United States
                 Environmental Protection
                 Agency  	
Atmospheric Research and
Exposure Assessment Laboratory
Research Triangle Park NC 27711
                 Research and Development
EPA/600/S3-89/007 Sept. 1989
                                                                             I V
c/EPA         Project Summary
                  The  National Atmospheric
                  Deposition Program/National
                  Trends Network  (NADP/NTN)
                  Site  Visitation  Program  (October
                  1986  through  September  1987)
                  W. Gary Eaton, Curtis E. Moore, Dan A. Ward, and
                  Richard C. Shores
                  The  proper collection  of  pre-
                 cipitation and the  accurate meas-
                 urement of  its constituents are im-
                 portant steps in attaining a better
                 understanding of the distribution and
                 effects of "acid rain" in the United
                 States. One of NAPAP Task Group
                 IV's major programs concerns wet
                 deposition monitoring.  One of that
                 program's projects, 4A-15,  "Quality
                 Assurance Support  for Wet Deposi-
                 tion Monitoring," is sponsored by
                 EPA to evaluate the sample collection
                 process of the National Atmospheric
                 Deposition Program/National Trends
                 Network (NADP/NTN) precipitation
                 networks through a site visitation
                 program. Research Triangle Institute,
                 as contractor to EPA, conducts these
                 visits. If deficiencies or nonstandard
                 procedures are noted, the  site
                 operator and supervisor are notified.
                 Brief reports are sent to the EPA
                 Principal Investigator and the
                 NADP/NTN Quality Assurance Man-
                 ager. In this way, necessary changes
                 can be made promptly.
                  All NADP/NTN sites were visited in
                 1985-1986. A second round of visits
                 began in October 1986, with the goal
                 of visiting approximately one-third of
                 the  200 sites  each year over the
                 three-year span 1986-1989. This docu-
                 ment is a summary report of the find-
                 Ings from the 1986-1987 (fiscal year
                 1987) site visitation program to 62 of
the sites that comprise the
NADP/NTN precipitation networks,
referred to collectively as the
NADP/NTN network.  In its present
configuration, the NADP/NTN net-
work's research and  monitoring pro-
grams are supported and operated by
the U.S. Geological Survey, State
Agricultural Experiment Stations, the
Departments of Agriculture, Interior,
Commerce, and Energy, and the En-
vironmental Protection Agency. Addi-
tional support is provided  by state
agencies,  public  utilities, and
industry.
  Protocols and procedures followed
in conducting the site visits are de-
scribed. Results of systems and per-
formance audits are discussed for
siting, collection equipment, and the
field support laboratories.
  Where exceptions  are found,  the
potential effects of nonstandard sit-
ing, improperly operating equipment,
and improper sample handling or
analysis technique on the data base
are discussed. Recommendations are
given for improvement and standardi-
zation  of individual  sites  and  the
network as a whole.
  This Profect  Summary was  devel-
oped by EPA's Atmospheric Research
and Exposure Assessment Laboratory,
Research  Triangle  Park, NC, to
announce key findings of the research
project that is fully documented in a

-------
separate report of tfie same title (see
Project Report ordering Information at
back).

