GUIDELINE  SERIES
           OAQPS NO.
1.2-020
     GUIDANCE FOR DECENTRALIZATION AND
     "CONTINUED OPERATION OF THE NASN

             (DRAFT)
                    one
   US. ENVIRONMENTAL PROTECTION AGENCY
     Office of Air Quality Planning and Standards



      Research Triangle Park, North Carolina

-------
GUIDANCE FOR DECENTRALIZATION AND CONTINUED
           OPERATION OF  THE NASN
               September  1974
            OAQPS Number  1.2-020
I
I
I
I
I
I
I
I
I
I
I
I
I
•                       Monitoring and Data Analysis  Division
                     Office of Air Quality Planning  and  Standards
•                                        and
                Quality Assurance and Environmental  Monitoring Laboratory
•                        National Environmental Research Center
                          Research Triangle Park, North  Carolina
•                                      Environment.-! Pretest ion Agency
                                        Region V0 Library
                                        f23Q South Dearborn Street
                                        Chicago9 Illinois
I
                                                  DRAFT

-------
ENVIRONMENTAL PROTECTION AGENCY

-------
I
I
I
I
I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
    X
                                 TABLE OF CONTENTS



                                                                      PAGE
               INTRODUCTION                                             +


               QUESTIONS CONCERNING  THE NASN  NETWORK      •              *


               QUESTIONS CONCERNING  DATA                                4


•              QUESTIONS CONCERNING  TOTAL  SUSPENDED PARTICULATES        6


               QUESTIONS CONCERNING  SULFUR DIOXIDE  (OR N©2)             7


m   '           APPENDIX A  Quality Assurance  Performance Audit Procedures


               APPENDIX B  Stations Which  Are Designated As Critical
                          To Retain Indefinitely

-------
I
I
I
I
I
I
I
I
I
 I
 I
 I
 I
 I
  I
  1
  I
  I
  I
I.  INTRODUCTION
    The Monitoring and Data Analysis Division, OAQPS in co-
operation with the Quality Assurance and Environmental
Monitoring Laboratory of the NERC-RTP, has prepared this
guideline document concerning the operation of the KASN
network.                .

    The purpose of this guideline is to assure uniformity
of the RO's in the operation and continuation of the network
and to allow a number of important uses of the data to con-
tinue on an uninterrupted basis.

    The continued operation of this network is vital to
EPA since the data are used in many decision making areas
such as the development of air quality and emission standards
Further, it is our only data source for the establishment of
long term trends in S02 and TSP as well as non-criteria
pollutants such as sulfates, nitrates, BaP, etc.  The hi-vol
filters provide a mean for retrospective analysis of parti-
culate pollutants such as the ten year lead study and for
determination of long term concentration on new particulate
pollutants which may become of interest to the agency.  In
addition it provides the major source of nitrogen dioxide
data during a period for which no reference method for NO2
currently exists.

    Because many questions have been raised by the Regional
Offices over the last few months concerning the operation
of the NASN, the guideline is structured to pose and answer
the majority of these questions.


II.  QUESTIONS CONCERNING THE NASN NETWORK
    Q.  Can the NASN network be decentralized to the States?

    A.  It is our desire to have the Regional Office maintain
the responsibility for operation of the network.  However, if
resources are not available and if in the judgement of the
Regional Office the state monitoring program is capable and
has demonstrated adequate performance, then the network may
be decentralized.

-------
    The following conditions would apply if the RO turned
over the NASN operation to the State.

        1.  If the station was not part of the SIP network,
then the RO should obtain an agreement from the state to include
it in their SIP network and to be responsible for any necessary
maintenance.  The RO should provide grant funds for this pur-
pose.

        2.  If the state decides at a later point in time
not to analyze the samples from a station, the Regional
Office will have to make ... committment to continue the
operation of the station by performing the necessary
laboratory analyses.

        3.  State agrees to participate in a periodic quality
control check.

        4.  All TSP filters currently being received by QAEML
will continue to be sent by the state.

    Q.  How can the state monitoring program demonstrate
adequate performance?

    A.  The state must pass independent audit checks as
described in the Guidelines for Development of a Quality
Assurance Program Series EPA-R4-73-028 and summarized in
Appendix A.

    Q.  What should the RO be prepared to do to effectively
achieve decentralization of the NASN network?

    A.  1.  The RO should be capable of performing inde-
pendent audit checks  (Appendix A).

        2.  The RO should be prepared to devote approxi-
mately two man years of effort toward the smooth decentra-
lization of the NASN network to the'states.  This effort
would be in the form of guidance, training and audits
similar to that done by QAEML when they decentralized the
network to the RO's.

    Q.  What guidance is available to the RO to give to the
states?

    A.  1.  The series "Guidelines for Development of a
Quality Assurance Program," EPA-R4-73-028-a,b,c and d is

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
method specific for the criteria pollutant, of S02, CO,
TSP and Oxidants using the reference method.

        2.  "Quality. Control Practices in Processing Air
Pollution Samples" APTD-1132 provides general guidance for
minimizing sources of variation inherent in all analytical
functions and to evaluate and document the ability of an
analytical system or person to produce results which are
valid within predetermined acceptance limits.

        3.  The Air Pollution Training Institute is avail-
able, upon request, to provide training both at NERC-RTP
and the Regional Offices for Regional Office and state
and local personnel.

        4.  QAEML, quality control activities are also avail-
able to provide limited assistance to the states upon request
by a RO.

    Q.  Which NASN stations should be retained in operation
and for how long?

    A.  1.  All sites should remain in operation through
CY 1975.

        2.  After 1975, although it is desirable to maintain
all stations, we have identified in Appendix B, a list of
stations which we feel are critical to retain indefinitely.
This list consists of all nonurban stations and urban stations
which have a long history of data.

