equal
volumes are taken from surface, mid-depth, and bottom sample (B-1 or B-2).
b For a stratified water column, equal volumes are taken from the surface, 5 m, 10 m, and lower epilimnion (LEP). If
the epilimnion is very shallow, equal volumes are taken from a maximum of four sampling depths and a minimum of
two sampling depths. The underlying strategy is to collect a representative sample from the epilimnion. See in
Appendix P, Integrated Sample for a Stratified Water Column.
c These parameters are not determined every year.
NOTE: A suite of physical parameters, such as specific conductance and temperature, also are monitored with the
SeaBird CTD (see section 8.1 and LG302, Operating the SeaBird CTD).
7.3 Sampling Sequence/Frequency Strategy
Sampling events during the cruises follow the general outline described below, with minor variations
made on-site as necessary to accommodate weather, sea, and other conditions.
Sample Type Sampling Strategy
Visual and • Record wind speed and direction, wave height and direction, air temperature,
Physical barometric pressure, visibility, weather conditions, and heading (when
Observations underway) each hour on the hour
Record station identification, arrival time, departure time, wind speed and
direction, wave height and direction, barometric pressure, water depth, air
temperature, geographic location upon arrival, and final location (if vessel
drifted during sampling) at each sampling station.
Record deviation of ship time from Greenwich mean time daily.
Rosette • Run Rosette/CTD down to define the temperature profile and determine the
samples thermal structure. The Chief Scientist or Shift Supervisor will examine the
CTD profile. The Chief Scientist or Shift Supervisor and the Marine
Technician will select sampling depths according to depth selection strategy
for each lake (See LG200, Section 4.0, Sample Depth Selection, for detailed
information on selecting station depths and Appendix B, Attachment A for a
detailed list of monitoring stations and depths). During unstratified conditions
that occur in the Spring, most of the sample depths are constant, however,
the CTD profile is used to determine whether inverse stratification is present
and sample accordingly. During stratified conditions that occur in the
Summer, the CTD profile is used to determine the depths of discrete positions
in the thermal structure of the lake such as the mid-epilimnion.
Trigger sample bottle at correct depths, while verifying the temperature profile
(See Appendix B, Attachment A for a detailed list of monitoring stations and
depths).
Split Rosette Niskin samples into the required sample bottles/preservatives.
An integrated sample3 is taken for phytoplankton, chlorophyll a, total
suspended solids, particulate organic carbon, particulate nitrogen, particulate
phosphorus, chloride, silica, total phosphorus, total dissolved phosphorus,
and nitrate-nitrite, by compositing Niskin samples at various depths.
3For an unstratified water column, the integrated sample is prepared by taking equal volumes of
water from SRF (1 m), 5 m, 10 m and 20 meters unless the depth is less than 20 meters. If the total depth
is between 15 and 22 meters, the 20 meter sample is replaced by the bottom sample (B-1 or B-2). If the
total depth is less than 15 meters, equal volumes are taken from surface, mid-depth, and bottom sample
(B-1 or B-2). For a stratified water column, equal volumes are taken from the surface, 5 m, 10 m, and
lower epilimnion (LEP). If the epilimnion is very shallow, equal volumes are taken from a maximum of four
sampling depths and a minimum of two sampling depths. The underlying strategy is to collect a
representative sample from the epilimnion.
March 2003
Page 31
-------
Sample Type
Sampling Strategy
Zooplankton
samples/
Secchi
readings
Benthos
Dissolved
Oxygen
Conduct the 20 meter and B-2/100m vertical tows
Rinse the net
Pour into bottles and transfer to onboard laboratory for preservation
After tows are completed, measure transparency using the Secchi disk.
Collect four ponar grab samples at each benthos stations during the summer
surveys and three Ponar grab samples at each station in the spring surveys.
During the summer survey, use one of the four grabs off for physical
measurements and the remaining three for benthos analysis. (Physical
measurements are not collected from benthos samples during the Spring
surveys.)
During summer surveys, when the lakes are stratified, collect samples for DO
determination at each master station in each lake. Record a full SeaBird
profile DO. In addition, the surface and B-2 sample is used for duplicate
analysis by Winkler titration. Simultaneously, analyze an oxygen-saturated
sample, either from a sampled same depth or from the reagent water system,
by Winkler titration.
Note: A separate dissolved oxygen survey is conducted in the Lake Erie Central Basin. See Appendix D,
Dissolved Oxygen and Temperature Profiles for the Central Basin ofLake Erie QAPP, for detailed
information regarding this survey.
Table 7-3 summarizes the types of samples (grab vs. composite), type of field QC, and parameters
sampled at each depth.
Page 32
March 2003
-------
Table 7-3. Sampling depth strategy for target parameters
Parameter Group
Sampling Depths
Sample Type
Total Suspended Solids,
Integrated Sample
Composite
Particulate Organic Carbon,
Deep Chlorophyll Layer
Grab
Particulate Nitrogen, Particulate
Phosphorus
Chloride, Silica, Total
Surface
Grab
Phosphorus, Total Dissolved
Mid-depth
Grab
Phosphorus, Nitrate + Nitrate
Mid-epilimnion
Grab
Lower epilimnion
Grab
Thermocline
Grab
Deep Chlorophyll Layer
Grab
Upper hypolimnion
Grab
Mid-hypolimnion
Grab
Discrete Depths*
Grabs
B-10
Grab
B-2, B-1
Grab
Integrated
Composite
Board Chemistry (Alkalinity, pH,
Surface
Grab
Turbidity and Conductivity)
Mid-depth
Grab
Mid-epilimnion
Grab
Lower epilimnion
Grab
Thermocline
Grab
Upper hypolimnion
Grab
Mid-hypolimnion
Grab
Discrete Depths*
Grabs
B-10
Grab
B-2, B-1
Grab
Winkler DO
Surface
Grab
B-1
Grab
Zooplankton
Tow 20 m to surface
Integrated
Tow B-2 to surface or from 100 m to
Integrated
surface, whichever is less
Benthos
Sediment
Grab
Phytoplankton
Integrated Sample
Composite
Deep Chlorophyll Layer
Grab
Chlorophyll a
Surface
Grab
Mid-depth
Grab
B-2, B-1
Grab
Integrated
Composite
* Discrete depths include 5 m, 10 m, 20 m, 30 m, 40 m, 50 m, 100 m and 200 m (as station
depth allows)
NOTE: A suite of physical parameters, such as specific conductance and temperature, also are
monitored with the SeaBird CTD (see section 8.1 and LG302, Operating the SeaBird CTD).
8.0 Sampling Method Requirements
The primary data quality objectives when sampling are to:
Collect representative samples at the locations described in Sections 7.1
March 2003
Page 33
-------
Collect representative samples at the depths described in Section 7.2
• Preclude contamination from the equipment and sample handling processes
Verify that the sampling techniques yield reproducible results by collecting field duplicate
samples from each basin and field blanks from approximately 25% of the stations sampled in
each survey.
Station Locations: GLNPO will use a digital global positioning system (GPS) to ensure that sampling is
conducted at the designated station locations. GPS is a satellite-based radio-navigation system developed
and operated by the U.S. Department of Defense (DOD). It permits land, sea, and airborne users to
determine their three-dimensional position, velocity, and time 24 hours a day, in all weather, anywhere in
the world and is accurate within 25 feet. The same digital GPS system used on the bridge is connected to
the SeaBird so that consistent readings are made at each location. When the R/VLake Guardian arrives
on station, the Captain will notify survey participants of the station name and ship time so that sample
collection can begin. Note: For surveys in Lake Erie, sampling locations should sometimes be revisited
due to windy conditions that can stir up the lake bottom. The Chief Scientist should review the results of
turbidity measurements for Lake Erie to determine if another sampling event should be conducted on the
return trip through the lake. If the turbidity is more than 3.5 NTU for any Lake Erie station, the results
may be artifacts of sediment resuspension and may not accurately reflect conditions of the water column.
In these cases, the Chief Scientist will, if possible, collect additional samples for the full suite of analyses
during the return trip through Lake Erie.
Sampling Depths: Samples are collected at all stations at a series of depths. The exact depths are
determined by the type of station (master and non-master), the season of the survey, the station depth, and
the thermal profile. A generalized list of the samples collected is provided in LG 200, Table 1.
Attachment A to this QAPP further details the sampling depths at the various stations. Prior to the
survey, labels and other paperwork are prepared designating the sampling depths for the different stations.
Once on station, depths will be determined using the CTD to define the temperature profile. During
unstratified conditions that occur in the Spring, most of the sample depths are constant, however, the CTD
profile is used to determine whether inverse stratification is present and sample accordingly. During
stratified conditions that occur in the Summer, the CTD profile is used to determine the depths of discrete
positions in the thermal structure of the lake such as the mid-epilimnion. The Chief Scientist or Shift
Supervisor and the Rosette operator will interpret these readings and determine the depths of samples to
be collected.
Further information on sample depth selection can be found in the WQS manual LG200, Field Sampling
Using the Rosette Sampler, which includes a figure that illustrates an example thermal profile.
Controlling Contamination: Concentrations of chemicals in lake water are very dilute. A small amount
of sample contamination can have a large effect on the results. Avoiding contamination is, therefore, a
major goal of the Water Quality Surveys QA program. To reduce contamination from atmospheric dust,
empty bottles will be capped during preparation from sampling. Care also should be taken in the storage
of bottles to reduce exposure to "dirty" environmental conditions. The Niskin bottles used on the Rosette
sampling device are opened as the device is dropped. This has the effect of flushing the bottles with
thousands of gallons of water and effectively rinses the bottles clean of any carryover from previous
sampling activities. Historical data has confirmed the effectiveness of this approach. As described
below, samples collected in Niskin bottles with the Rosette device are transferred to pre-rinsed 500 mL
brown and 1 gallon clear polyethylene containers (PEC) when the device is brought back on ship.
Aliquots are then processed (filtration and preservation) from the 1 gallon containers for storage in 125
mL bottles for nutrient determination. Techniques to control sources of contamination during these field
activities are described in Table 8-1.
Page 34
March 2003
-------
Table 8-1: QA Techniques to Control Contamination During Field Sampling Activities
Contamination
Source
Control Strategy
Sampling
Equipment/
Bottles
Open Niskin bottles as Rosette is lowered
Use new PEC containers to hold the sample for onboard analyses
and laboratory preparations
Rinse each PEC bottle with sample before filling the container.
Atmospheric
Cap empty bottles during all on-site activities (except actual
collection or transfer of sample)
Store bottles in clean environments (e.g., avoid exhausts, dirt, dust,
organic vapors, etc.)