Introduction
  This document is the summarizing re-
port of quality assurance assistance and
findings from site  visits  made to the
National  Atmospheric  Deposition Pro-
gram/National  Trends  Network
(NADP/NTN)  precipitation collection sta-
tions in the period October 1986 through
September 1987. Each site is located and
operated  according to protocols  and
procedures as  given in the  siting  and
operating manuals for the networks L2^.4.
The purposes of the site visitation  pro-
gram,  sponsored  by the  U.S.  Environ-
mental Protection Agency,  are  to verify
that each site is operated according to
established procedures and to provide
technical assistance as required.
  Sixty-two of the 201  sites that were in
operation  as  of June  30, 1987,  were
visited during  this time frame.
  The goals of the site  visitation program
for  quality assurance assistance  to  the
NADP/NTN collection sites are to:
1.  Provide a qualitative assessment of
    each site  and  its surroundings, the
    operator's adherence to sample cot-
    lection and analysis procedures, and
    the condition of the site's collection
    and analysis equipment through an
    on-site systems survey;
2.  Provide a quantitative assessment of
    the operation of the precipitation col-
    lector and the accuracy of response
    of field and laboratory measurement
    devices for precipitation depth, mass,
    temperature, conductivity,  and  pH
    through an on-site  performance sur-
    vey;
3.  Provide technical assistance to the
    operator by verbal explanation, minor
    troubleshooting, repair and calibration
    of equipment, and by making recom-
    mendations for sources of corrective
    action;
4.  Prepare brief reports for each site de-
    tailing site characteristics, results of
    quality assurance tests, and technical
    assistance provided;
5.  Computerize results of all information
    gatfiered  from  each site and  submit
    this to the NADP/NTN Quality Assur-
    ance (QA)  Manager on a  quarterly
    basis;
6.  Document the  sites and their  sur-
    roundings by assembling a collection
    of site maps and color photographs.
  This project summary describes proce-
dures and results from quality assurance
visits made to the sites in 1986 and 1987.
Recommendations for improvement  are
also given.

Procedures

Scheduling
  Each NADP/NTN site was to be visited
once in a three-year period. About one-
third of the 201 active sites were visited
in the first year (1986-1987). Prior to the
scheduling  of site visits, RTI consulted
with the  NADP/NTN  QA Manager and
Central Analytical Laboratory (CAL)  site
liaison to determine  which  sites,  if any,
should be seen on a priority basis. When-
ever possible, visits were planned so that
several sites in the same vicinity could be
seen in the same trip.
  The  following  sequence was  followed
when arranging a visit to a collection site.
• About two months before  the  an-
  ticipated date of visit, RTI selects a set
  of sites in a location suitable  for a trip
  lasting up to two  weeks and  sets  a
  tentative schedule for  visits. As many
  as eight  sites  may be visited  in  this
  time frame, depending on the prox-
  imity of the sites.
• The NADP/NTN QA Manager, the CAL
  of the Illinois State Water Survey, and
  the  sponsoring agency are notified by
  letter  of the  proposed visits. CAL
  supplies  pertinent  information  to  RTI
  concerning each site.
• Each site's supervisor is telephoned to
  set  up the visit.  Depending on  the
  wishes  of the site  supervisor,  the
  supervisor contacts the operator  or
  RTI contacts the  operator to confirm
  the  date, time, and place of  the visit.
  When the supervisor  makes the  ar-
  rangements, he contacts  RTI prior to
  the  planned visit to confirm or alter the
  initial plan.
•  EA  form letter is  sent to each site's
  supervisor and operator  that  confirms
  the  date, time, and place of  the visit.
  This letter also gives a  brief agenda
  for  the  visit and  an estimate of  the
  time to set aside for the visit (usually 4
  to 5 hours).


Site Survey Visits
  An auditor accompanied the supervisor
and/or operator  to each collection  site
and field laboratory with the dual aims of
(1)  documenting the site and its imme-
diate surroundings, its operation, and tii
accuracy of its instrument's responses 1
various  quality  assurance tests and (i
providing  information,  training,  an
instruction for operators and supervisor:
equipment  calibration and minor mair
tenance as needed, and establishing cot
tacts for further information and/or maj<
repairs.


Systems  Survey
  A quality assurance systems  surve
was conducted at each  site to qualiU
lively  assess the site, its  surrounding!
and the operator's adherence to procc
dures specified in the NTN design doci
ment' and  in the NADP/NTN site open
tor's  instruction  manual3.  Criteria fc
siting  an NADP/NTN precipitation static
are illustrated in Figure 1. The operate
was asked to demonstrate sample collet
tion and analysis procedures. These wer
observed with specific attention given t
calibration  procedures and sample  har
dling technique. Site equipment was e>
amined  for signs of wear or faulty opera
tion. It was noted whether solutions an
equipment were properly stored. Site log
books and  rain gauge charts (if presen
were  examined  for legibility,  complete
ness, and accuracy.
  Information from the  systems  surve
was entered in the systems survey ques
tionnaire. Photographs (color  slides)  c
the sites were taken. The directions N, E
S, and  W  were photographed with th
precipitation collector in  the foregrouru
Additional views were taken as specific
in the questionnaire.