    Q.  Can any of the NASN stations be moved?

    A.  In general, the NASN stations should not be moved
from their present location.  However, the station may be
moved for any of the reasons listed,below.  The Regional
Office should inspect the site and submit a report to the
QAEML and MDAD for concurrence before any changes are made.

             a.  razing of supporting structure

             b.  sampling location has become inaccessible

             c.  restrictions to air flow due to new buildings
                 in area of sampling inlet

             d.  undue or unreasonable influence of emission
                 sources in the immediate vicinity

-------
                            4

             e.  repeated vandalism

             f.  extensive urban renewal project in imme-
                 diate area (station may be moved back
                 when urban project is completed).

    If a station is moved, a new SAROAD site description
form should be submitted to NADB.

    Q.  Can state sites be co-located with existing NASN
sites?

    A.  If the Regional Office is responsible for the net-
work, duplicate sites would be legitimate for quality con-
trol purposes for short time periods (on the order of 3 to
9 months).  After appropriate quality control checks have
been made, the NASN equipment could be used to check other
state sites or be pulled back to Regional Office equipment
pool for use in special studies.

        If the state has taken over the NASN, then the state
should move their equipment and use it for another SIP moni-
toring station.

    Q.  Can the State substitute one of their existing sites
for an existing NASN site?

    A.  No.  The purpose of the network is to provide infor-
mation gathered at particular locations for an extended time
period.  By switching or substituting sites, this time history
of data would be lost and the site would be like any other
in the SIP network.

        The only exception to this would be if the state could
demonstrate over a two year period that the proposed replace-
ment sites exhibits the same kind of air quality trends
and absolute concentrations.

    Q.  Can the state substitute their own equipment for
Federal equipment at an existing NASN site?

    A.  Yes, if the state uses EPA procedures and methods
as described in April 30, 1971 Federal Register.


III.  QUESTIONS CONCERNING DATA
    Q.  How frequently and where should air quality data
be submitted?

-------
I
I
I
I
I
I
I
I
I
I
I
I
 I
 I
 I
 I
 I
 I
 I
    A.  All data should be sent directly to NADB no later than
45 days after the close of the calendar quarter.  We encourage
sending data on a shorter time schedule if practicable.

    Q.  Should any other information be submitted along
with the data?

    A.  Yes.  The Quality Assurance Performance Audit forms
for the appropriate pollutants should be submitted on forms
on a quarterly basis directly to the Quality Assurance Staff,
QAEML.  For your convenience, copies of these forms are
included in Appendix A.

    Q.  Are new SAROAD Site Forms required?

    A.  If the analysis will be performed by the RO, each
site must be redefined with the Supporting Agency listed as
Regional Office.  This requires a new site form, not a revised
one, since the Supporting Agency Code will change.  If the
analysis will be performed by a State or local Agency and a
site already exists with the proper Supporting Agency Code,
then no new site form should be submitted.

    Q.  What will the new Supporting Agency Code be if the
Regional Office is doing the analysis?

    A.  If the Regional Office is conducting the pollutant
analysis, the Supporting Agency Code will be a "P."  The "P"
site will have to be defined on the site file and all data
should carry the "P" code.

    Q.  When should the new data start using the "P" Agency
Code?

    A.  Arrangements were originally made with QAEML for
all 1973 KASN data to be coded with the old "A" code and sent
directly to QAEML.  The 1973 data was subsequently sent to
NADB.  Beginning with 1974 data, all data should be coded
with the new Agency Code.

    Q.  Why should all 1973 use the "A" Agency Code even if
decentralization had already taken place?

    A.  First, many Regional Offices were not able to complete
the decentralization until sometime into 1973 and it was
felt that one consistent starting date would be appropriate.
Secondly, data split between Agency Codes in the middle  of a
calendar year could never satisfy the annual summary criteria
and therefore annual statistics could never be calculated on
the 1973 data.

-------
    Q.  What if the monitoring was decentralized to the
State or local level?  Should the 1973 data have been
submitted to QAEML as "A" data?

    A.  Yes, because QAEML, in effect, was still responsible
for the laboratory analysis until the end of 1973 even though
another agency was actually performing the laboratory work.

    Q.  What should be done if the above instructions were
not followed?

    A.  Contact the National Air Data Branch.


IV.  QUESTIONS CONCERNING TOTAL SUSPENDED PARTICULATE
    Q.  How often and where should' the TSP-hi-vol filters
be submitted?

    A.  The filters should be submitted monthly to the
Quality Assurance and Environmental Monitoring Laboratory
at the NERC-RTP.

        If the state has taken over the network and the
sampling is being conducted on a six day schedule (as per
the requirements of the August 14, 1971, Federal Register),
then every other filter should be returned.  A minimum of
26 filters per year should be submitted, collected to insure
uniform seasonal coverage.  However, the states should be
encouraged to return all filters, if they have no plans
for trace material analyses.

    Q.  What portion of the TSP hi-vol filter should be
returned to QAEML?

    A.  The entire filter should be returned.  The routine
nonurban analysis requires over two-thirds of the filter.
It is necessary to retain a portion of the filter in the
filter bank for retrospective analysis, for analysis of new
pollutants or for other special studies.

    Q.  Can a state use their own filters instead of those
provided by QAEML?

    A.  No.  In order to be able to analyze the hi-vol filter
samples for trace metals etc., it is necessary to know the
trace constituents in a blank filter.  If the states used
their own filters, this information would be unknown and
thus the emalysis would be biased.

-------

    Q.  Should any speciaj. precautions i>& Lukeji wiLli the
hi-vol filters before submitting it to QAEML?

    A.  The laboratory doing the analysis should obtain
the total weight as soon as practical (after equilibrating
for 24-hours at 50% relative humidity) to reduce the loss
of organic material.

    Q.  How often should hi-vol samplers be calibrated?

    A.  Samplers should be calibrated when first purchased,
after maintenance, any time the flow rate measuring device
has been replaced or repaired, or any time-a one point cali-
bration check deviates more than + 6 percent from the cali-
bration curve.  (See Quality Assurance Guideline EPA-R4-73-028b
for more details).