Replace bottle caps immediately after adding preservative or, for
those that do not require preservation, immediately after transfer to
the permanent containers
Avoid repeated transfer of samples from one container to another
Perform sample preservation and filtration activities in the ship
laboratory
Preservatives
Use automatic pipettes or dispensers when adding preservative
Confirm that preserving agent has the same colored tags as that
used on the sampling bottle. Investigate/implement corrective
actions if mismatch occurs.
"Set up" dissolved oxygen samples immediately (see Section 9.1)
Filtration
Use pre-cleaned filters that are stored in a contaminant-free
environment
Specific sampling methods to be used in the Surveys are detailed in Sampling and Analytical Procedures
for GLNPO 's Open Lake Water Quality Survey of the Great Lakes (updated regularly, maintained
onboard the R/VLake Guardian during each cruise, and available from GLNPO). Sampling requirements
also are described in the QAPPs prepared by the GLAS contractor, the biology grantee, and other
organizations tasked with implementing sampling activities that support the Water Quality Surveys. Key
requirements from these documents are highlighted below.
8.1 Sampling with the Rosette
A 12-bottle Rosette sampler system (SeaBird Electronics Carousel Water Sampler) is used to collect
water samples for the following Survey parameters:
all nutrient parameters • total suspended solids
phytoplankton • turbidity
chlorophyll a • specific conductance
dissolved oxygen • pH
temperature
The Rosette sampling system consists of an A-frame structure, 1000 feet of multi-conductor cable, a
variable speed winch, a SeaBird Electronic Deck Unit with attached computer, a conductivity,
temperature, and depth (CTD) sensor, an optical transmission sensor, a sensor for measuring photo-
synthetically active radiation, a fluorometer with filters for chlorophyll detection, a pH electrode, and a
dissolved oxygen sensor attached to the end of the cable. The sampling system allows an operator to
remotely actuate a sequence of up to twelve 8-liter Niskin sampling bottles. The bottles can be closed
March 2003
Page 35
-------
remotely from the deck in any pre-determined order while the array is submerged at various sampling
depths.
The CDT measures water depth and temperature, which are graphically displayed onboard the research
vessel. Sampling depth is detected by a pressure transducer on the CTD. To ensure that the display
parameters are set to include the entire water column, the bottom sounding will be compared to the
Rosette sample reading at each station. The Rosette winch operator obtains a depth sounding from the
bridge and documents this on the Rosette form, then adjusts the computer program parameters,
controlling the depth range to be displayed. The Rosette sampler is then lowered to the bottom at
between 0.5 and 1 meter/second.
If the bottom sensor is operational the Rosette is lowered to 1-2 meters from the bottom, and
sampling may commence.
If the bottom sensor is not providing usable results, the Rosette is lowered so that it touches the
bottom and is then raised at least five meters from the bottom. In this case, the operator waits at
least three minutes to allow the sampler to drift away from the disturbed area before dropping the
Rosette back down to the B-2 (two meters from the bottom) depth for sampling.
When the Rosette is lowered into the water, the Chief Scientist and marine technicians monitor
the sampler to maintain a vertical cable. If the cable moves off vertical, the Chief Scientist and
the ship Captain coordinate in an attempt to maneuver the ship to maintain a vertical cable. The
ship GPS system provides a constant indication of distance off station. The final location is
recorded in the Captain's log and in the hard-copy forms. (As per Section 8.3 below, benthic
samples should always be collected within 50 meters of the station location.)
Samples are taken at each of the depths discussed in Section 7.2 as the Rosette is raised. Duplicate
samples are taken sequentially at the randomly-determined duplicate sites and depths. (Duplicate
sampling locations and depths in each basin are selected by the Chief Scientist for Board Chemistry prior
to each survey using a random numbers generator.)
Additional details regarding required Rosette sampling procedures are given in GLNPO SOP LG200,
Field Sampling Using the Rosette and SOP LG301, Operating the SeaBird 25.
8.2 Zooplankton Sampling Tows
Two sampling tows will be performed at each station. The first tow is done from 20 meters below the
water surface using a 63 |_im net. The second tow is a 'full' water column tow, from 2 meters above the
bottom of the lake or 100 m, whichever is less, using a 153 |_im net. If the station depth is less than 20 m,
both tows should be taken from one meter above the bottom. Triplicate tows are made at each of the
master stations to provide QC samples. Note: Due to the limited number of organisms in Lake Superior,
two tows must be done with the 63 |_im net for the 20 m sample and combined into one bottle. The 100 m
tow is performed as in the other Lakes.
The tow net, with a screened sample bucket attached to the bottom, is lowered to the desired depth, and
raised at a constant speed to collect zooplankton from the water column. The net is rinsed to free
organisms after being lifted out of the water. The sample is concentrated into the sample bucket and is
transferred to a pre-labeled 500 mL sample bottle. After zooplankton tows have been completed,
transparency measurements are made using Secchi disks. Note: Secchi depth measurements are made
only for tows performed during daylight hours (e.g., more than one hour after sunrise and more than one
hour before sunset) and on the shady side of the boat, out of direct sunlight.
Additional details regarding sampling methodology for zooplankton are given in GLNPO SOP LG402,
Zooplankton Sample Collection and Preservation and Secchi Depth Measurement Field Procedures.
Page 36
March 2003
-------
8.3 Benthic Invertebrate Sampling Methods
Three separate samples of benthic invertebrates are taken with a Ponar grab sampler at each designated
sample station and each of the three samples is processed, preserved, and stored separately. During the
spring surveys, samples are taken at five sites: Saginaw Bay, 2 in Green Bay, and 2 in the Western Basin
of Lake Erie and are analyzed specifically for the mayfly Hexagenia. During the summer surveys, all
stations are sampled and analyzed for all benthic invertebrates. During the summer cruise, a fourth Ponar
sample is taken for grain size and chemical analysis.
Samples are collected by lowering the Ponar to the sediment surface, in accordance with procedures
outlined in GLNPO SOP LG406, Benthic Invertebrate Field Sampling Procedure. The sampling device
is then raised to the deck where the sample is emptied into a plastic bucket. The sediment and animals are
rinsed from the top screen and the interior of the Ponar. Rinsing is done at very low pressure to avoid
damaging the organisms.
All of the survey sampling stations are located in depositional areas, so the bottom surface should be soft
enough for sampling. If, however, the Ponar comes up empty, it is likely that the bottom substrate is hard
packed clay, sand, or bedrock. In such cases, the station should be relocated to deeper water (moving as
short a distance as possible) and the procedure repeated. If problems persist, the GLNPO Chief Scientist
must be consulted immediately to determine if sampling should be discontinued at the station, and if the
station location should be revised for the next survey. Note. The vessel should stay within 50 meters of
the station location for benthic sampling because benthic sites are typically near shore, and sites greater
than 50 meters away may be non-depositional zones.
If the samples contain large amounts of sand, zebra mussel shells, or other debris that prevent sediment
from quickly passing through a 500 |_im mesh, the samples should be elutriated using the procedures given
in GLNPO SOP LG406.
8.4 Meteorological Methods
The ship officer in charge of the bridge (the bridge officer) is responsible for recording wind speed and
direction, wave height and direction, air temperature, barometric pressure, visibility, present weather
conditions, and heading (when underway). These recordings are made each hour, on the hour, and are
recorded in the ship's log. In addition, the officer in charge is responsible for recording the time and
description of any unusual event (e.g., man overboard).
At each station, the bridge officer also is responsible for recording the station identification number,
arrival time, departure time, wind speed and direction, wave height and direction, barometric pressure,
water depth, surface water temperature, air temperature, geographic location (loran and/or GPS), and final
location (if vessel has drifted during sampling). This information is recorded in the Captain's log, which
yields carbon copies that are provided to GLNPO at the conclusion of the surveys.
Collection of meteorological data will be conducted in accordance with GLNPO SOP LG300,
Meteorological Data Aboard the R/VLake Guardian.
9.0 Sample Handling and Custody Requirements
Once the Rosette is brought on deck, samples are immediately aliquotted, filtered, and preserved as
described in Sections 9.1 through 9.5 and detailed in corresponding SOPs. Samples are labeled as
described in Section 6. Formal chain of custody procedures are not required for the surveys.
March 2003
Page 37
-------
Table 9-1 summarizes sample preservation and holding times for survey parameters. All survey samples
collected during the Water Quality Surveys must be preserved in accordance the requirements
summarized in this table and detailed in the corresponding SOPs. For samples that are shipped frozen to
laboratories, a frozen solution of ethanol can be used where dry ice is not available.
9.1 Rosette Sample Handling
After the Rosette is brought on deck, aliquots are immediately transferred from the Niskin bottles into
appropriate containers for each type of analysis. Sample collection and transfer is performed by the Chief
Scientist (or Shift Supervisor) with assistance from contractor, or occasionally, grantee staff. Staff
performing these operations wear latex gloves to preclude contamination of the samples during handling.
They rinse each cubitainer or polyethylene bottle with sample drawn from the Niskin bottles before
filling. Sample rinsing is performed by removing the container cap, partially filling the bottle with
sample, capping the container, and shaking to ensure the entire surface is flushed with sample. The cap is
then removed and the rinse poured out prior to filling the container.
Aliquots for nutrients, pH, conductivity, alkalinity, turbidity. Aliquots are transferred from the Niskin
bottles to pre-labeled 1 gallon polyethylene cubitainers on deck. The cubitainers are pre-labeled to
minimize transcription errors. Cubitainer caps also are pre-labeled with the same sample number. This
provides a backup means of identifying the sample if the label comes off the container. After collecting
the sample in the 1 gallon cubitainer, the cubitainer is taken to the ship laboratory for sample preservation
and storage according to the scheme described in Table 9-1 pending analysis by the GLAS contractor
(nutrients) or the board chemist (pH, specific conductance, alkalinity, turbidity). In the laboratory,
nutrient samples are processed according to the procedures in the WQS manual, LG212, SOP for
Nutrient Samples Processing.
Aliquots for chlorophyll a: The chlorophyll a aliquots are transferred directly from the Niskin bottles into
300 mL brown polyethylene bottles while on deck. Once transferred, they are immediately sent to the
onboard lab for filtration and preservation. The entire filtration and preservation procedure is carried out
as much as possible in subdued light (green) to prevent photodecomposition. Samples are treated with a
MgC03 solution during filtration to eliminate acid-induced transformation of chlorophyll to its
degradation product, pheophytin. Filtration is ideally performed within 30 minutes of collection by
vacuum filtration at <5 psi. (Though less desirable, filtration may be delayed as long as 2 hours without
compromising survey results.) The filter is stored in darkness in a capped glass tube. The filtered
samples are grouped by station and completely wrapped in aluminum foil during storage and transport to
the land-based GLAS laboratory. The samples are stored in a freezer pending transfer to the GLAS lab,
and during transit, are stored in a cooler containing dry ice.