Performance Survey
  A quantitative performance survey wa
conducted  at each site. Criteria for eva
uating performance are specified in th
NADP Quality Assurance Plan*. All into
mation was recorded  in the performanc
survey questionnaire.  Several items hav
ing to do  with  quality  assurance tes
equipment, materials, and procedures an
discussed in the next section.


Sequence Of Site Visitation
Activities

/. Select Site For Visit And Inltla
Communications
• Advise QA Manager, CAL,  site spor
   sor of plans; request site informatioc
   receive go-ahead

-------
                                                 5 m • No objects greater lhan 1 m In height
                                                         20m  • Stops* 115H
                                                              • Natural v«g*Mion<0£m
                                                              • No grazing animals
Conversion
m - H
03
0.6
S
20
30
500
- 10
• 20
- 16
• 66
- 96
• 1640
kin • ml
10
20
40
• 62
• 12.4
• 246
  Spacing batman ma gauge and cosecax at S to 30 n
  No rasManW buUngs wNNn upwind 30* com
  II mom Kim 20H of pradpMton is snow, gauge muel haw an Mar wM
  ahMd. ptml wk * Mm* **atm n g«ug« Mat
  In tncn»«r»«i.col»cloimj)»ihouldb« property oounlaf nnelgt*^
  QuMfonkilimlinduM
  ChwgM mo* to submitted to coordinator's offic*
30m • No sudden changes In (tap* greater thin tISH
       Farm aim should be nothing «xcepl mgMMkxi maintoinad H teas
       than 06m

      100m •Nosunteeiloraoialagricullunlpfoducls.hMls.wniclM.
            psfking Ms, of miJntonmca ywds
           • No moving ioucoi ol poManH wch n rummy, tailway. road or
            navtgibtortvw

         500 m • No iMd lot*, dairy bvra or targi eonomiratlon ol anlmalt

                   10km 
-------
•  Contact site  supervisor, site  operator,
   sponsoring agency
•  Send letter of confirmation to super-
   visor, operator, agency
•  Advise  EPA  that trip  plans are
   complete

//. Pre-TWp Preparation
•  Make  travel  arrangements (air, car,
   hotel)
•  Prepare  and  test  quality assurance
   materials
•  Review site-specific quality control in-
   formation (maps, QC test results, etc.)
•  Check equipment and supplies
•  Prepare site visit notebook

///. Conduct S/te  Visit
•  Outline activities  to supervisor and
   operator
•  Assess site  and  surroundings  (map,
   photographs, obstructions, sources)
•  QA tests  of precipitation collector and
   gauge
•  Adjust or calibrate collection equip-
   ment as  required
•  Assess operator handling and  trans-
   port of collection bucket
•  QA tests of  conductivity  and  pH
   meters, temperature, mass
•  Examination of  site  records, rain
   gauge charts
•  Answer questions; provide information
•  Prepare  short  report; conduct exit
   interview

IV.  Reporting
•  Short report  prepared; left with super-
   visor or operator
•  Copy  of  short  and extended reports
   forwarded to NADP/NTN QA  Manager,
   to CAL, and to EPA Project Officer
•  Copy  of site visit notebook  sent  to
   NADP/NTN, QA Manager; file original
•  Summary reports  prepared  monthly
   and annually
•  Report presented  to NADP/NTN com-
   mittees


