SERA
United Stales
Environmental Protection
OHiee ot
Toxic Substances
Washingjon DC 2W60
EPA-56Q/5-66-017
May. 1986
Toxic Substances
FIELD MANUAL FOR
GRID SAMPLING OF
PCB SPILL SITES TO
VERIFY CLEANUP
a
D D D D
nan
D a D a
:
;ff
a
•
> -"
•• •
-------
it
L_
F1ELD FOR OF PCS
SPILL SITES TO VERIFY CLEANUP
By
Gary L, Kelso
Mitchell 0, Erickson
MIDWEST RESEARCH INSTITUTE
and
David C. Cox
WASHINGTON CONSULTING
INTERIM NO. 3
37
EPA Contract No.
MR! Project No, 8501-A(37)
and
EPA Contract No, 68-01-6721
WCG Subcontract to Battelle Columbus Laboratories
No. F4138(8149)435
Prepared for
U.S. Environmental Protection Agency
Office of Toxic Substances
Field Studies (TS-798)
401 M Street, S.W.
Washington, DC
Attn: Mr. Daniel T. Heqqem, Work Assignment Manager
Or. Joseph J. Breen, Project Officer
Richard A. Levy, Work Assignment Manager
Cindy Stroup, Project Officer
-------
DISCLAIMER
This document has been reviewed and approved for publication by
the Office of Toxic Substances, Office of Pesticides and Toxic Substances,
U.S. Environmental Protection Agency. The use of trade names or commercial
products does not constitute Agency endorsement or recommendation for use.
-------
PREFACE
This Interim Report was prepared for the Environmental Protection
Agency under EPA Contract No. 68-02-3938, Work Assignment 37. The work
assignment was directed by Mitchell D. Erickson. This report was prepared
by Gary Kelso and Dr, Erickson of Midwest Research Institute (MRI). David C.
Cox of the Washington Consulting Group, 1625 I Street, N.W. , Washington, O.C.
20006, contributed to the sampling design (Section 5.0) and compositing strat-
egies (Appendix) sections under subcontract to Battelle Columbus Laboratories,
Subcontract No. F4138(8149)435, EPA Contract No. 68-01-6721 with the Design
and Development Branch, Exposure Evaluation Division,
This report is a revision of a previous draft report entitled
"Field Manual for Verification of PCB Spill Cleanup" (Draft Interim Report
No. 3, Task 37, EPA Prime Contract No. 68-02-3938, June 27, 1985). Both
English and metric units are used in this document, where appropriate. EPA
field inspectors will most commonly measure the site in English units; there-
fore these units were used for the site measurements in this report.
The EPA Work Assignment Managers, Daniel T. Heggem, Richard A. Levy,
and John H. Smith, as well as Joseph J. Breen and Cindy Stroup of the Office
of Toxic Substances, provided helpful guidance, Ms. Joan Westbrock and
Mr. Ted Harrison of MRI and Mr. David Phillippi and Mr. Robert Jackson of
EPA Region VII assisted in the field validation of this manual.
MIDWEST RESEARCH INSTITUTE
3aul C. Constant
Program Manager
Approved!