    Q.  How soon will trace material analyses be available
after receipt of hi-vol filters?

    A.  QAEML will have the data from trace metal analyses
available within five months after receipt of the filters.
Those pollutants for which quarterly composites are run,
will be available five months after the receipt of one
quarter's filters.  The data will be entered into NADB two
weeks after validation by QAEML and can be accessed by the
RO at that time.
V.  QUESTIONS CONCERNING SULFUR DIOXIDE  (OR NITROGEN DIOXIDE)


    Q.  Is QAEML doing any analysis of NASN sulfur dioxide
samples?

    A.  No.  The decentralization of the operation of the
NASN to the Regional Offices included the analysis of the
bubblers.

    Q.  May the Regional Office give the states the responsi-
bility to do the SO2 analysis as well as the operation of
the NASN? -

    A.  Yes.  However, the RO must be assured of the capa-
bility of the state to produce valid data.  This assurance
would come from applying the independent audit checks as
described in the "Guidelines for Development of a Quality
Assurance Program" EPA-R4-73-028d and summarized in Appendix A.

-------
                            ^>*

    Q.  Are there any special precautions to be taken in
the sampling and analysis of S02?

    A.  1.  General quality control considerations are des-
cribed in previously mentioned guideline APTD-1132 and a
step-by-step operations manual and quality assurance program
are described in EPA-R4-73-028d.

        2.  While it would be desirable for a single laboratory
to perform all the bubbler analysis for the entire region,
a few precautions will go a long way in ensuring uniformity of
the analysis.  These are described as follows.

             a.  Adhere to a rigid sampling schedule so
the absorbing reagent is the same age for all samples.

             b.  Save a portion of the absorbing reagent
to be used as a blank for the subsequent analysis.  Do not
use fresh absorbing reagent as a blank for past samples.
The same applies for constructing calibration curves as well.

             c.  The analysis of the SC>2 samples should be
structured so that there will be a uniform amount of elapsed
time between sampling and analysis, preferably the analysis
should occur as soon as possible after sampling.

             d.  Where more than one laboratory performs
the analyses, periodic interlaboratory comparisons should
be made using standard samples.

             e.  Since SC>2 samples have a daily decay rate
which varies from season to season,  (see Federal Register,
April 30, 1971) the sample should be refrigerated if there
is undue delay on the order of 5 to 7 days between sampling
and analysis.  If SC>2 samples are not refrigerated, a daily
decay rate established on a seasonal basis should be used to
correct the analyses.

-------
1
I
I
I
I
                                    APPENDIX A

-------
I
I
                             QUALITY ASSURANCE PERFORMANCE AUDITS FOR NASN NETWORK
I

I

I
 I
 I

 I
                                       (Independent Duplicate Checks)



               Introduction               •

                    To attain quality assurance for any measurement system, a number of

               activities need to be considered with respect to quality.  These activities

               include but are not to be limited to:
                          1.   procurement of equipment and materials

 •                        2.   training of field operators and laboratory analysts

                          3.   documentation and control of sampling, calibration and

 I
                              analytical procedures

                         4.   preventative maintenance of equipment

                         5.   control of the design of tiie equipment to assure uniformity

•          .                  of the "hardware" and records of any changes

                         6.   adequate records of equipment failures and major repairs

fl                       7.   procurement and care of appropriate calibration standards

m                       8.   adequate record keeping (using standard forms for recording

                              raw data, performing computations and reporting results)

•                       9.   Performance of special checks and adjustments by field

                              operators and laboratory analysts during the sampling,

|                            handling, and transport of samples, calibration and

*|                            analysis of samples (intralab quality control)

                        10.   review of measurement data for completeness and

•                            reasonableness (data validation)

-------
                                      -2-
          11.  performance of quality problem investigations (trouble-




               shooting)




        •  12.  Conductance of independent performance audits




          13.  participation in interlahoratory comparison testing




               programs




     Many of the above activities are prevention and correction aspects




of quality assurance—prevention, with  the objective of preventing bad




data from occurring, and  correction,  with the objective of taking




appropriate corrective action when deficiencies or bad data are identjfied.




Items 1 through 11 are aiued at assuring both the precision, or repeatability,




and the accuracy of the measurement" system.  Item 12 above, the conductance




of independent performance audits is  one of fh^ key activities in assessing




or quantitatively measuring the quality of data generated from the




measurement system.  Item 13 above—the performance of interlaboratory




comparisons is particularly aimed at  assessing the accuracy (or relative




bias) of participating laboratories.




     The conductance of independent performance audits is primarily a check




on accuracy if different  calibration  <=rnnHards arp used, and primarily as




a check on precision if the independent audits involve most of the elements




in common with the original measurements.  Although it is not always




practical, the objective of the independent performance audit is to attain




as complete an independent check as possible, thereby providing a




quantitative afssessraent of the variability in the measurement involved.




Where perl"ornnnce audits involve the total measurement system, the agreement

-------
I
I
«l





•
               between the original checks and the independent checks, provide a basis


               for statistical confidence limits which can be placed on the individual


      .         reported data from the measurement system.  Further, if lack of expected


               agreement is obtained, corrective action is indicated 'so that the measure-


               mcnt system can be corrected to bring it within expected control.  And if


               the data obtained since the last acceptable agreement cannot be corrected,


f             the validity of the data may be questioned.


_                  The independent performance audit is therefore considered as a


               management tool (1) to assess quantitatively how good the measurement and


•             the resulting data are, and (2) to indicate when the system is in need of


               correction.


I
               Scope of Pen o rmance Audi ts for NASN


•      .            EPA guideline documents relating to the NASN network methods are:


                         EPA-R4-73-028b for High Volume Method and

      f,
                         EPA-R4-73-028d for Sulfur Dioxide Method.