Results and Discussion

Siting Criteria Survey
   Collector Height Standard—The collec-
tor should be installed on its standard 1-
meter high  aluminum base. Any of sev-
eral methods can  be used to stabilize or
level the collector such as concrete pads
or stakes. However,  the bucket height of
the  collector should not vary  from  its
standard height by  more than  ±0.5  m.
An exception to this  criteria is  permitted
in areas with significant accumulations of
snow. In this case, the collector may be
placed on  a platfbrm that is  no higher
than the highest anticipated snow pack.
To prevent obstructions to wind flow, the
collector base should not be enclosed.
  Eleven of 62 collectors checked (18%)
were not at the standard  1-meter above
ground height. Of these 11. all were on
platforms. All of  these sites,  with  the
possible  exception  of  Georgia  Station,
were  in  areas where snow  pack  ac-
cumulates and warranted  being  on  plat-
forms.  The  base  was  enclosed on five
collectors.
  In most  cases, platforms were  short,
not more than 2 to  4 feet in height. The
higher  platforms  (Buffalo  Pass  and
Snowy Range)  were necessary  to  raise
the collector above  snow packs which
could  exceed  10  feet.  The  effect  of
shorter  platforms  on the  sample is be-
lieved to be minimal.
  Wet  Bucket Orientation—The collector
should be  mounted with the  wet-side
bucket to the West and the sensor facing
North.  In this way, the wet-side bucket is
generally upwind  of the dry-side bucket
(winds generally being from the S to SW
in the  eastern  United  States), and the
sensor  is  downwind  of  the  wet-side
bucket. This placement  is designed  to
lessen the  chance for contamination and
to minimize  the obstruction of  the col-
lector itself to sample entry.
  Of  61 sites examined in  1987, 47
(77%)  were  correctly installed with the
wet bucket facing to the SW, W, or NW.
Three were installed with  the wet bucket
facing SSW. These  southerly installations
probably have  no  effect on  the  data.
Eight collectors that were installed with
the collection bucket facing N or E may
cause  an  aberration in the collection ef-
ficiency or sample chemistry. A statistical
study of the long-term data base  would
be  required to  discern this and may be
complicated by other factors.
  Ground  Slope—The collector  should
not be located on  ground with a slope
greater than 15° or 27%.  The  slope at 7
of 62 sites  (11% of the  total) exceeded
this criterion. Six of the seven  sites were
located in mountainous regions and were
representative of their respective regions.
  It is difficult to say what the effect of
the  steeper  slopes is  on  collection
efficiencies. However, to  adhere too
rigidly to the criterion would  effectively
eliminate many regions, especially those
at higher elevations, from this network.
  Collector-Gauge  Separation—The col-
lector should be located within a distance
of 30 m of  the rain gauge but not closer
than 5 m. This guideline is set so that the
collector  and  gauge  "see"  the same
precipitation event  and so  that  neith
piece of equipment offers  an  aerod
namic interference to the  other's colle
tion ability.
  Twelve  of 60  sites (20%) had ra
gauges  less than 5  m from the collectc
The  collector and rain gauge we
mounted on a platform at several of the:
sites and  separating them would requi
the construction  of  a  second platfor
(one site had separate platforms). Son
sites not  meeting the  distance  requir
ments  were  enclosed  by chain-Mr
fences for security;  meeting  the  distam
requirements for these sites  would  n
quire enlarging the  enclosure or buildir
a separate one. Many sites in this ne
work may  not be able to comply  with th
criterion due  to  cost.  However, thos
gauges  which  can  be removed to th
proper distance should be and the mo\
should be documented. None of the sit<
exceeded  the  recommended  separatic
distance of 30 m. The closest separatic
was 0.5 m.
  Collector  and  Rain  Gauge at San-
Height—The heights above ground of tf
collection bucket and  the  rain gauj
orifice  should  be within  1  foot  of  eac
other. Seven of 61   sites  (11%)  did n
meet this criterion. However, in all case
the criterion  was  exceeded by snru
amounts and the effect on the data ba:
is probably negligible.
  Immediate  Site  Surroundings — Th
purpose of  these criteria is to  prevei
sample  contamination or obstruction froi
occurring. The presence of  urban are*
or industry  at  distances  of 10 km  i
greater  from the site is not considered
this  report  since this  information  he
already  been given  in the report of th
first round of site visits  made in 1985 ar
1986, and the site locations have m
changed.  Documentation of the presenc
of new sources after 1980 is  not yet aval
able as  an NPAP emissions inventory.
  Vegetation Within 30 m  and  the  4
Degree Rule—Vegetation within  30 m i
the collector should not be more than tw
feet tall, and no object should project c
the collector from an angle greater the
45°. These  criteria  are intended  to kee
windblown contaminants such as seec
or splashing water  out of the  collectic
bucket.  Fourteen of 62 sites (23%) he
vegetation of height greater  than two fe<
within  30 m of  the collector.  Six  site
(10%) had trees or  meteorological towe!
too near the collectors that violated  th
45°  siting criterion. Under  certain win
conditions,  it  is  possible  that rai
splashing from these objects  into  th
collection bucket could significantly alt
the sample chemistry.