v > f-
John E. Going, Director
Chemical Sciences Department
May 1986
11
-------
TABLE OF CONTENTS
1.0 Scope and Application, , , , 1
2.0 Summary, .....,,..«,...,.,., 2
3.0 Safety , , . 3
4.0 Sampling Equipment and Materials ... 4
4.1 Personnel Equipment. ,...,...,...,.. 4
4.2 Sampling Equipment and Materials .... 5
4.3 Documentation Materials. .... .... 6
4.4 Trip Preparation 6
5.0 Sample Design, ....................... 6
5.1 Step 1: Diagram the Cleanup Site 7
5.2 Step 2: Diagram All Cleanup Surfaces in the
Same Plane ....... 9
5.3 Step 3: Find the Center and Radius of the
Sampling Circle 9
5.4 Step 4: Determine the Number of Grid Sample
Points to Use 12
5.5 Step 5: Lay Out the Sampling Points on the
Diagram Constructed in Step 2. ......... 15
5.6 Step 8: Lay Out the Sampling Locations on
the Site 22
5.7 Step 7: Consideration of Special Cases. . 23
5.8 Example of Laying Out the Sample Design. ..... 24
6.0 Sample Collection, Handling, and Preservation 28
6.1 Surface Soil Sampling. 30
6.2 Soil Core-Sampling ....... 31
6,3 Water Sampling . , 32
6.4 Surface Sampling ........ . . 33
6.5 Vegetation Sampling. ....... 35
6.6 Compositing Strategies ........ 35
7.0 Quality Assurance 35
8.0 Quality Control 36
8.1 Field Blanks 3?
8.2 Sampling Without Contamination ...,..,.,. 37
8.3 Sample Custody 38
8.4 Documentation of Field Sampling 39
111
-------
TABLE OF CONTENTS (Concluded)
Page
9.0 Documentation and Records. ,,,.,... 40
9.1 Equipment Preparation Log Book 40
9,2 Sample Codes ..,,.,.,. 40
9,3 Field Log Book . 41
9.4 Site Description Forms 41
9.5 Chain-of-Custody Forms 42
9.6 Sample and Analysis Request Forms 42
9.7 Field Trip Report. . 43
10.0 Validation of the Manual ....... 43
11.0 References 45
Appendix - Strategies for Compositing Samples. ............ A-l
L
j TV
-------
LIST OF TABLES
No. Title Page
I Required Number of Grid Samples Based on the Radius
of the Sampling Circle .,....,,,. 12
2 Geometric Parameters of the Hexagonal Grid Designs,
for Sampling Radius r , 15
LIST OF
Ho_ Title Page
1 Example PCB spill site diagram, ....,.,...,,,. 8
2 Example spill cleanup site diagrammed in the same plane , . 10
3 Locating the center and sampling radius of the example
spill cleanup site. . 11
4 Method to find center and radius of the sampling circle . . 13
5 Locating the center and sampling circle radius of
irregularly shaped spill areas. . . 14
8 Location of sampling points in a 7-point grid . . 16
7 Location of sampling points in a 19-point grid 17
8 Location of sampling points in a 37-point grid, ...... 18
9 Construction of sampling grid on a site diagram ...... 19
10 Sampling locations on the example PCB spill site. ..... 21
11 Scale diagram of PCB spill site 25
12 Determining center (C) and sampling radius (r) of
sampling circle ........... 26
13 Diagram of 19-point grid superimposed on the PCB spill
site '...... 29
-------
1,0 SCOPEAND APPLICATION
The purpose of this manual is to provide detailed, step-by-step
guidance to EPA staff for using hexagonal grid sampling at a PCB spill site.
Emphasis is placed on sampling sites which have already been cleaned, although
the sampling methods presented may also be used at PCB spill sites which have
not been cleaned. Guidance is given for preparing the sample design; collect-
ing, handling, and preserving the samples taken; maintaining quality assurance
and quality control; and documenting and reporting the sampling procedures
used. An optional strategy for compositing samples is given in the appendix,
This is a companion document to the report "Verification of PCB
Spill Cleanup by Sampling and Analysis" (EPA 560/5-85-026, August 1985, Second
Printing). That report provides an overview of PCB spill cleanup activities
and guidelines for sampling and analysis including: sampling designs, sam-
pling techniques, analytical techniques, selection of appropriate analytical
methods, quality assurance, documentation and records, and reporting results.
The previous report provided the rationale and background for the techniques
selected and describes many options in greater detail.
This "how-to" report concentrates on detailed guidance for field
sampling personnel and does not attempt to provide background information on
the techniques presented. This manual addresses field sampling only and does
not provide information on laboratory procedures, including sample analysis,
data reduction and laboratory data reporting. The types of field sampling
situations discussed in this manual are those typically found when a PCB spill
results from a PCB article, PCB container, or PCB equipment spill. Unusual
PCB spill situations, such as elongated spills on highways from a moving
vehicle, large spills in waterways, and large, catastrophic spills, are not.
addressed.
-------
2.0 SUMMARY
This manual is divided into the following sections:
• Safety
Sampling Equipment and Materials
Sample Design
Sample Collection, Handling, and Preservation
Quality Assurance
Qua!i ty Control
Documentation and Records
Validation of the Manual
Safety aspects of field sampling include wearing proper protective
equipment, practicing good hygiene, using safe work practices, and training
field inspectors in safety procedures. Sampling equipment and materials in-
clude personnel equipment, sampling equipment, and documentation materials.
Prior to making the field sampling trip, the EPA inspector should ensure that
all sampling equipment and materials are available, and that all sampling
containers and equipment have been properly precleaned.
The sample design is based on a hexagonal grid of 7, 19, or 3? sam-
ple points. A step-wise method describes how to construct a diagram of the
PCS spill site on graph paper; determine the radius and center of the sam-
pling circle; determine which grid size to use; lay out the grid on the dia-
gram; and then lay out the sampling grid on the site.
After the sampling grid has been laid out on the site, a sample
must be taken at each grid point. Methods to collect, handle, and preserve
different types of samples, including surface soil samples, soil core samples,
surface and subsurface water samples, wipe samples from nonporous hard sur-
faces, destructive samples from porous hard surfaces, and vegetation samples,
are suggested. For each type of sample to be taken, methods are recommended
to prevent cross-contamination between samples.