M       .      The guidelines for sulfur dioxide can be used as a general guideline for


               the nitrogen dioxide because of the similarities of these two bubbler methods,


•                  Although the guidelines outline a number of items  for performance


               auditing,  only three are included on the forms for reporting the NASN results


I             to EPA.   These were selected to include (1) a check involving the sampling


g|             part of  the  system, (2) a  check involving the analytical part of the system,


               and (3)  an "alternate" check involving the total measurement system.
I

I

I

-------
                                      -4-
            Sampling
                         Independent Auditing Program
Analysis
Total System
(Alternate)
Hi-Vol      Flow rate calibration

SO          Flow rate calibration


NO,,         Flow rate calibration
Exposed filter weight

Measurement of refer-
ence samples

Measurement of refer-
ence samples
Dual system

Dual sy.str;?.


Dual system
     Although the use of the dual system is not suggested by the guideline

for SO ,  it is desirable that such a check be established as an alternate

auditing procedure for SO  and for NO  measurements for consistency with

the hi-vol, and because it provides a siiaple way of checking the total

measurement system.  Similarly, other procedures and practices recommended

by the guidelines should be followed.  It is emphasized that quality

assurance practices, including the independent performance audits,  are

primarily for the use of the local agencies in maintaining, assessing and

correcting if necessary of measurement systems to assure production of

valid (accurate and precise) data.  The performance audits for the NASN

operations delineated herein suffice only to provide EPA with a means of

quantitatively assessing the quality of the data in the National Air Data

Bank.

     It is not suggested that the performance auditing activities of each

agency be limited to those required for NASN reporting.  It wouJd be to

each agencies benefit to perform as r.any of the performance audits outlined

in the guidelines as resources of the agency will permit.

-------
I
I
I
I
                                                     -5-
               Indcpendence of Performance Audi ts


                    Performance auditing is an independent check performed by an


M             auditor. . The auditing should not be done by the regular field operator


               or laboratory analyst who performs the original routine measurement or
I

 I

 I

 I

 I

 I

 I

 I

 I

 I

 I

 I

 I

 I
               analysis.   The auditor should be a person at least equally capable as


               the regular operator or analyst and should be at least as familiar with


               the details of the procedures for performing the measurements or analysis.


               Ideally,  the auditor would be a more senior and experienced operator or


               analyst,  or a supervisor.   Where such an individual cannot be assigned


               as an auditor, the responsibilities of the auditor could be rotated among


               existing  operators or analysts, and where there is only one operator-analyst,


               the p'-'pervT «=or ^hovld perform the civili*r function or train another individual


               to perform the audits.


                    Whenever practicable,  the equipment,  supplies, calibration materials,


               etc.  used  by the auditor should not be the same as those used by the


               routine operator or analyst.   For example, in checking the hi-vols, the


               flow meter used by the  auditor should not  be same one  as used by the


               operator,  and the auditor's  flow meter should have been calibrated at a


               different  time and desirably  with a different standard (e.g.,  wet test meter)


               than  th.it  used for calibrating the operator's flow meter.   Similarly,


               when  auditing the NO^ method,  the reference sample prepared by the auditor


               should  be  based on a calibration curve separately obtained by the auditor


               using a different reagent  batch than that  used  by the  regular analyst.   In


               the case of the "alternate" dual system arrangement the spare equipment,


               materials,  and checking  devices should be  independently maintained by the auditor.

-------
                                      -6-
I
£
i
 i
I
     In short, the work by the auditor should be as independent as possible         •


from that of the regular operator or analyst.  In other words, the


differences in equipment, materials, reagent, calibrations and other                V
                                                                                      |

variables associated with time should reflect as much as possible the               M


differences which would or might normally be encountered in the measurement


process from time to time for different samples measured at the facility            I

                                                                                      i
involved.  This is desirable so that confidence limits placed on a given


measured value are realistic in reflecting as many of the normally                  w
                                                                                      i

encountered variables in the system as practicable.                                 fl|



Defect Criteria                                                     .                  *


     The defect criteria given in the guidelines were based upon the best           ™
                                                                                      i
available technical, design and statistical information at that time.  It


will be noted, however, that the data forms for each of the methods do not


list the defect criteria in the guidelines.  An analysis will be made by             •


the Quality Assurance Staff at Research Triangle Park of an initial body


of actual operational data to establish realistic defect criteria for


future use.
I
  »

I


I
Frequency of Performance Audits


     The guideline documents generally recommend seven audits for each


J.OO measurements to attain a given statistical confidence.  Although this            •


sample size is desirable, there are likely situations where these                     ;


frequencies may 'be impractical from equipment, cost and time standpoints.            j§
                                                                                      t




                                                                                     I

-------
I
I
I
                                                   -7-
«|                For items checked once each 6 days, as for monitoring for state
             implementation plans, the guidelines suggest a sample size of 3 (one per
•   '        month) from a total number of 15 monitoring dates.  This corresponds to
             3/15 or 20%.  Because the monitoring frequency for the NASN network is
             one analysis each 12 days, only about 8 analyses vould be made each
             quarter for each site.  To provide consistency with the guidelines it. is
             suggested that for NASN sites one or two samples be selected each quarter;
             with a total of three checks out of the total of 15 for each 6 months.