Dissolved oxygen aliquots: During the Lake Erie surveys and at master stations, aliquots for dissolved
oxygen are immediately transferred from selected Niskin bottles by inserting an 8 to 10 inch length of
flexible plastic Tygon tubing connected to the Niskin bottle outlet plug into the bottom of a 60 mL glass
BOD bottle. Flow will be regulated by the outlet plug so as to minimize turbulence and mixture of
oxygen with the sample. The BOD bottle is filled to overflowing, allowing overflowing to continue five
seconds before adding, in series, the first two reagents, allowing the floe to settle, mixing, and allowing
floe to settle again. The dissolved oxygen sample analyses are then completed in the main laboratory.
(This procedure is used to calibrate/verify the dissolved oxygen readings collected from the SeaBird.)
Integrated composite samples for phytoplankton, chlorophyll a, particulate nitrogen, particulate
phosphorus, particulate organic carbon, and total suspended solids and nutrients (chloride, silica, total
phosphorus, dissolved phosphorus, and nitrate/nitrite): For an unstratified water column, the integrated
sample is prepared by taking equal volumes of water from SRF (1 m), 5 m, 10 m and 20 meters unless the
depth is less than 20 meters. If the total depth is between 15 and 22 meters, the 20 meter sample is
replaced by the bottom sample (B-l or B-2). If the total depth is less than 15 meters, equal volumes are
Page 38
March 2003
-------
taken from surface, mid-depth, and bottom sample (B-l or B-2). For a stratified water column, equal
volumes are taken from the surface, 5 m, 10 m, and lower epilimnion (LEP). If the epilimnion is very
shallow, equal volumes are taken from a maximum of four sampling depths and a minimum of two
sampling depths. The underlying strategy is to collect a representative sample from the epilimnion. (see
Section 7.2). The 1 liter aliquots are transferred to a 1 gallon cubitainer to create a composite from the
euphotic zone. Samples for phytoplankton analysis are preserved with 10 mL Lugol's solution (to a final
concentration of 1%) and stored in the dark. Note. Due to the low number of organisms in Lake
Superior, two 1 liter composite samples are collected for phytoplankton at each station.
9.2 Zooplankton Sample Handling
Zooplankton samples are collected and transferred to pre-labeled 500 mL sample bottles as described in
Section 8.2 and immediately transferred to the biology lab onboard the ship for sample preservation. The
organisms are narcotized with 20 mL soda water and allowed to stand for 30 minutes in the refrigerator
before being preserved with 20 mL sucrose formalin solution.
9.3 Benthic Invertebrate Sample Handling
Samples for benthos determinations are brought to the ship board laboratory for preservation after being
elutriated or sieved on deck. (Elutriation is rarely, if ever, required during the surveys.) The samples are
preserved in the lab with 50-100 mL of 37% buffered formaldehyde with Rose Bengal. The sample is
then topped off with tap water and inverted 3 times before storage in the onboard walk in refrigerator.
Samples for grain size and chemical analysis are stored unpreserved in the shipboard freezer.
March 2003
Page 39
-------
Table 9-1. Sample Preservation Methods, Holding Times, and SOPs
Parameter
Maximum
Unpreserved
Sample
Holding Time
Maximum
Preserved
Holding
Time
Operational
Storage Methods
and Holding Time
Limits
Preservation/
Storage
Method
SOP
Reference
Turbidity
not established
15 hours
2 hours,
12 hours if control
standards are out
(4°C)
Refrigerate (4°C)
LG200 or
LG201*,
LG500
Dissolved
Oxygen
perform ASAP
none
1st two reagents
immediately. Add
acid w/in 8 hours.
Titrate w/in 30
minutes of acid
addition
not applicable
LG301,
LG303,
LG304;
LG200 or
LG201*,
LG501
Specific
Conductance
unstable
15 hours
2 hours,
12 hours if control
standards are out
(4°C)
Refrigerate (4°C)
LG200 or
LG201*,
LG301,
LG500
PH
unstable
15 hours
2 hours,
12 hours if control
standards are out
(4°C)
Refrigerate (4°C)
LG200 or
LG201*,
LG301,
LG500
Alkalinity
unstable
15 hours
2 hours,
12 hours if control
standards are out
(4°C)
Refrigerate (4°C)
LG200 or
LG201*,
LG500
no3-no2
24 hours
3 months
2 hours
at land-based lab <3
months
1 mL H2S04/L in
filtered sample
LG200 or
LG201*,
LG203,
LG212
Total Dissolved
Phosphorus
24 hours
3 months
2 hours
at land-based lab <3
months
1 mL H2S04/L in
filtered sample
LG200 or
LG201*,
LG204,
LG212
Total
Phosphorus
24 hours
3 months
2 hours
at land based lab <3
months
1 mL H2S04/L in
filtered sample
LG200 or
LG201*,
LG204,
LG212
Chloride
not established
3 months
2 hours
Refrigerate (4°C)
LG200 or
LG201*,
LG205,
LG212
SiOz
not established
3 months
2 hours
at land-based lab <3
months
Refrigerate (4°C)
LG200 or
LG201*,
LG205,
LG212
Page 40
-------
Parameter
Maximum
Unpreserved
Sample
Holding Time
Maximum
Preserved
Holding
Time
Operational
Storage Methods
and Holding Time
Limits
Preservation/
Storage
Method
SOP
Reference
Particulate
Organic Carbon,
particulate
organic nitrogen,
particulate
organic
phosphorus, total
suspended
solids
filter ASAP
not
established
not established
Frozen at-10°C
LG200 or
LG201*,
LG206,
LG207,
LG208,
LG209,
LG210,
LG302
Dissolved
Organic Carbon
filter ASAP
24 hours
filtered immediately
and stored at 4°C
1 mL H2S04/L
LG200 or
LG201*,
LG210,
LG211
Chlorophyll a
30 minutes
approximate
ly 3/4 weeks
if frozen (-
20°C) in
darkness
30 minutes
unpreserved (ideal)
2 hours unpreserved
(maximum)
filter w/ MgC03.
Freeze in
darkness.
LG200 or
LG201*,
LG404,
LG405
Phytoplankton
1-2 hours
Several
months if
stored in the
dark
2 hours until
preservation
10 mL Lugol's
fixative/L sample
(store in the dark
at 4°C)
Formalin upon
arrival at the
laboratory
LG200 or
LG201*,
LG400,
LG401
Zooplankton
1 hour
Store at 4°C
ASAP
unlimited
Narcotize and
refrigerate
immediately and not
later than 1 hour after
sample collection.
Preserve 30 minutes
later.
Narcotize with 20
mL soda water; let
stand 30 minutes
at 4°C; preserve
with 20 mL
sucrose/formalin
solution
LG402,
LG403
Benthic
invertebrates
1-2 hours
unlimited
1 hour until
preservation
50 to 100 mL 37%
formaldehyde with
Rose Bengal
solution at 4°C
LG406,
LG407
Sediment
Samples for
chemical and
grain size
analysis
Preserve ASAP
not
established
1 to 2 hours
Freeze
LG406,
LG600,
LG601,
LG602
* LG201 when Rosette is not operational.
Page 41
-------
10.0 Analytical Methods Requirements
10.1 Nitrate/Nitrite
Nitrate and nitrite forms of nitrogen are determined on grab and composite samples taken from the
Rosette. (Sample preparation is conducted on board the R/VLake Guardian as described in Section 9.1.)
The analyses are performed in a land-based laboratory by the GLAS contractor, in accordance with
GLNPO SOP LG203, Nitrate-Nitrite in Lake Water, QuickChem FIA + 8000 Method. Briefly, nitrate is
quantitatively reduced to nitrite by passing the sample through a column containing copper coated
cadmium. The nitrate (reduced nitrite plus original nitrite) is determined by diazotizing with
sulfanilamide dihydrochloride. The resulting water soluble dye has a magenta color which is read at
520 nm.
10.2 Total Phosphorus and Total Dissolved Phosphorus
Total phosphorus and total dissolved phosphorus are determined on grab and composite samples taken
from the Rosette and on sediment samples taken during the summer surveys. (Sample preparation is
conducted on board the R/VLake Guardian as described in Section 9.1.) The analyses are performed in a
land-based laboratory by the GLAS contractor, in accordance with GLNPO SOP LG204, Total and Total
Dissolved Phosphorous (Lachat Method 10-115-01-1-F for QuickChem FIA+8000). In this procedure,
samples are digested in the presence of sulfuric acid and persulfate to convert (hydrolize) polyphosphate
and organic phosphorous to orthophosphate. The orthophosphate ion reacts with ammonium molybdate
and antimony potassium tartrate under acidic conditions to form 12 molybdophosphoric acid. This
complex is reduced with ascorbic acid to form a blue complex that absorbs light at 880 nm. The
absorbance is proportional to the concentration of orthophosphate in the sample.
10.3 Chloride and Silica
Chloride and silica are determined on grab and composite samples taken from the Rosette. (Sample
preparation is conducted on board the R/VLake Guardian as described in Section 9.1.) The analyses are
performed in a land-based laboratory by the GLAS contractor, in accordance with GLNPO SOP LG205,
Chloride and Silica in Lake Water (LachatMethod). In this procedure, thiocyanate ion is liberated from
mercuric thiocyanate by the formation of soluble mercuric chloride. Free thiocyanate ion forms the
highly colored ferric thiocyanate in the presence of ferric ion. The highly colored ferric thiocyanate
absorbs strongly at 480 nm, producing a response that is proportional to the chloride concentration. The
calibration curve is non-linear.
Soluble silica species react with molybdate under acidic conditions to form a yellow silica molybdate
complex. This complex is subsequently reduced with l-amino-2-napthol-4-sulfonic acid (ANSA) and
bisulfite to form a heteropoly blue complex that has an absorbance maximum at 820 nm.
10.4 Particulate Organic Carbon
Particulate organic carbon (POC) is determined on grab and composite samples taken from the Rosette.
(Sample preparation is conducted on board the R/VLake Guardian as described in Section 9.1.) The
analyses are performed in a land-based laboratory by the GLAS contractor, in accordance with GLNPO
SOP LG207, Analysis of Particulate Phase Organic Carbon. In this procedure, subsamples of the
exposed glass fiber filters are placed into tin capsules and loaded into a Carlo Erba Elemental Analyzer
1108 equipped with an autosampling device. The elemental analyzer comprises a combustion furnace,
GC oven, and thermal conductivity detector. Once loaded into the autosampler, the samples are
automatically dropped in a vertical quartz reactor tube inside a 1000°C furnace. The temporarily oxygen-
enriched carrier gas oxidizes the sample. Quantitative oxidation is achieved when the resulting mixture of
Page 42
March 2003
-------
gases passes over a tungsten anhydride catalyst. The gas mixture then passes over the elemental copper
where excess oxygen is removed and nitrous oxide is reduced to elemental nitrogen. The sample gases
then pass through a packed chromatographic column to be separated, eluted, and detected by a thermal
conductivity detector. Organic carbon is quantified by the external standard method.