-------
Quality assurance (QA) and quality control (QC) must be an integral
part of any sampling scheme, A quality assurance plan must be developed by
appropriate EPA offices according to EPA guidelines and be submitted to the
regional QA officer or other appropriate QA official for approval prior to
sampling PCB spill sites. Each EPA office must, operate a formal QC program
and all QC measures should be stipulated in the QA plan. Some of the require-
ments of quality control are discussed in this report, including field blanks,
sampling without cross-contamination, sample custody, and documentation of
the field sampling activities.
All sampling activities should be thoroughly documented and reported
as a part of the verification process. Each EPA office is responsible for
preparing and maintaining complete records, including an equipment preparation
log book, a field log book, site description forms, chain-of-custody forms,
sample analysis request forms, and field trip reports.
Section 10.0 briefly describes a field study which was conducted to
test and validate the sample design given in this manual. The study showed
that the sampling design is easy to follow and understood by those unfamiliar
with the manual prior to reading it, and that the grid sample points can be
correctly laid out in a relatively short period of time.
The appendix gives strategies that may be used to composite the
samples taken at a PCB spill site when compositing is deemed to be desirable.
3,0 SAFETY
A PCB spill site which has been cleaned up should have very low
levels of PCBs present. The EPA inspectors) who sample the site to verify
that the site has been properly cleaned up should, however, take pre-
cautions to minimize any exposure to PCBs or other potential hazards at the
s i te.
-------
In order to ensure that the inspectors understand and practice good
safety procedures, a training and education program should be established and
a health and safety manual provided by the responsible EPA officer. The pro-
gram should inform inspectors of the potential hazards of exposure to PCBs,
and the proper safety procedures to follow when sampling PCB spill sites.
4 • ° SAMPLING. EQU1PME_NT AND MAT E RIA L S
The equipment and materials required to sample a PCB spill site
will vary with the types of samples to be taken. The general lists of equip-
ment and materials given below must be adjusted for the specific requirements
of each spill. The lists include personnel equipment, sampling equipment and
materials, and documentation materials which should be taken to the spill site
by the EPA inspector. These equipment and materials must be assembled prior
to making the site visit, and all sampling containers and sampling equipment
must be precleaned.
4,1 Personnel Equipment
The inspector should take the following personnel equipment to the
spill s i te:
Disposable rubber gloves
Plastic overshoes
Safety glasses
Impervious paper-like coveralls
Hardhat
Safety shoes
First-aid kit
Other safety equipment specified by safety officer
-------
kets
:'lot,ru
b a r j s
-------
-------
grid has a 98% chance of detection of this contamination level, while analysis
of six random grab samples from the site has only a 3% chance of detection
(Boomer et al. 1985).
The hexagonal grid sampling design is to be laid out within a sam-
ple circle centered on the spill site, and extending just beyond its boun-
daries. Preparation of the design requires the following steps:
Ste_£_l: Diagram the Cleanup Site
Step 2: Diagram All Cleanup Surfaces in the Same Plane
Find the Center and Radius of the Sampling Circle
Determine the Number of Grid Sample Points to Use
Lay Out the Sampling Points on the Diagram Constructed in
Step 2
Lay Out the Sampling Locations on the Site
Consider Special Cases and Use Judgment for Sample Points
The discussion which follows gives the methods to be used in accom-
plishing each step of the hexagonal grid sampling design, using a three-
dimensional spill surface as an example. Following this discussion, a simple
example of laying out the sample design on a rectangular two-dimensional sur-
face is given.
5.1 Step 1: Pi agram the C1eanug Site
Draw a scale diagram of the cleanup site on graph paper, including
vertical surfaces (walls, fences, etc.), noting important dimensions and dif-
ferent types of surfaces (sod, cement, asphalt, etc.). Such a diagram may
sometimes be found in records of the cleanup. If not, site measurements
should be taken. Great accuracy (e.g., using surveying instruments) is not
necessary, however; the use of a tape measure and pacing should be adequate.
An example diagram is shown in Figure 1 on a scale of 1 in, = 4 ft.
-------
figure 1, Example PCB spill site diagrar
-------
The site diagram should include as many reference points as neces-
sary to relocate the spill area in the future, if necessary. For example,
a spill site in an open field should be located with respect to nearby struc-
tures such as roads, telephone poles, buildings, etc. The direction of north
should be indicated on the diagram.
If available, a detailed drawing or a survey plot of the spill site
should be obtained from the individual(s) that cleaned the site.