             Randomization of Audit Checks
I

I

I

I
                  Performance audit checks should be made on a random basis with respect
•           to the item or site checked.  It is not necessary, however, that the checks
             for sampling and the checks for analysis coincide for the same date and
W      •     site combinations.
                  §The purpose of performance auditing is to check or measure the
     x
             variability in the agencies system.  Even though the decentralization applies
•      "     only to the NASN sites, the sampling applies to the total system of the
             agencies including all other sites.  As a result, in the sample randomization
V           process each NASN sample would be given an equal chance of being selected
             as any other sample.
             Corrective Action
I

*                When comparing  the  original  and  the  auditor's  results,  excessive
•           differences  may  be encountered,  in which  case  the  local  agency may desire
             to  take  corrective action.   The  nature  of the  corrective actions  will vary
P           depend ing upon  the Item  chocked  and  the magnitude  of  the differences.  Althoi.-yh,

I

-------
                                      -8-
thc guidelines provide some direction for trouble-shooting, those routinely




working with tho measurement systems will generally know the most potentially




fruitful avenues to explore.  And although corrective actions should be




taken to prevent future excessive differences, some discrepancies such as




data processing errors are immediately correctable.  It should not be




considered sufficient however just to correct the data involved for one




particular audit.  Corrective action should be pursued to correct the




system against future like occurrences.   Such corrective actions as




additional training, procurement of better equipment, more frequent




calibration, more frequent prcventative  maintenance, more specific




instructions or procedures, more personnel assigned to a task, replacement




cf defective, pcrtr of cq-iprp.cr. t 2re cnly ccrac of the types of corrective




action that should be taken.  Records should be made of the nature, scope,




and dates of corrective actions taken.




     The audit is only a sample and as such when an excessive difference is




encountered for a given apparatus or equipment, a given operator or analyst,




or a given monitoring site, it is an indicator -of a possible problem which




needs investigation and possible corrective action.







Reco rd-Kcc*ning and Reporting




     It is important that complete records be k-ept of audits.  Such




records should include the names of the  auditors, the dates and locations




of the audits, and other identification  information such as the serial




numbers of equipment used and the container number of reagents used.  Records




concerning the audits should be as complete, if not more so, than those for




the- original rnensurt-nients.

-------
I
I
I
I
I
I
I
I
                                                    — 9 —
^                  It  is  suggested  that  a separate  systematic  file  or  logbook  be


™             maintained  to  keep  such  data  for  the  audit  check as well as  for  the


M             original measurements.   The individual  reported  measurements  for .the
              original and the audit check should be recorded along with the signed


              difference and the percentage difference.


                   It is important that the results of all audits be reported  to


              EPA on the data forrr,3 provided in order to properly assess the quality


              of data from the various sources and to perform analyses of the  data
I
               to  arrive  at  criteria  for  expected agreement of audit  results.



               Detailed Procedures  and Reporting Formats  for Hi-Vol,.S00  and NO,
                                                  __^..  ,           j=-,12

               •Measurement Method.-j
                   The  following procedures are essentially an  integration  of  the


               information on performance aduits for  the NASN contained  in the  guideline


               documents EPA-R4-73~028(b) and  (d)  for hi-vol and SO  ,  respectively.   This
I

I
w            integration of the material was accomplished to provide all of the
•            information concerning the performance audits in a separate single
              integrated document  in a logical order for ease and facility  in  use  and


              reference.  Even so,  references to pages of  the original guideline


              document were necessary for completeness.  The procedure for  the  alternate


              dual systems for NO.,  and S00 is simply to independently install,  calibrate


              and service a spare  instrument with a separate take-off from  the  common


              Dianifold, to independetly calibrate the colorimeter and indepcntly


              analyze the absorbant solution.


                   The reporting formats which follow each of the procedures, are
™     •       ledesigns of the forms contained in tho guideline documents.  These redesign'.


I

-------
                                     -10-
were accomplished to obtain more reporting detail for identification




and subsequent data analysis purposes.  It is likely that the forms will




be simplified at a later date after analyses of the distributional




behavior of the data have been accomplished.




     The completed forms for performance audits should be submitted each




quarter along with the results of the NASN data to the EPA Regional Office.




Initially, the performance audit data should then-be submitted to the




Quality Assurance Staff, QAEML, for review and analysis.  Appropriate




analysis of the data will result in:




          1)   establishment of realistic defect criteria for the




               performance audit data




          2^   £!*"*"rc~ric.tc rcc?cslr""i cf *^2*"H f^^^s  3i^d




          3)   appropriate routivie analysis of performance audit data.




     Subsequently, the performance audit data will be routinely submitted




from the Regions directly to OAQPS for routine analysis and reporting.

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
           I*Fron r.lV--R4-73-028h, "Guidelines  for Development  of  a  Quality  Assurance
            Progrnr, Ref^runoe Method for  the Determination of Suspended ['articulate;
            in the Auosphere  (High Voluruo Method)."  June 1973.
I
I
           QUALITY ASSURANCE PERFORMANCE AUDIT FOR HI-VOL FOR NASN*




INDEPENDENT CI1CCKS FOR AUDITING PURPOSES



1.    Wei ^ 11 i ng_Ch_ec_k'> °f Exposed Filters


          To check t'ae total weighing system, rather than just checking the

     sensitivity of the weighing instrument as might he done using weights,


     weighing checks of the filters should no^ be made immediately before

     or after the regular weighing.  A time lapse such as between morning


     and  afternoon or from one day to the next should be introduced.  The


     filters to be checked should not be removed from the conditioning


     environment.


          The check should be independent;, i.e., performed by a person


     other  than the one doing the regular weighings.

          Because of loss of volatiles after the 24-hour conditioning

     period, check weights made as late as the next day might show a

     downward bias.  If such weight losses occur, it is necessary to

     perform the audit as the filters are weighed.

          a)   Randomly select and reweigh the sample of exposed filters.

          b)   Record original and check weights in the audit logbook.