-------
  Parking  Lots, Chemical  Storage-The
site  should not be within  100 m of
parking  lots or chemical storage. Ten of
62  sites  (16%)  did not  satisfy  this
criterion.
  Transit Sources Closer than 100  m —
Transit  sources such  as well-traveled
roads,  railroads, and shipping channels
should be no closer than  100 m to the
sample  collector.  Five of 61  sites (8%)
were judged to be too near such sources.
  Grazing  Animals or Feedlots—The site
should be more than 30 m from  grazing
animals and more than 500 m from large
concentrations  of  animals  such  as
feedlots, dairy farms, or poultry  farms.
These criteria  are  intended to  keep
sources of compounds which could buffer
acidic solutions removed from the  site.
Five of  the 62 sites examined in 1987
were only  affected  by grazing animals,
and the effect here is probably negligible
since animals in these situations  tend to
roam over large areas  and are not
concentrated near the site. None of the
sites, however, were near  feedlots or
other large concentrations  of animals
which could affect the sample chemistry.

Equipment and Sample
Collection Survey
  System  and  performance checks were
made at each  site to assess the opera-
tional fitness  of the Aerochem  Metrics
precipitation collector* and the Belfort rain
gauge. The process of  sample retrieval
and care was also examined. These are
discussed  below.

Precipitation Collection System
Checks
  Collector Level—The precipitation col-
lector  should  be  level  so  that the
collection efficiency will not be biased by
wind direction or variations  in effective
bucket opening  areas. Only  five of 60
collectors  examined  (8%) were not level.
In most cases, the collectors were off
level by small amounts. It is not believed
the effect of  this  variance  would  be
measurable.
  Collector Stable—The precipitation col-
lector should be mounted firmly so that it
will not move in strong winds. Each of the
61 collectors examined was judged to be
stably  mounted (i.e.,  it could  not be
rocked easily by hand).
•Mention of trade names or commercial products
does not constitute endorsement or  recom-
mendation for use.
  Sensor  Clean—The collector's sensor
should be clean. Each of the 61 collector
sensors examined was clean. The  usual
effect of a dirty sensor is to prolong the
length of time  the  wet bucket stays
uncovered after an event, increasing the
likelihood of contamination.
  Counferwe/ghf—The moving bucket lid
should be properly counterweighted so
the lid can be moved without excessive
motor strain  or clutch slippage. Improper
counterweighting is usually found at sites
which have added  snow roofs. Only two
of 61  site collectors were found  to have
improper counterweights. For these two,
there  was no sign  of  excessive clutch
wear.
  Clutch Wear—The motor box clutch
assembly should  not show signs of
excessive slippage or wear  as evidenced
by a shiny clutch surface or rounded-off
indent. Fourteen of 60 sites  (23%) had
evidence  of  clutch  wear  or slippage.
However, other tests,  such  as the
assembly's ability  to lift  two or  more
Belfort gauge weights  without  slipping,
showed  that eight of these 14  were
operating  properly. There are  several
causes  of clutch slippage in  addition to
improper  counterweighting  discussed
above. The usual result of serious clutch
slippage is that samples are not collected
because the bucket lid fails to move off
the wet  bucket  at the  start  of  a
precipitation event.
  Bucket  Tie-Down—Both  the wet and
dry collection buckets should be secured
to the collector  with tie-down straps to
prevent their being blown out during
strong winds. Nine of the 61 sites (15%)
did not secure the buckets.  However, the
actual  incidence of sample loss due to
this variance is probably rare.