10.5 Particulate Nitrogen
Particulate nitrogen is determined on integrated composite samples taken from the Rosette. (Sample
preparation is conducted on board the R/VLake Guardian as described in Section 9.1.) The analyses will
be performed in a land-based laboratory by the GLAS contractor, in accordance with GLNPO SOP
LG208, Particulate-Phase Total Nitrogen by Alkaline Persulfate Oxidation Digestion (LachatMethod).
In this procedure, nitrogen compounds are spontaneously oxidized to nitrate in alkaline persulfate when
the media is autoclaved under 15 psi and 121°C. Nitrogen losses are not observed when oxidation occurs
under these pressure conditions. Nitrate is quantitatively reduced to nitrite by passing the sample through
a column containing copper coated cadmium. The nitrite (reduced nitrate plus original nitrite) is
determined by diazotizing with sulfanilamide dihydrochloride. The resulting water soluble dye has a
magenta color that is read at 520 nm.
10.6 Particulate Phosphorus
Particulate phosphorus is determined on the integrated composite samples taken from the Rosette.
(Sample preparation is conducted on board the R/VLake Guardian as described in Section 9.1.) The
analyses are performed in a land-based laboratory by the GLAS contractor, in accordance with GLNPO
SOP LG209, Particulate-Phase Total Phosphorous by Persulfate Oxidation Digestion (Lachat Method).
In this procedure, the filters are digested in the presence of sulfuric acid and ammonium persulfate to
hydrolyze polyphosphates and organic phosphorous to orthophosphate. The orthophosphate ion reacts
with ammonium molybdate and antimony potassium tartrate under acidic conditions to form 12-
molybdophosphoric acid. This complex is reduced with ascorbic acid to form a blue complex that
absorbs light at 880 nm. The absorbance is proportional to the concentration of the orthophosphate in the
sample.
10.7 Dissolved Organic Carbon
Dissolved organic carbon (DOC) will be determined on grab and composite aliquots from the Rosette
sampling device. (Sample preparation is conducted on board the R/VLake Guardian as described in
Section 9.1.) The analyses are performed in a land-based laboratory by GLAS contractor staff in
accordance with GLNPO SOP LG211, SOP for Dissolved Organic Carbon. Determination of organic
carbon requires the removal of inorganic carbon which is present in the samples as carbonate. This is
accomplished with an automated pre-treatment system in which a high velocity stream of organic-free air
transforms the sample into a thin turbulent liquid film that is rapidly transported through a large bore coil
that provides the necessary surface area for efficient removal of carbon dioxide. Up to 500 mg of
inorganic carbon can be removed at a purge rate of 500 mL/minute with minimal loss of volatile organic
compounds. An aliquot of the carbonate-free sample is then segmented and presented for analysis. The
aliquot is mixed with a stream of acid and potassium persulfate and subjected to ultraviolet (UV)
radiation. The resultant C02 generated from the organic carbon present in the sample is then purged with
a stream of C02-free air or nitrogen and measured with a non-dispersive infrared analyzer. The dissolved
organic carbon concentrations are calculated from peak heights generated by a chart recorder connected to
the infrared analyzer.
March 2003
Page 43
-------
10.8 Total Suspended Solids
Total suspended solids (TSS) concentrations are determined on integrated composite samples collected
from the Rosette sampling device. TSS filtration is performed on board the ship by GLNPO scientists in
accordance with GLNPO SOP LG302, Sampling and Analysis of Total Suspended Solids in Lake Water.
Aliquots from each desired depth are poured from the Rosette and filtered under vacuum through a 47
mm diameter pre-weighed glass fiber filter. The suspended solids are retained on the filter and frozen at
-10°C until final weighing by the GLAS contractor on an analytical balance in the contractor's land-based
laboratory.
10.9 Specific Conductance, Total Alkalinity, Turbidity, and pH
Grab samples collected with the Rosette sampling device as described in Sections 8.1 and 9.1 are
analyzed on board the R/VLake Guardian by GLNPO scientists with assistance from the marine
technicians. The equipment used for these analyses is standardized for each of these parameters at the
beginning of each lake survey. Samples are analyzed in accordance with the procedures detailed in
GLNPO SOP LG500, Standard Operating Procedure for GLNPO Board Analyses. Briefly, this
procedure involves 1) placing a 250 mL sample on the specific conductivity apparatus and heating it with
allO volt immersion heater to exactly 25.0 °C. The conductivity reading is recorded, and the pH
electrode is then placed in the sample. As soon as the pH meter indicates that the pH reading is stable,
that reading is recorded, the beaker is removed from the apparatus, and the content is used to fill the
turbidity cuvette and the 100 mL alkalinity flask. The content of the alkalinity flask is transferred to the
alkalinity apparatus and titration is performed. The readings from the alkalinity titration and the
turbidimeter are recorded on the board chemistry data forms.
pH measurements also are collected from the SeaBird using procedures described in the QAPP Addendum
to The Great Lakes Water Quality Studies of Lakes Michigan, Huron, Erie, Ontario, and Superior. This
QAPP is included as Appendix C to the Sampling and Analysis Manual for the Great Lakes Water
Quality Surveys. Results are averaged every half meter and saved with the SEASAVE software program
(described in the Appendix C QAPP) to convert the raw data files to graphic or tabular format. Note: The
pH measurements taken off the SeaBird are primarily used, along with other SeaBird data, to define lake
conditions and determine appropriate sampling depths. pH measurements collected from the SeaBird are
not considered to be as reliable as those determined in the laboratory and are, therefore, not typically used
by GLNPO for purposes other than depth selection.
10.10 Dissolved Oxygen
SeaBird Method: DO measurements are collected using the SeaBird in all lakes during the Summer
surveys. During summer surveys, when the lakes are stratified, samples are collected for DO
determination each master station in each lake. A full SeaBird profile is recorded for DO. In addition, a
surface and B-2 sample will be collected and analyzed in duplicate by Winkler titration (see below).
Simultaneously, an oxygen-saturated sample, either from a sampled same depth or from the reagent water
system, will be analyzed by Winkler titration. The B-2 samples is collected only when the nepheloid
layer is present. For monitoring stations in Lake Erie a B-l sample is collected instead of the B-2. An
additional DO survey is conducted in the central basin of Lake Erie. Refer to Appendix D, Sampling and
Analytical Procedures for GLNPO's Open Lake Water Quality Survey of the Great Lakes, for DO and
temperature profiles for this survey. Detailed procedures for operating the SeaBird can be found in
Appendix D and also in SOP LG200, Field Sampling Using the Rosette Sampler.
Winkler Method: Dissolved oxygen is measured on water samples from selected depths in Lake Erie at
each station on the summer survey, and from the master stations on the other lakes. Analysis of these
samples is made by the azide modification of the Winkler test. The sample is sequentially treated with
manganous sulfate, a potassium hydroxide/potassium iodide solution, and sulfuric acid. The initial
Page 44
March 2003
-------
precipitate of manganous hydroxide combines with the dissolved oxygen to form a brown precipitate,
manganic hydroxide. Upon acidification, the manganic hydroxide forms manganic sulfate, which acts as
an oxidizing agent to release free iodine from the potassium iodide. The iodine, which is
stoichiometrically equivalent to the dissolved oxygen in the sample is then titrated with sodium
thiosulfate. These analyses are performed to calibrate/verify the SeaBird readings.
10.11 Phytoplankton
Phytoplankton identification and biovolume are determined on the integrated sample (a composite of the
epilimnion in stratified conditions and in unstratified conditions a composite of several upper depth
samples, see Section 7.2) collected at each site using the Rosette sampling device and also on an
additional sample from the deep chlorophyll layer (DCL) during Summer surveys. Analysis is performed
by grantee staff in a land-based laboratory in accordance with GLNPO SOP LG 401, Phytoplankton
Analysis. The method, called the Modified Utermohl method, involves the microscopic examination of a
preserved water sample. Initially, a preliminary scan is made to determine the volume of sample needed
for all portions of the procedure. A settled sample of appropriate volume is then examined at a
magnification of 500x for non-diatom algae and Urosolenia species (hereafter referred to as 'soft algae').
A second examination is performed on a cleaned diatom preparation for identification and enumeration at
a magnification of 125 Ox. The soft algae portion of the settled phytoplankton samples are examined and
analyzed using an inverted microscope (Leitz Diavert or equivalent microscope). During examination of
the settled sample, most diatoms are enumerated and identified only as live pennates, empty pennates, live
centrics, and empty centrics, with the only exception being species of Urosolenia (=Rhizosolenia). Actual
species identification of diatoms (excluding Urosolenia) and cell volume measurements are done under oil
immersion (1250x) by another method. While not included in the regular counts, note should be made of
the presence of other identifiable species. Diatoms should be identified and enumerated at 1250x.
Identification should be down to the lowest taxonomic rank possible. Results for the sample
sedimentation procedure are reported as cells per mL. Biovolume is calculated using formulas
representing the closest approximation of geometric shape. The data from the diatom slides is reported as
percent composition of the 125 Ox count. This percent is applied back to the live diatom counts at 500x to
determine a cells/mL count for each species.
10.12 Zooplankton
Zooplankton biomass and identification are performed by grantee staff in a land-based laboratory in
accordance with GLNPO SOP LG 403, Zooplankton Analysis. The method is a microscopic examination
of a preserved zooplankton sample collected from the conical tow nets. Microcrustacea are examined in
four stratified aliquots under a stereoscopic microscope. Rotifera are examined in two equal volume
sub-samples under a compound microscope. Four sub-samples are examined and enumerated. All
microcrustaceans are identified and enumerated under a dissecting microscope. Rotifers and nauplii also
are counted but only from the tow taken with the 63 mm mesh net. Tows taken with the larger mesh
(153 mm) will not capture sufficient numbers of the smaller rotifers. All rotifers, microcrustacean nauplii
and Dreissena veligers and post-veligers are identified and enumerated under a compound microscope at
lOOx magnification.
10.13 Chlorophyll a
Chlorophyll a determinations are made by grantee staff in a land-based laboratory in accordance with
GLNPO SOP LG 405, In vitro Determination of Chlorophyll a in Freshwater Phytoplankton by
Fluorescence. Chlorophyll-containing phytoplankton in a measured volume of sample water are
concentrated onto a glass fiber filter by low vacuum filtration. After sonication, pigments are extracted in
a 90% buffered acetone solution for 16 to 24 hours at -20°C. The extracted slurry is removed by
filtration, and the fluorescence of the extract is read using a digital fluorometer. The concentration in the
March 2003
Page 45
-------
natural water sample is reported in |ig/L.