-------
                                     -2-
2.   Flow Rate Calibration Check
          c)    Convert rotameter reading to flow rate,  Q ,  using the
3.    Alternate Check
     as a reerence.
operated in accordance with the existing procedures applicable to
                                                                               I
                                                                               I
                                                                               I
                                                                               I
          Independent calibration checks should be made  en-site.   Portable


     calibration-equipment is used.   Perform calibration checks according            •


     to the following procedure.                    •                                  _


          a)    Set-up equipment.                  .


          b)    Select one of the  resistance plates and obtain the   •                 B


               actual flow rate,  Q ,  and the rotameter reading, following


               the calibration procedures given on page  30,  Section 2.2              •


               of EPA-R4-73-028(b).
                                                                                     I
               calibration curve and making any standard corrections for             M


               ambient temperature and pressure.


          d)    Record the reference rate,  Q ,  and the sampled rate,  Q ,  in           •
                                           cl                         I"               ^w

               the audit logbook.
     Assessment by Auditing with a Mobile Sampler


          An alternate method of auditing the High Volume Method, which


     in certain situations might be feasible, is to use a mobile sampler


                                                                                     •
                                                                                I

                                                                                I

                                                                                I
          A high volume sampler equipped with a continuous flow rate


     recorder,  a constant voltage regulator,  elapsed time indicator, and             ||


     a constant flow regulator is maintained  by the office of the laboratory         «


     director and used as a reference.   The reference sampler should be
                                                                                     •



                                                                                     I

-------
 I
 I
 I
—3—
 •                other samplers.   A record should be maintained of the checks performed
                  on the reference sampler.  An audit consists of placing the reference
 •                sampler adjacent to (but no closer than 3 feet) the field sampler
 _                and sampling simultaneously for a sample period.   The reference
 *                sampler should be oriented so that the long (10") dimension of the
 V                filter is in the same-direction as that for the routine sampler.
                       After determination of the particulate mass  for the reference
 I                sampler,  record  the results for the regular sampler and the
 —                reference sampler in the audit logbook.


 I

 I

 I

 I

 I

 I

 I

 I

 I

I

-------
1
1
1

1



1
1
^v
1

1



1

1

•




1

1




1
1


QUALITY ASSURANCE i'F.KfCI'.MANCE AUDIT FOR SULFUR DIOXIDE AND
MTKCSLN DIOXIDE FOR NASN*
•
IKDI:PJ:;;DI:?;T C;::TKS FOR AUDITING PURPOSES

1. Flo*- n.-.te C.ilibrcu'cn Chock
For 24-hour samples a flow rate check is recocr.p.ided as a
means of auditing the sample collection phase of the measurement
process. The check is performed as follows.
a) The regular operator prepares the sampler for sample
collection as usual.
b) The individual performing the audit inserts a calibrated
rotarneter in the sample inlet line.

c) With the calibrated rotamet^r in place the sample is
collected in the usual manner.
d) The individual performing the audit reads the calibrated
rotameter before and after the sampling period.
e) Convert the rotaneter readings to flow rates using the

calibration curve making corrections for ambient
..
*•
f) Compute the average of the initial and final flow rates.
g) Record the operators and the auditors average flow rates
.in the audit logbook.

*Frr>-\ r:'A-K'.-7'i- ^:\i, "r-iid'-l incs for Dovnlop-. or, t of a Cirility Assurance
Pi. -.I."1, .' •:••:> • " ':.,,: tcr t l.c .';-. '^i- i::ili..'n of Sulfur Dioxide ii.
t--.1' '••!' -•-..-!-»> .' ....... r. Jv!'3. I'c.r N,r:..,;^a Di.jMje, subs t i Lute "Ni i i\ ^
i:\ {-,];u-i; c:" "jul.'ur" ,i;t : tVc-r it appo.-;r:-i .



-------
                                '-2-
Measurcn:.•>nt of Ro fo rcnee Sar:p 1 os




     Measurement of a reference sample prepared independently of




normal operations can be used Lo evaluate the precision and




accuracy of the analysis phase of the measurement process.  A




reference- sample as used here refers to a sample prepared by an




individual other than the regular operator using reagents prepared




independent of those used for normal operations and used for auditing.




purposes.   A control sample as referred to in the operating procedures




implies a solution prepared by the regular operator from the normally




used reagents and used periodically to verify that the analysis




process ir; under control.




     The procedure for performing the check is given below.  The




frequency of performing the check will be specified by the




supervisor.




     a)   Prepare a reference sample with a known concentration




          in the sar.e manner as was used when developing a




          calibration curve.  The concentration should be varied




          over the range of normally measured values from audit to audit,




     b)   Have the regular operator measure the reference sample.




          The operator must not know the true concentration of the




          reference sample and preferably not know that it is a




          reference sample.

-------
I
I
I
-3-
                        c)   Obtain the operator's  finding (jjg SO )   and the
 •              •             true concentration  (ug  SO  )   ol  the saraple.

                        d)   Record  (ug SCO  and  (ji'g SO-)   in the audit logbook.



 I


 I

 I


 I


 I

 I


 I

 I


 I

 I


 I

 I


 I


 I

-------
S3
W
O
o
w
H
O
 3
 O
 .C
 I
        E-i
     W M
     O Q
W   ^ S
S   D C^
                             n
                             CNl

                             .•I
                             O
                             CM
                 01
                 •o
2;


O
Measure
r— 1
cn
•H
to
•H
Ul
O
v7
f-ti
r^
n
f-i
n
o
<~i
.1.J4J
-fH-H
— >- •— * — i— t
i r i j
• 1 ,' .-, ^~?)
                                                                 .J   _...,_i_;_J  L._i_L.U
                                                                 I ^4   -, 1^1, -t-t rt  I .-. nn . ,   (.
                                                     "   1 ' !  'i - j  '1  ['_(. -I.J  V,]  I 'I^f  f '•[
I
I
I
I
I
I
I
I
 I
 I
 I
 I
 I
 I
  I
  I
  I
  I
  I

-------
I
I
I
I
I
I
I
I
I
I
I
I
 I
 I
 I
 I
 I
 I
£S
W
O
o
M
E-i
O
g
 CO
 c
W H

P
co W ^
s§|

si!
3le
a M
sp
     o
     H
     VJ
     o
     cv,
od
           ra I r i
           1-11—
           to H
           o0
                                                  C3
                                                  3
                                                  Q
                                                      re
                                                      a
                                                      a)
                                                      a
                                                      3
                                                      CO
                                                n

-------
                               •o
                                'o
                               Si
                                u

                               $
                                01
                               •o
                                o
                               u

                                V4
                                a
                                4J
                                0)


                                I
                                H)
                               Pi
                                                                           Sl'.il
I
1
T





Ln|u*i m




!/l




Li

                                                                                                                                                                 til
                                                                                                                                                                 C3
§
H
s
H
i
W  H
O  l-l


11
!=>
w  W
p lil
O- A<
    no
    q
   •rl





















	 1







' 	 1



...