Precipitation Collector
Performance Checks
  Cover Seats Properly on Wet Bucket—
The collector's bucket cover  should fit
tightly  and evenly on the rim of the wet
(and dry) bucket so  that  dust cannot
enter during  dry periods (and so  that the
lid liner is protected during  wet periods).
None  of the 58 collectors that could be
examined in  1987 had  bucket lid  seal
problems.
  Lid Tension—The force that the bucket
cover  exerts  against  the  rim  of the
collection  bucket may  be  assessed by
lifting  the lid slightly  above  the bucket
and reading the force (in grams) required
to  do so.  A spring  scale  is  used.
Generally, tensions of 1500 g  or greater
are found. Three sites had  lid tension of
1500 g. The average lid tension was 2384
± 434 g.
  Lid Drop Distance—Another measure
of adequate lid/ bucket seal tension is the
lid drop distance -- the distance the lid
drops  when   the   wet  bucket is
momentarily removed.  The CAL of the
NADP/NTN network  has found that a
distance of 3 mm or greater is required to
give good, dust-free seals.
  Not all collectors were checked in this
manner  because the test was not  used
initially  in the program.  Of  56  site
collectors  checked, none had a lid  drop
distance  of 3  mm or  less.  Minimum,
maximum,  and  mean  distances were 5,
21,  and 14.2 ±  3.7 mm.
  Voltage  to  Event Marker—\n the  wet-
side open mode, the  Aerochem  Metrics
unit should send signal of 14  ± 3 volts to
the  event  marker solenoid of  the Belfort
rain gauge.  Zero  voltage  or  values
outside  the  prescribed  range  are
indications  of  problems  with  the
Aerochem Metrics motor  box  or the
battery,  when one is used to power the
unit. Of 58 units checked,  three  were
lower and one higher than the acceptable
range. The mean voltage was 12.74 ±
1.15. The  highest voltage was 173 at the
Snowy Range-Glacier  Lake site. It  may
be  that  the solar  panel  there  is  over-
charging the  battery.  The low  voltages
were found at the Niwot Saddle,  Quincy,
and Verna Well Field sites.
  Unactivated  Sensor  Temperature —
Generally,  the  temperature of the  pre-
cipitation collector sensor is  at ambient
level when there is no  precipitation. If the
ambient temperature is below 4°C, the
sensor heater should come on, at a  lower
power level, to melt ice or snow that may
fall. A sensor should  not be heating at
ambient temperatures  above  4°C unless
it is raining. If  it is heating  prior to the
rain, light  rainfall striking the  sensor may
evaporate before the circuit can be made
to open the lid. None  of the  53  sensors
checked showed irregular heating at am-
bient temperatures.
  Activated Sensor Temperature—When
activated by precipitation, the Aerochem
Metrics sensor causes the cover to  move
off  the  wet-side collection  bucket. To
speed precipitation evaporation and thus
reduce the time the wet-side bucket is
open  after precipitation ceases, a heater
circuit beneath  the base plate of the
sensor is energized and the temperature
of the sensor (as measured  at the base
plate) rises. The circuit is  thermistor-
limited and is adjusted to 50-60°C  (122-
140°F) at the factory.