5. 2 Step 2: Diagram Aj 1__C_1eanup Surfaces in the Same PJane
The purpose of this second diagram is to determine and show the
dimensions of the total cleanup area, including vertical surfaces, so that
the required sample size can be found. The diagram also facilitates the
determination of sampling locations on vertical surfaces. Constructing the
diagram is analogous to flattening a cardboard box. All vertical surfaces
are placed in the same plane as the adjoining horizontal surfaces. Figure 2,
also on a scale of 1 in, = 4 ft, shows the example spill cleanup site dia-
grammed in the same plane. The actual site dimensions are shown in feet.
5.3 Step 3: Find the Center and Radius of the Sampling Circle
In practice, the contaminated area from a spill will be irregular
in shape. In order to standardize sample design and layout in the field,
samples are collected within a circular area surrounding the contaminated
area. The sampling circle is, approximately, the smallest circle contain-
ing all cleanup surfaces diagrammed in Step 2,
A recommended procedure for finding the center and radius of the
sampling circle is illustrated in Figure 3 and is described below;
1. Draw the longest dimension, lj, of the site diagram in Step 2.
2. Find the midpoint, P, of Lj.
-------
N
r
20'
,
!
|
Wall
(Concrete)
1
I
Dirf i
Driveway
(Asphalt)
i^-"
1
Sidewalk (Concrete)
V
V
>
-*<
N
L
\
\
1
\
\
\
\
>
\
\
*
DJrt
I o— 1/2'
v
!
Lot
(Dirt)"
/
N
/
/
/
^.]
A
/
-
-------
N
\
4'
-------
3. Draw a second dimension, L2, through P perpendicular to LI.
L2 extends to the boundaries of the site diagram.
4. The midpoint, C, of L2 is the center of the sampling circle.
5. The distance from C to either end of the longest dimension, Lj,
is the samplingradius. r.
Figure 4 illustrates the application of this procedure to a site
with an irregular shape, and Figure 5 shows the procedure for a variety of
irregularly shaped areas. These figures show that the center and radius
determined are generally reasonable.
5. 4 Step 4: Determine the_ Number_of Gri_dSamplePoints to Use
The number of grid samples to be taken at a site depends upon the
radius of the sampling circle, which is determined from the scale diagram
shown in Figure 3. The number of samples to be taken at a spill site should
increase as the radius of the sample circle increases. The reason for this
is that the probability of detecting residual PCB contamination at a given
site increases as the number of grid samples increases. Table 1 shows the
required number of grid samples for sampling circles with a radius of 4 ft or
less (seven samples); greater than 4 ft to 11 ft (19 samples); and greater
than 11 ft (37 samples),
Table 1. Required Number of Grid Samples Based
on the Radius of the Sampling Circle
Sampling radius, r (ft) Number of Samples
54 7
> 4 - 11 19
> 11 37
12
-------
(a) Draw longest dimension, I],
on site diagram.
(b) find midpoint, P» of l|.
(c) Draw line, t2t through P
perpendicular to L|,
(d) The midpoint, C, of L2 is the
center of the sampling circle.
(e) The distance from C to the end of I]
is the sampling radius, r.
Figure 4, Method to find center and radius of the samplinq circle.
-------
=„ y°""
\
Figure 5. Locating the center and sampling circle radius of
irregularly shaped spill areas.
14
-------
The radius, r, for the example site is 3-1/4 in, in Figure 3. Thus
ihe actual site sampling radius is 13 ft (3-1/4 in. x 4 ft/in.) and the num-
ber of grid samples required is 37,
Figures 6, 7, and 8 illustrate the hexagonal grid sampling design
for the three sample sizes given in Table I, for a sampling radius of 4, 10,
and 20 ft, respectively,
5. 5 Stej3__5j__Lay_Out the Sampling Points on the Diagram Cons true ted _j n
The geometric properties of the hexagonal designs can be used in
many ways to lay out the sampling points. Perhaps the simplest way to pro-
ceed is as follows. Define s to be the distance between adjacent points and
u to be the distance between successive rows of the design. The distances
s and u are given in terms of the sampling radius, r, in Table 2 below for
the given number of samples defined by the radius rule and listed in Table 1,
Table 2. Geometric Parameters of the Hexagonal Grid Designs,
for Sampling Radius r
Number of samples
7
19
37
Distance, s, between
adjacent sample points
0.87r
0.48r
0.30r
Distance, u,
successive
0,75r
0.42r
0,26r
between
rows
The recommended method for laying out the sample points of the
hexagonal grid on the scale diagram is illustrated in Figure 9 and is de-
scribed below,
1. Draw a diameter of the sampling circle on the scale diagram.
The orientation of the diameter (e.g., east-west) should be chosen to maximize
the number of sample points which fall within the spill area, when practical.