I 	
















'*
j

                                                                                                                                                       t)
                                                                                                                                                       o
                                                                                                                                                       •o

                                                                                                                                                       4
                                                                                                                                                        CO
                                                                                                                                                        C
                                                                                                                                                       •o
                                                                                                                                                        
-------
I
 i
 I
 i
i

 i
 *

i
 i
 i
i
i
i
i
i
i
i
                                    APPENDIX B
   NASN Stations Which Are
Designated As Critical To Retain
          Indefinitely

-------
I
i
I
1
I
i
I
I
I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
           •r\uyiuii i UP urudii, 4 nori-uroan;
   07  OOGO 001  A01*    Conn., Bridge-port
   07  0420 001  A01*           Hartford
   07  07CO 001  A01*           New Haven
   07  1240 001  A01~           Waterbury
   20  0950 002  A3!      Me., Portland
   22  02 40 CO i  A01*    Mass., Boston
   22  05SO GO?  ADI             Fall River
   22  21 CO 002  A01*           Springfield
   22  2C40 001  A31*           Worcester
   30  0120 001  AC!      N. H., Concord
   41  0120 Co'l  A01      R. I., E.  Providence
   41  C3CO 001  Adi*           Providence
   47  .0140 001  A01      Vt., Burlington

   20  0010 001  A03      Ma., Acadia National Park
   3D  0140 001  A03      N. H., Coos County
   -11  03?0 CZ2  A03      R. I., Washington Co.
   47  0360 CO'i  AG3      Vt., Orange Co.
*Gas bubbler  at this  site.

-------
       IT (20 urban, 1 nor, urban) (11 S02/N0?2)
31 0550 002 A01*    N. J., Burlington Co.  (Karlcto.i)
31 0720 GOT A01*           Camden
31 1300 002 ACT    '        Elizabeth
31 17CO 001 A01*           Glassboro
31 2320 001 A01*           Jersey City
31 3480 C01 A01*   •      •  Newark
31 4140 001 A01*           Paterson
31 4220 001 A01            Perth Amboy
31 5400 001 A01            Trenton
33 0560 001 A01*    N. Y., Buffalo
33 4680 CO! A01*      '     New York City
33 4740 001 A01            Niagara Falls
33 5760 001 A01*           Rochester
33 C620 031 A01            Syracuse
33 6830 001 A01         .   Utica
40 0330.002 A01*    P. R. , Bayamon
40 0560 002 AQ1            Cateno
40 1030 002 A01*           Guayanilla
40 1920 GC2 A01   "         Ponce'
40 2140 001 A01            San Juan

33 3340 001 A03     N. Y., Jefferson  Co.

-------
I
I
 i
I
I
i
i
 -i
i
i
i
i
I
i
i
i
i
i
i
i
Re Mic.n. II! ^28 urbsn, 3 nen-urb?n^
03 0140-001 A01*    Del.,  Newark
09 0020 001 A01     D.  C., Washington
                   /
09 0020 003 A01            Washington
21 0120 001 A01*    I'd., Baltimore
39.0120 001 A01*    Penn., All entov.n
39 0140 001 A01          .  Altoona .
39 07£0 002 A01            Bethleher.i
39 30GO 002 A01            Erie
39 3880 001 A01            Harrisburg
39 3960 001 A01            Hazleton
39 44SO 001 AC 1*      '     Johnstown
39 7140 001 A01*           Philadelphia
39 7200 001 A01*           Pittsburgh
39 7620 001 AQ1*           Reading
39 804 0 031 A01*        .   Scranton
39 9160 001 A01*           V.'arainster
39 9''; 30 001 A01            Wilkes Barre
39 9550 001 A01*  •         York
48 0320 001 A01     Vs. , Danville
48 K40 001 A01          Harr-pton
48 1840 001 A01      -    Lynchb'jrg
n8 2120 001 A01          Kev/ Port  Nev/s
48 2140 001 A01*         Norfolk   •
48 2440 001 A01          Portsmouth
4826 G 0 002 AO 1 *         Ri c hra n d
48 P.700 C01 A01          Roanokc
50 0,20 001 A01*    W.  Va.,  Charleston
bO/i/GO 001 A01              S.  Charleston

39 1760 001 A03     Penn., Clarion  Co.
48 ?^[) GJl A03     Va., Shcrundor.h  N'itional  Park
48 3',-'0 001 A03          1,'ythc Co.

-------
 01  1450 001  AQ1      Ala.,  Gadsden
 01  I860 001  A01            Huntsville
 01  2460 001  A01*   '       Montgomery
 10  19GO 002  ;,01      Fla.,  Jacksonville
 10  2700 002  A01*          Miami
 10  39SO 002  A01*   •       St.  Petersburg
 10  4360 002  A01~          Tampa
 11  0200 001  A01*    Ga., Atlanta
 11  1230 001  A01*         Columbus
 11  4500 001  A01*         Savannah
 18  0030 002  A01      Ky., Ashland
 18  0320 001  A01           Bowling Green
 18  OSOO 001  A01*         Coving ton
"18  2300 001  A01*         Lexington
 18  2380 002  A01*        . Louisville
 341  0700-001  A01      N. C., Charlotte
 34  1160 001  A01             Durham
 34  1740 001  A01*           Greensboro
 34  4460 002  A01             Winston-Saleni
 42  1180 001  A01      S. C., Greenville
 44  0380 001  A01*    Tenn., Chattanooga
 44  1740 002  A01             Knoxville
 44  234o" 001  A01*           Memphis
 44  2540 001  A01*        .   Nashville