In situ chlorophyll measurements also are made from the SeaBird and averaged every half meter. Data
are saved in the SEASAVE program as described in the QAPP Addendum to The Great Lakes Water
Quality Studies of Lakes Michigan, Huron, Erie, Ontario, and Superior.
10.14 Benthic Invertebrates
Benthic invertebrates are identified and enumerated by grantee staff in a land-based laboratory in
accordance with GLNPO SOP LG 407, Benthic Invertebrate Laboratory Procedure. Prior to analysis,
they are rinsed to remove the formalin. The rinsed sample is spread on a gridded 500 mm mesh
bottomed, stainless steel pan and subsampled using a 30-sided die or a random numbers table and a grid-
sized cookie cutter. One or more subsampled grids are transferred to a smaller container where individual
organisms can be examined under a dissecting microscope. Samples are sorted according to type and
counted as sorting occurs.
10.15 Meteorology Measurements
Meteorological measurements are collected on the bridge, as described in Section 8.4 and GLNPO SOP
LG 300, Meteorological Data Aboard the R/VLake Guardian.
11.0 Quality Control Requirements
GLNPO implements several types of field and laboratory QC measures to monitor and control the quality
of all data gathered from water column and sediment samples. These measures are used to identify and
correct problems as they occur and to define the quality of the data generated for the surveys. The
specific types of QC measures employed vary according to the sample collection process, the
measurement process, and the location of the analytical laboratory. The QC sample requirements are
provided in each analytical SOP in the WQS manual. For many parameters, a field duplicate, lab
duplicate, and field reagent blank are collected with each group of 3, 4, or 5 stations depending on the
lake. A Random Number Generator (RNG) is used to determine the stations and depths of these QC
samples. Where basins are well defined, at least one of each is collected from each basin. In general, QC
measures include one or more of the following:
• Field duplicates or field replicates are used to verify that the entire sampling and analysis process
is capable of yielding reproducible results. These duplicates also can be used to develop
quantitative estimates of imprecision associated with field and analytical activities. Field
duplicates are collected for the nutrients and many of the all total and dissolved parameters
collected off the Rosette sampling device (except for the integrated composite samples), and field
triplicates are collected for all benthos samples at all stations and for all zooplankton tows at the
master stations. Field duplicates also are collected for Secchi disk measurements whenever a
field duplicate is collected for a sample collected at 1 meter below the surface. Performance
criteria for these field duplicates are given Table 11-1. These performance criteria are under
ongoing review and can change following re-evaluation. The performance criteria also are listed
in each SOP in the WQS manual.
Lab duplicates or lab splits are used to verify that the laboratory measurement process is capable
of yielding reproducible results. These duplicates also can be used to develop quantitative
estimates of imprecision associated with analytical activities. This measure primarily serves to
identify and correct problems as they occur so that overall data quality objectives for the surveys
are not compromised. Lab duplicates are performed on all nutrient, many physical, and
biological parameters analyzed by the GLAS contractor and biology grantee. For most
Page 46
March 2003
-------
measurements except the particulate nutrients, the lab duplicate can be used in combination with
the field duplicate to separate sources of field and lab error. (True field duplicates are not
collected for the particulate nutrients.) Also, lab duplicates are not performed on the board
chemistry parameters because the field duplicates serve to capture the same information.
Performance criteria for required lab duplicates are given in Table 11-2. These performance
criteria are under ongoing review and can change following re-evaluation.
• Laboratory Check Standards are used to verify that the laboratory procedures and systems are in
control for all physical and chemical measurements (standards do not exist for phytoplankton,
zooplankton, and benthos). These check standards also can be used to develop quantitative
estimates of bias associated with analytical activities. Generally, these include a high
concentration check standard (CH) to verify that the process is yielding accurate measurements at
the upper end of the measurement range, and a low concentration check standard (CL) to verify
the process yields accurate results at the lower end of the measurement range. Both standards are
treated as much like field samples as possible to reflect laboratory procedures. (These standards
do not reflect field operations such as filtration and transfer from the cubitainer to the sample
storage bottle.) Bias limits, defined in terms of absolute or percent differences from the 'true'
concentration' are given in Table 11-3. Precision also can tracked over time by maintaining
control charts of all lab check standard analyses and calculating the standard deviation of these
analyses. The precision and bias results can be used to quantitate the accuracy of the
measurement process.
Lab spiked field samples (also known as matrix spike samples) are used to evaluate how well the
laboratory and method procedures are recovering the target parameters from the sample matrix.
These spiked samples also can be used to develop quantitative estimates of bias associated with
analytical activities. In the water quality surveys, such tests are performed for particulate
nitrogen and particulate phosphorous at a frequency of 1 matrix spike per 40 samples. Accuracy
limits for these spikes are given in Table 11-4.
Reagent blanks (also known as method blanks) are used to characterize the effects of
contamination that might be introduced by laboratory processes and systems (reagents,
instruments, atmosphere, etc.) on sample results. Such blanks are analyzed for many physical and
chemical measurements except the board parameters (field blanks capture any required
information concerning contamination about the board parameters). Reagent blanks are not
created and analyzed for phytoplankton, zooplankton, or benthos samples because the possibility
of contaminating samples with these parameters is minimal to non-existent. Performance criteria
for laboratory reagent blanks that are required in the surveys are given in Table 11-5.
Field blanks are used to characterize the effects of contamination that might be introduced during
either the sampling, sample handling, sample transport, or laboratory analysis processes on
sample results. One field blank is collected from each basin (where basins are well-defined) to
demonstrate the absence of contamination during the Rosette sampling, sample transfer, sample
preservation and storage process. Field blanks are not collected for the phytoplankton,
zooplankton, or benthos because the logistics involved in collecting such a sample are not
warranted by the remote possibility of contaminating samples with these parameters. Table 11-6
gives performance criteria that will be applied to all field blanks collected during the Water
Quality Surveys.
In addition, for the turbidity determination, the formazin calibration standard is prepared prior to each
survey and checked at least once prior to each cruise.
Analysts must compare analytical results to the performance criteria listed in the pertinent SOP to identify
March 2003
Page 47
-------
QC failures. If the results are outside the acceptance criteria, the analyst should first review their
calculations for errors and if none are identified, they must follow the corrective action procedures listed
in each SOP. Corrective action procedures will often be handled at the bench level by the analyst, who
reviews the preparation or extraction procedure for possible errors, checks the instrument calibration,
spike and calibration mixes, instrument sensitivity, and any other potential sources of error. If failure
occurs and an error is identified, the analyst should re-run quality control and RFS samples in the entire
analytical batch to confirm the results. Because analysis of field duplicates and lab duplicates usually
occurs after leaving a specific station and re-sampling is largely impossible, re-analysis of these samples
to confirm results may be the limit to corrective actions when all other QC samples within a batch meet
acceptance criteria. For analyses conducted onboard, if the problem persists or cannot be identified, the
matter must be referred to the Chief Scientist for further investigation. Depending upon the Chief
Scientist's evaluation, the analyst may or may not be required to prepare and re-run the samples. Once a
decision is made, full documentation of the corrective action procedures and assessment of the final result
must be filed with the WQS QM Technical Lead (Marvin Palmer) or the GLNPO QM. For analyses
conducted at contract or grantee laboratories, this information can be included with submitted data. When
contractor or grantee laboratories have a question regarding acceptable corrective actions, they should
contact the Biology Technical Lead or Limnology Technical Lead as appropriate for instruction at a time
when corrective action can still be taken.
Page 48
March 2003
-------
Table 11-1. Method Performance Criteria - Overall Precision Measures
Description
Survey Parameter
Minimum Frequency
Performance Criteria
Relative percent
difference between
field duplicates
Chloride
Nitrate + Nitrite
Silica
One per basin
Relative Percent Difference < 20%
(Measures precision
of sampling,
Total Dissolved Phosphorus
Total Phosphorus
One per basin
Difference < 2.0 |jg/L
transport, and
analysis procedures-
also referred to as
system precision)
Alkalinity
One per basin
Difference < (1 + 0.01 x mean reading) mg/L
pH
One per basin
Difference < 0.3 SU
Turbidity
One per basin
Difference < (0.1 + 0.1 x mean reading) NTU
Conductivity
One per basin
Difference < 2.0 pmhos/cm
Dissolved Organic Carbon
Particulate Phosphorus
Particulate Organic Carbon
Particulate Nitrogen
Total Suspended Solids
None
Samples are filtered in duplicate on ship. See Table 11-2
Phytoplankton
Dissolved Oxygen - Winkler Method
None
—
Zooplankton
Master stations collected in triplicate for each net
mesh size
Relative Standard Deviation criteria under consideration
Secchi Disk Measurement
Every time a field duplicate of the Surface sample
is collected
Difference < 0.5 meter; two different analysts should take
the duplicate measurements
Chlorophyll a
One per batch
Relative Percent Difference < 25% (interim limit)*
Benthos
100% of samples collected in triplicate
Relative Standard Deviation criteria under consideration
Meteorology
Not required
Not applicable
* Interim Limit = These limits will be used until there is enough data to calculate performance limits for this procedure.
Page 49
-------
Table 11-2. Method Performance Criteria -
Lab Precision Measures
Description
Survey Parameter
Minimum Frequency
Performance Criteria
Relative percent
difference of lab
Total Dissolved Phosphorus
Total Phosphorus
One per basin
Difference < 2.0 |jg/L
duplicates
Measures lab
precision
Chloride
Nitrate + Nitrite
Particulate Phosphorus
Particulate Nitrogen
Particulate Organic Carbon
Dissolved Organic Carbon
Silica
One per basin
Relative Percent Difference < 20%
Total Suspended Solids
At least once during the sampling of
each Great Lake
± (0.2 mg/L + 0.2 x mean)
Phytoplankton
10% of samples are analyzed by a
second analyst
Bray-Curtis Index of 60 (interim limit)*
Zooplankton
10% of samples are analyzed by a
second analyst
Percent similarity of 90% (see LG403 for calculations)
Chlorophyll a
One per batch
Relative Percent Difference < 25% (interim limit)*
Benthos
10% of samples
Identification corroborated by second analyst; < 10% difference (with
special consideration for samples with very low numbers of
individuals)
Dissolved Oxygen - Winkler
Each Master station
Absolute difference < 0.2 mg/L
Method
DO Surveys: All SRF and B-1 samples
Alkalinity
Conductivity
Meteorology
pH
Turbidity
None
* Interim Limit = These limits will be used until there is enough data to calculate performance limits for this procedure.