15
-------
4
3
2
Y 0
1
2
- o
a
0
X
The outer boundary of the contaminated area
Is to be 4 feet from the center (C)
of rtie spill site,
6, Location of in
a 7-point grid.
16
-------
10
8
6
4
2
Y 0
2
-a
I i i
i i i i t i i i j 1 j ' '
10 8 64202468 10
X
The outer boundary of the contaminated area is assumed to be
10 feet from the center (C) of the spill site.
Figure 7. Location of sampling points in a 19-point grid.
n
-------
20
16
12
8
Y 0
8
12
16 -
20
a
a
a
o a
a
a
a a
iC a
D
a a
a
D a
....... i ....................... .1 ......................... L_ i
i \ t
20 16 12 8 4 04 8 12 16 20
X
The outer boundary of the contaminated area Is assumed to be
20 feet from the center (C) of the spell site,
Figure 8, Location of in a 37-point grid,
18
-------
•c
(a) Center of cleanup area, C,
(b) Middle row of grid points
located distance, s, apart.
(c) Next two grid rows perpendicular
distance, u, from middle row.
(d) Completed 19 sample point grid,
figure 9. Construction of sampling grid on a site diaqram.
19
-------
A transparent overlay like Figures 6, 7 and 8 (using the appropriate scale)
may be helpful in determining the orientation of the diameter.
2. Place the center point of the hexagonal design at the center (€}
of the sampling circle, Lay out the middle row of the design along the diam-
eter with successive points a distance, s, apart.
3, To 'lay out the next row, find the midpoint between the last two
sample points of the middle row and move a distance, u, perpendicular to the
middle row as shown in figure 9, This is the first sample point of the next
row. Now lay out the remaining points at distance s from each other. By
systematically following this plan, the entire design can be laid out.
Figure 10 shows the sample point locations for the 37 grid points
for the example PCS spill site diagrammed previously in Figures 1, 2, and 3.
On the diagram, r = 3-1/4 in. so from Table 2 the grid spacing is s = 0.30r -
1 in, and the distance between the rows is u = 0.26r = 7/8 in.
In Figure 10, a horizontal diameter is drawn through C. Sampling
locations 1 through 7 are marked I in, apart. To lay out the next row of the
design, we first find location 8. Point D is the midpoint between locations
3 and 4. Then, as described, location 8 is a vertical distance u = 7/8 in.
{3 ft 6 in. on the site) above D. Now locations 9 through 13 are laid out
1 in. apart. In the same way, locations 14 through 18 are found. Continuing
so, the entire grid is marked on the diagram.
All of the sample points in Figure 10 are numbered (1 to 37), Any
type of numbering system can be used, but the points must each be identified
so that the location of the samples taken can be identified by reference to
the diagram points.
Note that sampling locations 4, 7, 8, 13, 23, 34, 35, 36, and 37
are outside the cleanup area. Of these, locations 4, 8, 23, 34, and 35 do
not correspond to a physical location—al 1 are in "thin air," so to speak--
and samples cannot be collected at these locations. Locations 36 and 37 are
concrete samples; locations 7 and 13 are dirt samples (from Figure 2).
20
-------
Sampling Circle
N
L
T;
a
9
D
23
II
33
34
a
1?
0
32
-e-
21
a
15
D
I]
22\
\
D
12
26 U
o a
31 30
D a -
35 36 *:
\
Diameter
4*
Figure in. Semolina locations on the exampl
-------
The orientation of the sample circle diameter shown does not ac-
tually maximize the number of points falling within the spill area, since a
45° clockwise rotation would result in only 8 points lying outside the spill
area instead of the 9 points shown, However, a 45° orientation would make
the sample points very difficult to locate on the actual site with little to
gain by the addition of one more sample point within the spill area.
5,6 Step 6: Lay Out the Sampjing Locations on the Site
To locate the sample points on the site, use the same procedure as
was used to construct the diagram of the sample points in Step 5, but use a
tape measure or pacing, as appropriate, to measure distance. Since s = 1 in.
in the diagram (figure 10), then s = 4 ft on the site. Similarly, u = 3 ft
6 in. on the site. It may be helpful to show the actual distances (in ft) on
the diagram before laying out the site sample points. For example, the sam-
ples on the wall are most easily found by measuring the distance on the scaled
diagram from one end of the wall and the height above the driveway, and then
converting these measurements to find the actual location on the wall. Con-
sider point 32, for example. On Figure 10, it is located approximately 3/4 in.
above the driveway and 5/8 in, from the left edge of the wall. On the site,
then, this point is 3 ft above the driveway and 2-1/2 ft from the left edge
of the wal1.