 10  1680 001  A03     Fla., Hardee Co.
 34  0590 001  A03     N. C., Cepo Hatteras
 44  OGoO 001  A03     Tenn., Cumberland Co

-------
I
I
I
I
 'i
 i
I
I
t
I
I
I
I
I
I
I
I
I
I
I
Jtegion V  (40 urban, 2 non-urban)  (2
14 1220 C01 A01      111.,  Chicago
14 1220 002 A01*           Chicago
14 5620 002 A01  f          N.  Chicago
14 5080 001 A01            Peoria
14 6700 001 A01            Rock  Island
14 7280 COV AOI            Springfield
15 1180 001 AOI*     Ind.,  E.  Chicago
15 1300 001 A01*           Evansville
15 1380 001 A01            Fort  Wayne
15 1520 001 A01*           Gary
15 1780 001 A01*           Hammond
15 2040 001 A01*           Indianapolis
15 2980 002 A01*           New Albany
15 3880 002 A01*           South Bond
15 '4080 001 A01            Tcrrc Haute -
23 1180 001 A01*     Mich.,  Detroit
23 1580 001 A01*            Flint
23 1820 001 A01*            Grand Rapids
23 2840 00! AOI*            Lansing
23 4860 001 AOI*            Saginav/
23 5120 001 A01             Trenton
24 1040 001 A01      Minn.,  Duluth
24 2260 C01 AOI*            Minneapolis
24 2320 001 AOI             Moorhcad
24 3300 001 AOI             St.  Paul
36 0060 001 AOI*     Ohio,  Akron
36 1000 CO] AOI*           Canton
36 1220 001 A31*           Cincinnati
36 1220 002 AOI*           Cincinnati
36 1300 001 AOI*           Cleveland •
36 1460 001 AOI-           Colu:.;bus
36 1650 GJ] AOI*           Day'  ;i
36 6600 001 AOI*           Toledo
3b 77 60 i/Ji AOI •'•           You ;',r;:. lo'..n

-------
51 0840 002 A01     Wise., Eau Claire
51 1540 001 A01            Kenosha
51 1850 001 A01            Madison
51 2200 001 A01* '          Milwaukee
51 2830 001 A01            Racine
51 3480 001 A01            Superior

15 2800 001 A03     Ind., Konroe Co.
15 32GO 001 A03           Parke Co.

-------
 1
I
i
            Region  VI  (13 urban,  4  non-urban)  (9  S02/H02)
            0-1  1440 001  A01     Ark,  Little Rock
            04  2740 001  A01           W.  Memphis
            19  02CO 001  A01  '   La.,  Baton  Rouge
            19  2020 032  A31*          New Orleans
            19  2740 001  A01     '      Shreyeport
            32  0040 00-1  A3!* '   N.  M.,  Albuquerque
            37  2200 001  A31*    Ok!a.,  Oklahoma  City
            37  3000 031  A01*            Tulsa
            45  1310 002  A31*    Tex., Dallas
            45  1830 00.1  AQ1*    .       Fort  Worth
            45  2560 001  A01*           Houston
            45  4063 002  A01*           Pasadena
            45  4570 031  A01*           San Antonio

            04  1760 001  A33     Ark., Montgomery  Co.
            37  0480 001  A03     Okla.»  Cherokee  Co.
            45  3530 001  A03     Tex., Matagorda  Co.
            45  5203 001  A03 .           Torn Green  Co.

-------
Region VII "(11 urba.,, 2 non-urban) (5 S02/N02)
16 0540 001 AOi     Iowa, Cedar Rapids
16 TOGO 001 AOI           Davenport        • "
16 1180 001 AOI-          Des Koines
17 1800 002 AOI     Kan., Kansas City
17 3560 001 AOI           Topeka
17 3740 00-1 AOI* •      •   Wichita
26 23SO 002 AOI     l-'.o., Kansas City
26 4280 001 AOI*         St. Louis
26 4230 002 AOI*         St. Louis
28 1560 CC2 AOI     -Neb., Lincoln
28 1880 001 AOI*          Omaha

26 4480 002 A03     Mo., Shannon Co.
28 2480 001 AG3     Keb., Thomas Co.

-------
   I
   I
   i
   I
   I
   t
   f
   I
  I
  I
   I
  I
 I
 I
 I
 I
 I
   i
 I
 I
 I
I
I
    UH viii \t uluw..}  ..._...
06 0580 001 A01*      Colo.  Dc-nvcr
35 0100 CO! A01       N.  D.,  Bisi-.cirk
43 1450 001 A01       S.  D.,  Sioux Falls
46 OG80 COT A01   '    Utah,  Cgccn
46 0920 C01 A01*            Sail Lake City
52 0120 C01 A01*      V.'yo.,  Cr.sper
52 0140 001 A01             Cheyenne
                                   •           i
05 1530 002 A02       Colo.,  M?sa Verde National Park •
27 0570 CC1 A03       Mont.,  Glacier national Park
43 0110 CC.l A03     .  S.  D. ,  Black Hills National Forest
52 OS60 COl A03       Wyo.,  Yellowstone national Park

-------
Kegion iA \
-------
I
 I ^


I
 1
I
I
I
 i
 i
i
i
i
 i
i
•j
i
i
i
i
i
i
i
VZ uu~y uuj jVji     Alas.»  Anchorage

13 0220 001 A01     Ida., Boise

38 1460 001 A01    t Ore., Portland

49 18',0 001 A01*    Wash.,  Seattle

49 2040 001 A01             Spokane

49 2140 001 A01             Tacosia


13 0340 001 A03     Ida.,'Buttc  Co.  •

33 0440 001 A03     Ore., Curry  Cu.

49 09SO 002 A03     Wash.,  King  Co.

-------