SRF = Surface (one meter below the surface)
B-1 = Bottom (station depth) minus one meter
Page 50
-------
Table 11-3. Method Performance Criteria - Lab Bias Measures**
Description
Survey Parameter
Minimum Frequency
Performance Criteria
Lab Check
Standards
(standard solutions
Nitrate + Nitrite
At the beginning & end of each batch or 1 per
40 samples, whichever is more frequent
Accuracy (mg/L):
High Check Standard: 0.60 ± 0.09;
Low Check Standard: 0.20 ± 0.03
analyzed at one or
more
concentrations to
verify that
Total Dissolved Phosphorus
Total Phosphorus
At the beginning & end of each batch or 1 per
40 samples, whichever is more frequent
Accuracy (|jg/L):
High Check Standard: 15.0 ±3.0;
Low Check Standard: 3.0 ± 2.0 (interim limit)*
laboratory
processes are in
control) and other
comparisons
(Measures lab bias)
Chloride
At the beginning & end of each batch or 1 per
40 samples, whichever is more frequent
Accuracy (mg/L):
High Check Standard: 17.3 ±1.2;
Low Check Standard: 5.6 ± 0.6
Silica
At the beginning & end of each batch or 1 per
40 samples, whichever is more frequent
Accuracy (mg/L):
High Check Standard: 0.467 ± 0.053;
Low Check Standard: 0.093 ±0.018
Particulate Organic Carbon
At the beginning & end of each batch or 1 per
40 samples, whichever is more frequent
Accuracy (|jg/L):
High Check Standard: 80.0 ± 16.0;
Low Check Standard: 5.0 ± 3.0
Particulate Nitrogen
At the beginning & end of each batch or 1 per
40 samples, whichever is more frequent
Accuracy (mg/L):
High Check Standard: 1.20 ±0.12;
Low Check Standard: 0.40 ± 0.04
Particulate Phosphorous
At the beginning & end of each batch or 1 per
40 samples, whichever is more frequent
Accuracy (|jg/L):
High Check Standard: 120.0 ± 12.0;
Low Check Standard: 40.0 ± 4.0
Dissolved Organic Carbon
At the beginning & end of each batch or 1 per
40 samples, whichever is more frequent
Accuracy (mg/L):
High Check Standard: 5.14 ±0.9.0;
Low Check Standard: 1.28 ± 0.60
Alkalinity
At the onset, starting with the initial calibration
of instruments for each lake survey, & after
the last station on each 12 hour shift
Accuracy (mg/L):
High Check Standard: 97-103 (control) or 98-102 (warning);
Low Check Standard: 38-42 (control) or 38.5-41.5 (warning)
Page 51
-------
Table 11-3. Method Performance Criteria - Lab Bias Measures**
Description
Survey Parameter
Minimum Frequency
Performance Criteria
PH
At the onset, starting with the initial calibration
of instruments for each lake survey, & after
the last station on each 12 hour shift
Accuracy (SU):
High Check Standard: 9.18 ± 0.3 (control) or 9.18 ± 0.2 (warning);
Low Check Standard: 6.86 ± 0.3 (control) or 6.86 ± 0.2 (warning)
Conductivity
At the onset, starting with the initial calibration
of instruments for each lake survey, & after
the last station on each 12 hour shift
Accuracy (pmhos/cm):
High Check Standard: 290-297 (control) or 291-296 (warning);
Low Check Standard: 193.5-199.5 (control) or 194.5-198.5 (warning)
Turbidity
At the onset, starting with the initial calibration
of instruments for each lake survey, & after
the last station on each 12 hour shift
Accuracy (NTU):
High Check Standard: 10 ± 2 (control) or 10 ± 1.4 (warning);
Low Check Standard: 0.5 ± 0.3 (control) or 0.5 ± 0.2 (warning)
Total Suspended Solids
Not required but being considered
—
Benthos
Meteorology
Phytoplankton
Zooplankton
None
Chlorophyll a
Daily
±10% of true value
Dissolved Oxygen -Winkler
Method
At the beginning of each lake of a regular
survey
± 0.5 mg/L compared to the theoretical
DO surveys: At the beginning & one per shift
* Interim Limit = These limits will be used until there is enough data to calculate performance limits for this procedure.
** Note: Lab precision also can be monitored from lab check standards by maintaining control charts and calculating the standard deviation of results.
Page 52
-------
Table 11-4. Method Performance Criteria - Other Bias Measures
Description
Survey Parameter
Minimum Frequency
Performance Criteria
Lab spike into field
sample (matrix spike)
(Measures accuracy of lab
and method procedures in
the sample matrix)
and other measures for
physical parameters as
described under
performance criteria
Alkalinity
Benthos
Chloride
Chlorophyll a
Conductivity
Dissolved Organic Carbon
Meteorology
Nitrate + Nitrite
Particulate Organic Carbon
PH
Phytoplankton
Silica
Total Suspended Solids
Total Dissolved Phosphorus
Total Phosphorus
Turbidity
Zooplankton
None
Particulate Nitrogen
Particulate Phosphorus
During each batch or 1 per 40 samples, whichever is more frequent
100% ± 20%
Wind Speed
Each reading
± 2.0 knots, comparison between multiple
instruments
Wind Direction
Each reading
± 1.5 degree, comparison between
multiple instruments
Barometric Pressure
Each reading
± 2.0 mB, comparison between multiple
instruments
Air Temperature
Each reading
± 0.3°C, comparison between multiple
instruments
Sea Depth
Each reading
± 5%, compared to the SeaBird 911 data
and charted soundings
Sea Temperature
Each reading
± 0.5°C, compared to the SeaBird 911
data
Page 53
-------
Description
Survey Parameter
Minimum Frequency
Performance Criteria
Ship's Heading
During river transits by range markers and underway
± 1.5 degrees, compared to GPS
generated course over ground
Ship's Water Speed
During river transits by range markers and underway
± 0.5 knot, compared to GPS generated
course over ground
Ship's Position
Each station
± 10 feet, compared to satellite health
messages and visual dockside position
Page 54
-------
Table 11-5. Method Performance Criteria- Laboratory Contamination Measures
Description
Survey Parameter
Minimum Frequency
Performance Criteria
Reagent Blank
Nitrate + Nitrite
At the beginning & end of each batch or 1 per 40 samples,
whichever is more frequent
0.00 ± 0.03 mg/L
(Measures the presence
or absence of
contamination introduced
by laboratory systems,
Particulate Phosphorus
Total Dissolved Phosphorus
Total Phosphorus
At the beginning & end of each batch or 1 per 40 samples,
whichever is more frequent
0.0 ± 2.0 [jg/L
reagents, or processes)
Chloride
At the beginning & end of each batch or 1 per 40 samples,
whichever is more frequent
0.0 ± 0.2 mg/L
Silica
At the beginning & end of each batch or 1 per 40 samples,
whichever is more frequent
0.000 ±0.015 mg/L
Particulate Organic Carbon
At the beginning & end of each batch or 1 per 40 samples,
whichever is more frequent
< 5 [jg/L
Particulate Nitrogen
At the beginning & end of each batch or 1 per 40 samples,
whichever is more frequent
0.00 ± 0.04 mg/L
Dissolved Organic Carbon
At the beginning & end of each batch or 1 per 40 samples,
whichever is more frequent
0.00 ± 0.60 mg/L
Alkalinity
Benthos
Chlorophyll a
Meteorology
PH
Phytoplankton
Total Suspended Solids
Turbidity
Zooplankton
None
Page 55
-------
Table 11-6. Method Performance Criteria- Field Contamination Measures
Description
Survey Parameter
Minimum Frequency
Performance Criteria
Field blank- water for
total forms in lake, filters
for particulate forms,
Total Suspended Solids
At least once during the
sampling of each Great
Lake
± 0.2 mg/L
filtered water for dissolved
forms
Nitrate + Nitrite
One per basin
0.00 ± 0.03 mg/L or < 1/10 associated field samples cone, whichever is greater
(Measures the presence
or absence of
contamination introduced
Particulate Phosphorus
Total Dissolved Phosphorus
Total Phosphorus
One per basin
0.0 ± 2.0 |jg/L or < 1/10 associated field samples cone, whichever is greater
in field sampling
procedures and includes
lab contamination)
Chloride
One per basin
0.0 ± 0.2 mg/L or < 1/10 associated field samples cone, whichever is greater
Silica
One per basin
0.000 ± 0.015 mg/L or < 1/10 associated field samples cone, whichever is greater
Particulate Nitrogen
One per basin
0.00 ± 0.04 mg/L or < 1/10 associated field samples cone, whichever is greater
Dissolved Organic Carbon
One per basin
0.00 ± 0.60 mg/L or < 1/10 associated field samples cone, whichever is greater
Alkalinity
One per basin
< 3 mg/L
Conductivity
One per basin
< 1 |j mho/cm
Turbidity
One per basin
< 0.15 NTU
Particulate Organic Carbon
One per basin
< 5 [jg/L
Benthos
Meteorology
Phytoplankton
Zooplankton
None
Chlorophyll a
One per batch
0.00 Mg ±0.11
Page 56
-------
12.0 Instrument/Equipment Testing, Inspection, and Maintenance
Requirements
All laboratory instrumentation will be assembled and tested before each survey. Testing will consist of
checking all control standards on the assembled systems to (1) verify proper concentration, and (2)
demonstrate that all analytical systems to be used on the R/VLake Guardian are capable of running
within the limits required using the current standards.
After the equipment is installed on the R/VLake Guardian, a series of stations, usually from Saginaw
Bay, will be used for shakedown purposes. In most cases, the entire sampling process is implemented
during this shakedown. The GLNPO Monitoring Team Lead will use the shakedown process and results
to evaluate the procedures and analytical systems to determine the status of equipment and personnel
readiness. Corrective actions will be implemented by the Monitoring Team Lead as warranted by specific
problems.
Reagent water for use in the onboard laboratory is produced on the ship. Filters in the water system
should be changed at least once per year. The super still is equipped with gauges that indicate when the
system is working properly. If questions arise regarding the adequacy of the water purification system,
the Chief Scientist or shift supervisor is responsible for evaluating the system.
During the cruise, the lead scientist for the functional area onboard ship or in the land-based laboratories
will be responsible for implementing preventative and corrective maintenance procedures necessary to
preclude instrument down time and produce precise and accurate data that meets the measurement quality
objectives listed in this QAPP. Most analytical instrument and equipment manuals have sections dealing
with preventative maintenance. These sections will be read by each person operating the equipment. In
addition to the recommendations offered by the instrument/equipment manuals, QC samples and
calibration requirements for the Survey parameters will be used as an indicator of necessary equipment
maintenance (see Table 13-1). Instruments requiring calibration above normal frequency will be
identified and evaluated for maintenance.
Preventative procedures also will include inspecting all onboard instruments for worn parts or erratic
behavior (as indicated by QC results) after each survey. To prevent equipment mis-uses all EPA,
contractor, and grantee staff will be required to follow all operational procedures for each instrument
utilized. This assurance is maintained by the development of detailed SOPs (section 4.2.5;
training/certification (section 5); adherence to contractor and grantee QA procedures for equipment,
glassware, and reagents; and shakedown activities described above.