The PCB spill site should be considered contaminated until labora-
tory analyses of the samples taken verify the site is clean. Therefore, cau-
tion should be exercised when marking the sample points on the site to prevent
possible cross-contamination. The inspector should make minimum contact with
the spill surfaces. One method for accomplishing this would be to cover the
surfaces with plastic sheeting.
22
-------
-------
sampling points include suspicious stains outside the spill area, cracks or
crevices, or any area where the inspector suspects inadequate cleanup,
5.7.4 Samp1ing Sma11 ftreas
The grid sample design specifies that seven samples should be taken
in areas which have a sample circle radius of less than 4 ft. In cases where
the spill area is very small, fewer than seven samples can be taken at the
discretion of the EPA inspector.
5,8 Exampleof Laying Outthe Sample Design
This section summarizes the step-wise procedures required to deter-
mine the locations of the grid sample points at a PCB spill site. The example
used is a simple 8 x 10 ft rectangular spill site.
Steps 1 and 2: Measure and Diagram the PCB Spill Cleanup Site
The PCB spill cleanup site must first be measured (usually with a
tape measure). Then the site should be drawn to scale on graph paper. In
this example, the site is assumed to be an 8 x 10 ft rectangle, as shown in
Figure 11. A scale of 1 in. = 2 ft is used.
Step 3: Determine theCenter and Radius of the Sampling Circle
The center and radius of the sampling circle is determined on a
separate diagram as follows, and is illustrated in Figure 12:
1. Draw the site diagram to scale (same as Figure 11).
2. Draw a line representing the longest dimension, Ll5 of the
site diagram.
3. find the midpoint, P, of Lj.
24
-------
N
Soil
2'
Scale
Figure 11, Scale of PCS spill site.
-------
N
i '
*
'igure 12.
Determini no center (C
of sampline circle.
) and samo 1 irtc rad 1 us ir'!
-------
4. Draw a second line, L2, perpendicular to Llt through point P.
Line L2 must extend to the boundaries of the site,
5, Find the midpoint, C, of line L2. Point C is the center of the
sampling circle, (In this example, points P and C coincide,
but will not coincide for many other types of configurations.)
6. Measure the distance from point C to either end of L1, which
is the samp!ing radius, r. The distance, r, should be measured
to the nearest 1/16 in,
7. Scale radius, r, up to actual size. In this example, the
radius, r, is 3-1/4 in, on a scale of 1 in. = 2 ft, or 6-1/2 ft
(3-1/4 in. x 2 ft/in.).
Step 4: Find the Nunber_gjf_Grj_d_Samgles_to be Used
The number of samples to be taken in a hexagonal grid depends upon
the length of the sampling radius, as shown in Table 1 and repeated here,
Sampling Radius, r (ft) Number of Samples
S 4 7
> 4 - 11 19
> 11 37
Since the radius in this example is 6-1/2 ft, the number of sampling
points would be 19.
Step 5: Plot the Sampling Points on the Site Diagram
The sampling points in a grid row are a distance, s, apart; and the
grid rows are a distance, u, apart. The distances s and u are determined from
the following table.
27
-------
Distance, s, Between Distance, u. Between
Number of Samples Adjacent Sampl_e Points Adjacent Rows
? 0.87 r 0,75 r
19 0.48 r 0,42 r
37 0,30 r 0-26 r
In this example, the distance, s, between the points in a row is
1-9/18 in. [(0.48) x (3.25 in.)] on the diagram, or about 3 ft 2 in,
[(1-9/16 in.) x (2 ft/in.)] on the actual site. The distance, u, between
rows is 1-3/8 in, [(0,42) x (3.25 in.)] on the diagram, or about 2 ft 9 in.
[(1-3/8 in.) x (2 ft/in.)] on the actual site.
The center point of the grid lies on the center, C, of the sampling
circle. Construct the hexagonal grid and superimpose it over the site diagram
(constructed on a third piece of graph paper), as illustrated in Figure 13 for
this example. The middle row of the grid (points 1 through 5} should be
oriented to maximize the number of sample points which lie within the bound-
aries of the spill cleanup site.
It should be noted that adjacent rows are staggered, and that the
sample points of one row are located midway (horizontally) between the sample
points of the other row.
Step 6: Hark the Sample Points on the jrtg
Starting at the center, C, of the spill cleanup site, mark the mid-
dle row points a distance of 3 ft 2 in. apart. Locate the adjacent rows a
distance (u) of 2 ft 9 in. from the middle row, and mark the four sample
points in each of these rows a distance of 3 ft 2 in, apart. Complete the
site sampling grid with the other two rows of sample points.