-------
  A total of 58  sensors  were checked
during  the  1987 visits. Sensors  at four
sites were not checked since heavy rains
were occurring at the time of the visit. Of
the 58, two were not heating (3%), one
(Snowy  Range-Glacier Lake)  was over-
heating (90°C), and 55 were judged to be
heating properly. Temperatures at 51 of
the 55  sites  were  checked using  a
thermo-couple and digital meter. Of these
51, the lowest temperature  was 33°C
(Meridian), the highest was 76°C  (Sugar-
loaf), and  the  average temperature was
58.2 ±  10.8°C. Of those 51, ten (20%)
heated below 50°C, 22 (45%) were in the
manufactured range of 50-60°C,  and  19
(37%) were above 60°C.
  Resistance to  Trigger  Sensor—The
sensor of the Aerochem Metrics collector
activates circuitry to open the cover when
water droplets bridge the gap between
the grid and the base plate elements of
the  sensor assembly.  According  to
factory specifications, this occurs when
the resistance  between the elements is
reduced to 80  Kohms. This resistance is
sufficiently  sensitive  to cause activation
by  relatively pure  water  (i.e., water of
very low conductivity, such as deionized
water).
  All of the 56 sensors checked  caused
the circuit to activate  when distilled water
was applied to the grid and base plate.
The low, high, and average resistances
measured  by bridging the grid  and the
base plate  through a variable resistance
box were  59.1, 110.0, and  82.3  ± 10.3
Kohms.
Rain Gauge System Checks
  Gauge Level—The rain  gauge should
be  level so that  precipitation  collection
efficiency is not biased by wind direction
or by variation in effective exposure area.
Five of the 61 rain gauges inspected
(8%) were out of level by small amounts.
  Alter  Shields—Alter  shields may  be
used with rain gauges in the NAOP/NTN
network to abate  strong winds near the
gauge  and improve collection efficiency.
Fourteen  of 62  sites (23%)  were  so
equipped.
  Chart Recorders—When checked at
the time of the visit, the chart  recorders
should indicate the correct time ± one
hour. The  rain gauge chart recorders
were off by more than one hour at only 1
of the 61 sites checked.
  Dampening Fluid Levels—The Belfort
gauge  dampening fluid reservoir should
be  filled to within  0.25 inch of the top to
reduce  pen  "noise" created by strong
winds.   Charts  showing excessive pen
noise are difficult to read accurately. The
dampening fluid  level was low  at  10
(17%) of the 60 sites checked.

Rain Gauge Performance Check
  The  rain  gauge  calibration  was
checked using Belfort gauge calibration
weights. Fifty-five of the 59 gauges (93%)
were in calibration (within  ±0.1 inch) up
to 5 inches depth. At a depth of 6 inches,
only 76 percent were in calibration. Errors
associated with  the  crossover point
increased  rapidly  at  the 6-inch  depth
point. Because most  rain amounts  are
measured  in the 0-5 inch range  (except
when winterized), and  because the event
depth is measured as the  difference in
chart reading  before and after the event,
few measurable  errors in  precipitation
mass measurements are expected due to
inaccurate rain gauge calibration.  Twenty
of the  51  rain  gauges checked  (34%)
were calibrated during the  site visits.

Sample Collection Procedures
  Most operators were using proper  col-
lection  procedures and no  instances of
contamination were  noted. Operators
were also  checking for sample contami-
nation while at the site.
  Only one site operator was not able to
make a weekly equipment check.  Seven
operators did not bag and  box the bucket
before transporting  it  to  the field
laboratory. In those cases, the sites were
within  walking distance of the field
laboratory.


Field Laboratory Survey

Systems Check of Field
Laboratory
  Field laboratories had adequate space
and were clean. Fourteen  sites (23%) did
not have  air  conditioning, but were
usually in  areas that would  need it only
rarely.  Required  records  were kept  and
report forms were filled out correctly in all
cases. Rain gauge charts  were annotated
fully  with site name, date, time on, time
off, etc. Field  samples were shipped to
CAL, generally within three  days. Proper
techniques for weighing  samples were
followed in all instances.


pH and Conductivity
Measurement Techniques
  In  order to make accurate and  precise
pH and conductivity measurements,  the
analyst must  be familiar  with the meas-
urement equipment,  follow  appropriate
calibration procedures, and use measure-
ment techniques  consistent with good
laboratory  practices.  Field  person
were observed while making pH and c
ductivity  measurements and  adherei
to technique was noted.
  In general, specified procedures w
adhered to and laboratory technique i
good. Only two sites (3%) had a variai
with pH measurement technique. Not
sites use the inverted cell technique
measuring  conductivity. This is  usu.
due to the fact that the site does not
the type of  electrode that may
inverted.
  Each site  operator rinsed the cond
tivity cell  with  sample before taking
second reading. Only one site  measu
the standard, deionized water,  and s«
pie in  the wrong order. NADP/NTN f
cedure requires the measurement of
ionized water after the 75 jiS/cm stand
and before  the precipitation sample
ensure there is  no carry-over from
standard to the lower ionic strength r
sample. Proper analysis procedures w
always discussed with site operat
whenever the need was evident.