To prevent instrument downtime during the cruises, GLNPO will maintain an onboard backup recorder,
sampler, colorimeter, pump, manifold, tubing supply and small replacement parts. The GLAS contractor
and the biology grantee are each responsible for maintaining a complete inventory of the equipment they
use on ship as well as a list of frequently used items (including current vendor and catalog number). This
inventory and list will be used to refine the list of back up equipment parts needed for each cruise. The
GLNPO Chief Scientists are responsible for keeping similar inventories of all other scientific materials
needed onboard ship.
Specific records of preventative maintenance inspections, problems, and corrective actions will be
documented by the laboratory analysts in instrument log-books maintained on-site in the laboratory.
These logbooks will be periodically reviewed by the biology grantee, the GLAS Contractor, or the
Limnology Technical Lead and will be available to the GLNPO QA Manager upon request.
In the event that instruments fail and cannot be repaired during the surveys, back up equipment will be
March 2003
Page 57
-------
used or ordered. In some cases, the instruments used for back up purposes may differ from those
specified in the QAPPs. To the extent feasible, SOPs will be revised during the survey. If such SOP
revision is not feasible during the survey, it will be revised immediately after survey conclusion. In any
case, all survey staff responsible for using the new equipment will be trained by the senior scientist
onboard ship for that functional area to ensure that all staff are using the new equipment and procedures
consistently in the absence a documented SOP.
13.0 Instrument Calibration and Frequency
Instrument calibration procedures and frequencies vary according to instrument type. Where possible, a
multi-point calibration will be used to establish the full range of the instrument before survey samples are
analyzed. The frequency of this initial, multi-point calibration varies across methods due to variations in
instrument stability and calibration procedures. While some methods, such as those that employ
colorimetric detection procedures, require daily calibration of the instrument, other methods, such as
those that employ factory calibrated devices, are more stable and require less frequent calibration. All
methods involving onboard calibration procedures require that the instrument be calibrated at least once
per working shift (e.g., day) during which samples are analyzed. All calibration standards used on board
the ship (or by the land-based laboratories) must be certified as to purity, concentration, and authenticity,
or prepared from materials of known purity and composition. Instruments, such as the SeaBird CTD or
the wind speed indicator are sent to the manufacturer for calibration and are considered stable enough to
require such calibration annually or biannually. Detailed instrument calibration procedures and
frequencies will be specified in each of the SOPs prepared to support the Water Quality Surveys. Table
13-1 summarizes these requirements.
Page 58
March 2003
-------
Table 13-1. Instrument Cali
bration Procedures and Frequencies
Instrument
Calibration
Procedure
Calibration
Standard(s)
Frequency
SeaBird CTD
Factory calibrated
Factory standard
annually
Wind Speed and Direction
Young Wind Monitor
Electric Speed Indicator
company
Factory calibrated
Factory standard
when errors are
indicated based on
comparison
Barometric pressure
Belfort Instrument Company
barograph
Chelsea Boston aneroid
barometer
R.M.Young, Model 61201
Factory calibrated
Factory standard
barograph and
barometer, annually
Secchi disk
not applicable
not applicable
NA
Turner Turbidimeter
Instrument manual
Formazin
shift
YSI Model 35 Conductivity
Bridge
Instrument manual
Shunts
When standards indicate
calibration is needed.
Standards measured
prior to first and second
lake of each survey.
Meter has not required
calibration in 5 years.
AB15 and AR15 pH meter
Instrument manual
buffers pH 7 and pH
10
shift
Fisher Accumet pH meter (for
determination of alkalinity)
Instrument manual
buffers pH 7 and pH
10
shift
Lachat - total phosphorus and
total dissolved phosphorus
Instrument manual
6 concentrations of
KH2P04 and a blank
batch
Lachat - N02-N03
Instrument manual
6 concentrations of
KN03 and a blank
batch
Lachat - chloride
Instrument manual
8 concentrations of
NaCI and a blank
batch
Lachat - Si02
Instrument manual
5 concentrations of
commercial silica
and a blank
batch
Technicon - DOC analyzer
Instrument manual
Potassium
Biphthalate
batch
Flowmeter for zooplankton
tow
SOP LG402
Water
At least annually
(summer) and if weather
is calm, each cruise.
March 2003
Page 59
-------
14.0 Inspection/Acceptance Requirements for Supplies and
Consumables
Supplies and consumables used during the Water Quality Surveys generally fall into one of the following
five areas:
1) Supplies used exclusively by contractors and grantees during performance of their contracts and
grants. In this case, the contractor/grantee is wholly responsible for inspection and acceptance
and assumes responsibility for the quality of those supplies. General requirements regarding such
inspection and acceptance are outlined in each contract or grant. Additional requirements
regarding the inspection and acceptance of laboratory supplies are described in the GLAS
contractor and grantee Quality Management Plans.
2) Consumable supplies obtained by either GLNPO , the GLAS contractor, or the biology grantee
and used to support the entire Survey operation. The quality of these supplies is managed
(inspected and accepted) by the GLNPO Property Officer.
3) Consumable supplies obtained by the ship operations contractor to support the entire Survey
operation. The quality of these supplies is managed (inspected and accepted) by the Ship
Operations Project Officer.
4) The quality of food to be consumed onboard ship is inspected by the Chemical Hygiene Officer
to ensure that it is properly handled and safe.
5) Non-disposable property, such as computers, instruments, etc., that are purchased to support the
Water Quality Surveys are inspected and accepted by the Monitoring Team Lead to verify that
they are of sufficient quality to meet survey needs.
15.0 Requirements for Acquisition of Non-direct Measurement Data
The sampling and analysis activities conducted during the Water Quality Surveys do not involve the
collection of data obtained from non-measurement sources such as computer databases, spreadsheets and
programs, and literature files.
16.0 Data Management
16.1 Field Data Management
The Chief Scientist has primary responsibility for assuring that all data gathered in the survey is
documented. Documentation includes raw instrument level printouts, summary bench sheets, and
electronic records generated on board ship and in the laboratories. All shipboard-generated strip charts,
bench records, and computer printouts are kept in a folder, indexed by station, until the remaining
samples are transferred to the land-based laboratories. All raw data are assembled and indexed by
parameter, by lake, and by survey leg. Analog charts and digital conversion printouts will be stapled
together. Each parameter will be placed in manilla folder and transferred to the GLNPO Chief Scientist
for Board Chemistry upon conclusion of the survey. Several parameters are recorded on the ship's log,
such as vessel position and weather conditions. This information is described in LG 300, SOP for
Meteorological Data Aboard the R/VLake Guardian. These parameters are then recorded on the
GLENDA Station Information Field Recording Form at each station visit and entered into the GLENDA
database.
During the surveys, sample collection, preparation, preservation, and board chemistry data will be entered
into GLENDA using the GLENDA data input tool. To the extent possible, this data entry will be done at
the conclusion of each shift, and in all cases, before the end of the cruise. A Data Flow Diagram is
provided in Figure 1 of SOP LG101 in the WQS manual and presents the data management process for
data collection and data verification for the WQS. Hard-copy Field Information Recording Forms
Page 60
March 2003
-------
(Appendix H of the manual) are used to record data as it is being collected during the survey. The data is
then entered into the GLENDA shipboard Remote Data Entry Tool. Output files from the Remote Data
Entry Tool are created at the end of each survey leg and the Chief Scientist delivers them to the GLENDA
Database Manager. If the tool is not functioning properly, the data can be entered into Excel electronic
files, which are a soft-copy version of the recording forms in Appendix H, as a back-up to the tool. If this
back-up option is used, the data are entered into the Remote Data Entry Tool from these electronic files at
GLNPO headquarters or onboard the ship if the Remote Data Entry Tool begins to function during the
survey leg. Once the data are in the Remote Data Entry Tool they are uploaded to GLENDA as
unverified data upon completion of each survey.
The Chief Scientist is responsible for transferring the original field information recording forms to the QA
Team at GLNPO (Marvin Palmer or his designee) for internal quality control checks. The QA Team
conducts checks of the recording forms against the data that has been uploaded to GLENDA and
addresses any data discrepancies. After completing these checks and resolving all discrepancies, the QA
Team transfers the forms to the Environmental Monitoring and Indicators Team for storage in the
designated file cabinet at GLNPO. The pertinent Technical Lead conducts a final review of the data and
notifies the Database Manager of approval. The Database Manager then finalizes the dataset as Version 1
and makes the data available. For each dataset, the status of this data management process is tracked by
the QA Team on the Data Status Tracking Sheet (Appendix N of the WQS manual).
For each staffing period, the Chief Scientist must assure that all data has been collected, analyzed, and
entered into the data entry tool. The Chief Scientist must assure that a copy of the hard-copy field
information recording forms is made and placed in the onboard designated file cabinet at the end of each
survey leg. The Chief Scientist assures that an electronic copy of all soft-copy field information
recording forms is made and transferred to the database manager at GLNPO (Ken Klewin) for storage.
The Chief Scientist also must assure that SeaBird data files are processed at the end of each survey leg
and returned to the database manager for storage. These items are included on the list of Chief Scientist
Roles and Responsibilities (Appendix L of the manual). For the bridge data, The Officer in Charge shall
be responsible for reviewing the previous watch entries in the Ships's Log, GLENDA Station Information
Field Recording Form and the GLENDA database to assure correctness of the data entered. Errors noted
shall be corrected immediately.
16.2 GLENDA Data Management
The GLENDA database manager is responsible for collecting all survey data from the Chief Scientists,
uploading these data into the GLENDA database, and maintaining the electronic results submitted by the
Chief Scientist in its original format for retrieval if necessary. The database manager also is responsible
for implementing any database or data entry system modifications necessary to accommodate the unique
requirements of the Water Quality Surveys. Once data are uploaded to the database, the Database
Manager is responsible for ensuring control and quality of the database, implementing appropriate
security and data release procedures, and for facilitating data retrieval and data dissemination to GLNPO
staff, contractors, grantees, and members of the scientific community.
A process has been developed for correcting errors identified in verified data in GLENDA and is
illustrated in Figure 2 of SOP LG101 in the WQS manual. Once a data error is identified, a Data
Discrepancy Form (Appendix P of the manual) is completed and submitted to the Quality Assurance
Manager and Database Manager. Data are revised in GLENDA and the revisions are verified as correct
by the QA Team. The pertinent Technical Lead also conducts a final review and notifies the Database
Manager of approval. The Database Manager then finalizes the dataset as the subsequent version and
makes the data available.
March 2003
Page 61
-------
17.0 Assessments and Response Actions
Several types of assessment activities and corresponding response actions have been identified to ensure
that data gathering activities in the Water Quality Surveys are conducted as prescribed and to ensure that
the measurement quality objectives established in this QAPP and the data quality objectives established
by EPA are met. These activities are summarized in Table 17-1 and discussed in greater detail in Sections
17.1 - 17.7 below.