8.0 SAMPLE COLLECTION, HANDLING fiND PRESERVATION
After the sampling grid has been diagrammed on the site description
forms and laid out on the site, a sample must be taken at each grid point.
28
-------
a
10
a
11
a
12
N
a
6
a
7
o
1
o
2
D
9
u = 1-3/8" - 2'9'
s* (midpoint)
3s= 1-9/16
-N3
O
13
O
14
O
15
D
16
O
5
a
17
a
18
D
19
Scale
Figure 13, of 19-point grid on the
PCB spill site.
-------
-------
template and scoop may then be placed in a plastic bag to be taken back to the
laboratory to be cleaned for the next field sampling job. The rubber gloves
should be discarded into a plastic bag which will be disposed of as PCB-
contaminated material if any samples exhibit PCB contamination.
If a sufficient number of templates or scoops are not available to
use only one item per sample, then each of these equipment items must be
thoroughly cleaned between samples. The template and scoop should be thor-
oughly rinsed with solvent and wiped with a disposable wiping cloth (which
should be discarded into the plastic bag intended for disposal of PC8-
contaminated materials).
6, 2 So i 1Core-Samp1ing
When core samples of sod or soil are needed, the samples may be taken
using a coring device such as a piston corer or King-tube sampler. Core sam-
ples should be taken to a depth of about 5 cm. The soil core can be pushed
out into a precleaned glass bottle and capped, or the tube containing the sam-
ple can be wrapped in solvent-rinsed aluminum foil, depending upon the type
of coring device used. The sample should be properly labeled, a yellow TSCA
PCB mark affixed, and placed in an ice chest (to keep the sample about 4°C).
If samples are to be analyzed soon, the cold storage requirements may be re-
laxed as long as sample integrity is maintained. The sample collection data
should be entered in the field log book and on the chain-of-custody form.
Core samples of soil or sod should be taken with individual core
tubes for each sample. If this is not possible, then the coring device should
be rinsed with solvent and wiped with a disposable wipe cloth to remove any
visible particles before taking another sample. After each sample, rubber
gloves and wipe cloth should be discarded into a plastic bag intended for dis-
posal of PCS-contaminated materials.
31
-------
8.3 Water Sampling
PCB spills on water may result in a surface film (particularly when
the PCBs are dissolved in hydrocarbon oils) or sink to the bottom (particu-
larly when the PCBs are in askarel or other heavier-than-water matrix). When
a surface film is suspected (or visible), the water surface should be sampled.
Otherwise, a water sample should be taken near the bottom of the body of water.
6.3.1 Surf_ace S amp 1ing
Surface water samples should be collected by lowering an open, pre-
cleaned glass sample bottle horizontally into the water at the designated sam-
ple collection point. As water begins to run into the bottle, slowly turn
the bottle upright, keeping the lip just under the surface so that only surface
water is collected. Lift the bottle out of the water, wipe the outside with
a disposable wiping cloth, and cap the bottle. Label the bottle, affix a
yellow TSCA PCB mark, and put the bottle in an ice chest (to keep the sample
at about 4°C). If samples are to be analyzed soon, the cold storage require-
ments may be relaxed as long as sample integrity is maintained. The sample
collection data should be entered in the field log book and on the chain-of-
custody form. The wiping cloth and rubber gloves should be discarded into a
plastic bag used for disposal of RGB-contaminated materials.
6.3.2 Subs u rfa ce Samp 1i ng
Water near the bottom of the body of water should be sampled by
lowering a sealed sampler bottle to the required depth, removing the bottle
top, allowing the bottle to fill, and removing the bottle from the water.
Transfer the subsurface sample into a precleaned glass bottle and cap. Wipe
the bottle with a disposable wiping cloth, fill out and label the sample bot-
tle, affix a yellow TSCA PCB mark, and put the sample bottle in an ice chest.
If samples are to be analyzed soon, the cold storage requirements may be re-
laxed as long as sample integrity is maintained. The sample collection data
should be entered into the field log book and on the chain-of-custody form.
32
-------
'"" sea r d e d 1n t o a p 1 a s t i c b a g
To prevent cross-contamination of samples, separate sampler bottles
should be used to take th<=> samples. Alternatively, the sampler bottle can be
rinsed three times with distil iecl w^te*", so i \ ent- r i nset;, and air^ar"ed beiweer
samp 1es.