Results of Field Stte Analysis c
Simulated Precipitation
  Each field laboratory was  asked
analyze a performance audit solution
conductivity  and pH.  These  solutic
were   prepared by dilution  of  EF
supplied  performance test solutions;
audit  value  is that  designated  by E
and lies between pH 3.8 and 4.8 and I
a conductivity between 20 and 77 pS/c
Designated quality limits are  ± 0.1 t
for pH  and  ± 4 iiS/cm for conductivity
  Ninety-seven percent of  the 60 fi
laboratories  checked  had  pH  rest
within  ±0.1 unit of the designated val
The  average  absolute difference v
0.036  ±  0.03 pH unit. Of the two si
that exceeded the accuracy requirenru
one had  a faulty pH electrode and
other  had a faulty  meter. Two sites I
inoperative pH electrodes and  could
be tested.  Ninety-seven percent  of
field laboratories  obtained  conducts
values which varied by no more than :
nS/cm from the designated value. 1
average absolute difference was 1.46
1.34 nS/cm.
  Balances are used at the sits to we
the mass of precipitation collected by
Aerochem  Metrics  collector.  They
usually triple  beam-balances.  The t
ances were checked with weights rang
from approximately 800 to  4000 grai
Only three of 58 sites checked had err
of greater than 5 grams over the range
test  weights. The worst  case was
 + 10.9 g disagreement at  a loading

-------
3292 g. This was discovered to be due to
 inding of the beam arm dampening at-
 ichment in the mechanism of the mag-
netic damper. This was corrected and the
Jisagreement was only 4.4 g.
  Of 58 balances checked, the absolute
 iverage of the worst case differences
was 2.2 g. This  usually  occurred at the
naximum  load applied and corresponded
o  less than 0.1  percent of the  average
load of 3736 g, or to less than 0.002 inch
of  rain (where 1724 grams =  1 inch of
rain for the Aerochem Metrics collector).
  The  full report was submitted in  ful-
fillment of Task  231B of  EPA Contract
No. 68-02-4125 by Research Triangle In-
stitute. This report covers site visits made
during the  period  October  1,  1986
through September 30,  1987,  and all
work was completed as of September 30,
1987.

References
 1. Bigelow, D.S.  "NADP Instruction
    Manuel  - Site  Operation." National
    Atmospheric Deposition Program, Ft.
    Collins, CO. January  1982.
 2. Bigelow,  D.S.  "Instruction  Manual:
    NADP/NTN Site  Selection  and In-
    stallation."  National  Atmospheric
    Deposition Program,  Ft. Collins, CO.
   July 1984.
 3. Robertson,  J.K.  and  J.W.  Wilson.
    "Design of the  National Trends Net-
   work for Monitoring the Chemistry of
   Atmospheric Precipitation."  U.S.
   Geological  Survey  Circular 964.
    1985.
 4. "The NADP Quality Assurance Plan."
    NADP Quality  Assurance  Steering
   Committee. 1984.

-------
W. Gary Eaton, Curtis  E. Moore, Dan A. Ward, and Richard C. Shores are with
  Research Triangle Institute, Research Triangle ParH, NC 27709.
Berne I. Bennett is the EPA Project Officer (see below).
The  complete  report,  entitled  "The  National Atmospheric  Deposition
  Program/National Trends Network (NADP/NTN) Site Visitation Program (October
  1986  through September 1987)," (Order No.  PB 89-151 5421 AS;  Cost: $15.95,
  subject to change) will be available only from:
        National Technical Information Service
        5285 Port Royal Road
        Springfield, VA 22161
        Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
        Atmospheric Research and Exposure Assessment Laboratory
        U.S. Environmental Protection Agency
        Research Triangle Park, NC 27711
United States                   Center for Environmental Research
Environmental Protection         Information
Agency                        Cincinnati OH 45268
Official Business
Penalty for Private Use $300

EPA/600/S3-89/007
         000085833   PS
                             fC*lO

                               «M
                                      60604

-------