17.1 Surveillance
All data gathering activities are subjected to surveillance activities during the course of sample collection,
sample analysis, and data reporting. The Chief Scientist and Shift Supervisors are responsible for
monitoring the activities of all sample collection and onboard analysis activities on a daily basis during
the surveys. This daily surveillance is intended to ensure that field and lab procedures are being
implemented correctly and that all required samples and data elements are being collected. Problems
identified are immediately corrected while in the field.
A second level of surveillance monitoring is performed by the Chief Scientists responsible for each
functional area. These scientists inspect and evaluate data as it is reported by laboratory staff. Missing
data, inconsistencies, or other anomalies are identified and discussed with the Chief Scientist or Shift
Supervisor who was onboard ship at the time of collection or with the laboratory analysts responsible for
making the measurements.
Finally, the QA manager is responsible for monitoring the quality of all data gathered during the surveys.
This surveillance activity is performed when data are reported and serves as a supplement to the
surveillance activities described above.
17.2 Peer Review
A formal peer review is conducted on all survey programs every four years to ensure that monitoring
strategies identified for the surveys continue to be appropriate, identify the need for new strategies, and
ensure that data gathered during the surveys are being gathered, interpreted, and utilized in a scientifically
valid manner.
At a more immediate level, 10% of the planktonic diatom measurements made by the biology grantee are
subjected to a real-time peer review that is performed by a senior biologist to verify that the analyst is
correctly interpreting sample results.
17.3 Quality System Audits
Quality system audits (QSAs) of the data gathering activities and processes will be conducted on a
periodic basis to ensure that the data gathered during each survey fulfills GLNPO's programmatic and
QA requirements. These QSAs will be scheduled and managed by GLNPO's QA Manager.
17.4 Readiness Reviews
Several types of 'readiness' reviews will be implemented to ensure that the sampling teams, laboratories,
and staff are capable of implementing required data gathering activities before the surveys begin. These
reviews include:
1) The use of on-site audits prior to award of any new laboratory contracts (i.e., the GLAS
contract). Such audits will be conducted by the contract Project Officer and/or the GLNPO QA
Manager
Page 62
March 2003
-------
2) Analysis of standard reference materials and calibration standards prior to each survey. These
analyses will be conducted by the laboratory staff normally responsible for survey sample
analyses. Results will be reviewed by the laboratory manager and the Technical Lead
responsible for that functional area.
17.5 Technical Systems Audit
The GLNPO QA Manager, or his designee, will conduct on-site audits of the facilities, equipment, staff,
and sample analysis, training, Record keeping, data validation, data management, and data reporting
procedures used at the GLAS contract lab, the biology grantee lab, and in the onboard ship labs. These
audits will be performed on an annual or biennial basis unless the results of the readiness reviews, data
quality audits, and surveillance suggest serious or chronic laboratory problems that warrant more frequent
on-site examinations and discussion with laboratory personnel. All technical system audits will be
conducted using a standardized audit checklist to facilitate an audit walk-through and document audit
findings. At a minimum, audit participants will include the EPA Quality Assurance Manager (or
designee). Other auditors may include the Technical Lead, the Environmental Monitoring and Indicators
Team Lead, the GLAS Project Officer, the Grant Officer, or contractor QA staff. In such cases, the QA
Manager (or designee) staff member will be responsible for leading the audit and conducting a post-audit
debriefing to convey significant findings to laboratory staff at the conclusion of the audit. The second
staff member will be responsible for gathering pre-audit documentation of problems that necessitated the
audit, customizing the audit checklist as necessary to ensure that those problems are addressed during the
audit, documenting audit findings on the audit checklist during the audit, and drafting a formal report of
audit findings for review by the QAM.
17.6 Data Quality Audits
Data quality audits of survey result files will be performed on a schedule to be jointly determined by the
GLNPO QA Manager, the Environmental Monitoring Team Lead, and the Technical Lead of each
functional area. These audits will be directed at verifying that data gathered during the surveys meet the
MQOs cited in this QAPP and at identifying trends in method and laboratory performance. Individuals
responsible for performing these audits, and procedures to be used in performing them, will be
determined prior to the next revision of this QAPP.
17.7 Data Quality Assessment
The QA Team Lead for the Surveys, or his designee, will conduct reviews of hard-copy data forms
against the data in GLENDA. This review will assess several aspects of data quality including data entry
errors, assessments of results against historical data, and evaluation of method performance. The
pertinent technical lead also will perform a review and provide final approval of the data to the database
manager who will make the data available as version 1 of the data. This process is described in Section
16.1 and illustrated in Figure 1 of SOP LG101 in the Manual. For each dataset, the status of this data
quality assessment process is tracked on the Data Status Tracking Sheet (Appendix O of the manual).
Statistical and scientific analyses of the GLENDA database will be performed on a periodic basis to be
jointly determined by the GLNPO QA Manager, the Environmental Monitoring Team Lead and the
Technical Lead of each functional area. The objective of these analyses is to verify that the data being
gathered are suitable for their intended use (water quality monitoring), monitor trends in survey results
and water quality to verify that survey objectives of detecting 20% changes in water quality are being
met, and identify areas for improvement.
March 2003
Page 63
-------
able 17-1. Assessment and Response Actions
Assessment
Measure
Definition
Frequency
Responsible Party
Rationale
Surveillance
Continual or frequent monitoring and
verification of the status of an entity and the
analysis of records to ensure that specified
requirements are being fulfilled
Throughout survey activities,
while samples are being analyzed
at GLAS laboratory
Chief Scientists
during surveys;
QA mgr when data
are reported
Identify and correct field and lab analytical problems as soon
as they occur to prevent problems at future sites or surveys;
identify data reporting or QA deficiencies prior to next
survey.
Peer Review
A documented critical review of work.
Conducted by qualified individuals who are
independent, but technically equivalent of those
who performed the work.
Performed on 10% of
phytoplankton, zooplankton, and
benthos determinations
Performed on all survey
programs every 4 years
GLNPO Peer
Review
Coordinator
• Ensure organism identification and enumeration is
activities are technically adequate, competently
performed, properly documented, and satisfy established
technical and quality requirements.
• Ensure data are being correctly interpreted and utilized;
identify monitoring strategies that should be modified
Management
Systems
Review
Qualitative assessment of a data collection
operation and/or organization to establish
whether the prevailing quality management
structure, policies, practices, and procedures are
adequate for ensuring that the type and quality
of data needed are obtained.
Periodically
QA Manager
Ensure survey practices continue to meet programmatic and
QA requirements established by GLNPO
Readiness
Review
A systematic documented review of the
readiness for the start-up or continued use of a
facility, process or activity. Typically conducted
before proceeding beyond project milestones
and prior to initiation of a major phase of work.
• On-site audits prior to t award
• Analysis of SRMs, calibration
standards prior to survey
• Collect health records prior to
surveys
• Contract PO
and GLNPO
QAM
• Contract or
Grantee lab
managers
• CHO
• Verify contractor or grantee is capable of producing
precise and accurate results with the methods they will
use during the survey
• Verify all survey participants are physically capable of
participating in offshore sampling
Technical
Systems
Audit
A thorough, systematic, on-site, qualitative
audit of facilities, equipment, personnel,
training, procedures, Record keeping, data
validation, data management, and reporting
aspects of a system.
Annually or biannually
QA Manager or
designee
Ability of each laboratory to adequately analyze and report
data will be assessed prior to analysis and continually
throughout analyses via other QA/QC measures described in
this QAPP.
Audit of
Data Quality
Systematic and independent examination to
determine if quality activities and related results
comply with planned arrangements and whether
these arrangements are implemented effectively
and are suitable to achieve objectives.
TBD
GLNPO QA
Manager
To verify that all data collected meet MQOs established for
this study; identify trends in data quality with respect to
method and laboratory performance
Page 64
-------
Assessment
Measure
Definition
Frequency
Responsible Party
Rationale
Data Quality
Assessment
Statistical and scientific evaluation of the data
set to determine the validity and performance of
the data collection design and statistical test, and
to determine the adequacy of the data set for its
intended use.
TBD
Technical
Leads/Data
Managers
Evaluate overall accuracy and quality of the survey results,
identify trends in water quality.
Page 65
-------
18.0 Reports to Management
The GLNPO Management Committee meets every week to discuss data gathering activities, status, and
priorities within the Office. The Environmental Monitoring Team Lead, the QA Team Lead, the
Information Management Team Lead, and the Safety Team Lead each are responsible for reporting to this
group on a monthly basis.
In addition, GLNPO will submit formal reports of survey activities and findings concerning the Water
Quality Surveys as part of GLNPO's annual reporting under the Government Performance Results Act.
19.0 Data Verification and Validation
Data verification activities are performed by GLAS contract managers and the GLAS QC Coordinator
before data are submitted to GLNPO. These verification activities include a detailed review of results to
identify QC measures that failed to meet the objectives outlined in Section 11 of this QA. Measures
failing to meet these objectives will be reviewed carefully to identify the cause of the problem and
determine if reanalysis is warranted. In the event that reanalyses is determined to be infeasible (due to
expired holding times, lack of sample volume, etc), or not warranted (QC failure is not expected to
compromise survey results), the data are submitted with QC flags to indicate the nature of the failure.
Data validation activities are conducted by GLNPO's Technical Lead of each functional area. These data
validation activities are conducted using best professional judgement. This judgement reflects in-depth
expertise and knowledge of the characteristics of each lake.
20.0 References
[1] EPA Requirements for Quality Assurance Project Plans, EPA QA/R-5, EPA Quality Assurance Division,
March 2001.
[2] EPA Guidance for Quality Assurance Project Plans, EPA QA/G-5, EPA Office of Research and
Development, Washington, D.C. 20460. EPA/600/R-98/018. February 1998.
[3] Kwiatkowski, 1980. Regionalization of the Upper Great Lakes with Respect to Surveillance
Eutrophication Data. J. Great Lakes Res 6(l):38-46.
[4] Lesht, B.M., 1984. Lake Michigan Eutrophication Model: Calibration, Sensitivity, andFive-Yar Hindcast
Analysis. Report ANL/ER-84-3, Environmental Research Division, Argonne National Laboratory,
Argonne, IL.
[5] Moll et al, 1985. Lake Huron Lntensive Survey 1980. Special Report No. 110. Great Lakes Research
Division, University of Michigan, Ann Arbor, MI.
[6] El-Shaarawi, 1984. Statistical Assessment of the Great Lakes Surveillance Program, 1966-1981. National
Water Research Institute, Inland Waters Directorate, Scientific Series No. 136.
Page 66
March 2003
------- |