A wipe $amp''e is taken by Mr'st apply-nq a suitable so"
as isooctanej to a piece of 11 cm filter paper (e.g., Whatman 40 ashless or
Whatman bO smear tabs) or gauze pad. The moistened filter paper or gauze
pad is ther held with a pair of stainless steel forcep1"-- or rubber gioves and
-------
-------
6- 5 yeggtation Samp]inj
The sample design or visual observation may indicate that samples
of vegetation, such as tree leaves, bushes, and flowers, are required. In
this case, the sample may be taken with pruning shears, a saw, or other suit-
able tool, and placed in a precleaned glass bottle, which should be capped,
labeled, affixed with a yellow TSCA PCS mark, and placed in an ice chest.
If samples are to be analyzed soon, the cold storage requirements may be re-
laxed as long as sample integrity is maintained. The sample collection data
should be entered into the field log book and on the chain-of-custody form.
After each sample is taken, the pruning shears should be rinsed with
solvent and wiped with a disposable wipe cloth to prevent cross-contamination
between samples. Also, rubber gloves and wipe cloths should be discarded into
a plastic disposal bag intended for PCB-contaminated materials.
6,6 Compositing Strategies
Compositing is the pooling of several samples to form one sample
for chemical analysis. In many circumstances it may be desirable to com-
posite samples to reduce the number of (often costly) analyses needed. The
suggested strategies for compositing samples are given in the appendix.
7.0 QUALITY ASSURANCE
Quality assurance must be applied throughout the entire sampling
program, including sample design and sample collection, handling, and preser-
vation. Each EPA office must develop a quality assurance plan (QAP) accord-
ing to EPA guidelines (USEPA 1980). The QAP must be submitted to the re-
gional QA officer or other appropriate QA official for approval prior to
sampling PCB spill sites.
35
-------
The elements of a QAP (USEPA 1980) include:
Title page
Table of contents
Project description
Project organization and responsibility
QA objectives for measurement data in terms of precision, accuracy,
completeness, representativeness, and comparability
Sampling procedures
Sample tracking and traceability
Calibration procedures and frequency
Analytical procedures
Data reduction, validation, and reporting
Internal quality control checks
Performance and system audits
Preventive maintenace
Specific routine procedures used to assess data precision,
accuracy, and completeness
Corrective action
Quality assurance reports to management
Each EPA inspector who will sample PCB spill sites should understand
and conform with all elements of the QAP,
8,0 QUALITY
Each EPA office that samples PCB spill sites must operate a formal
quality control (QC) program. The minimum requirements of this program con-
sist of preparing field blanks for the laboratory; sampling without contam-
ination of samples; maintaining a rigid chain-of-custody procedure for the
samples; and fully documenting the entire sampling program and maintaining
records of the documentation.
36
-------
The quality control measures taken by each EPA office should be
stipulated in the QA plan. The QC measures discussed below are given as ex-
amples only. EPA offices must decide which of the following measures, and
additional measures, will be required for each situation.
8.1 Field _Bjanks
Field blanks are given to the laboratory to demonstrate that the
sampling equipment has not been contaminated, A field blank may be generated
by using the sampling equipment to obtain a clean sample of solids or water.
for example, the scoop or soil coring device can be used to obtain a clean
solids blank sample. The water sampling equipment can be used to collect a
blank sample using laboratory reagent grade water. These field blanks should
be obtained both before and after field sampling.
Field blanks for wipe samples should be obtained in the field by
wetting a clean filter paper with the solvent and storing the wetted paper in
a clean sample jar.
One empty glass sample bottle and one filled with solvent should
also be given to the laboratory as field blanks,
8.2 Sampling Without Contamination
Samples collected from PCB spill sites which have been cleaned up
may become contaminated in two ways: (a) dirty sample containers, and (b)
cross-contamination of samples from the use of contaminated sampling equip-
ment. The first type of contamination can be eliminated by properly pre-
cleaning all sample containers prior to making the sampling trip. All glass
jars should be washed with soap and water, rinsed three times with distilled
water, rinsed with solvent (isooctane is recommended), baked in an oven at
350°C for 1 h, and sealed with a Teflon-lined cap. All aluminum foil used
should be rinsed with solvent.
37
-------
The sampling equipment should be precleaned before the site visit
by rinsing with solvent and thoroughly wiping the equipment down. Cross-
contamination during sampling can be avoided by using a separate sampler (such
as a scoop, spatula, corer, etc.) for each sample, or cleaning the sample
equipment between samples. Methods that can be used to clean the equipment
between samples are given in the sample collection, handling, and preservation
discussion (Section 6,0).
8.3 SampleCustody
As part of the quality assurance plan, the chain-of-custody proto-
col must be described. A chain-of-custody provides defensible proof of the
sample, and data integrity. The less rigorous sample traceability documenta-
tion merely provides a record of when operations were performed, and by whom.
Sample traceability is not acceptable for enforcement activities.
Chain-of-custody is required for analyses which may result in legal
proceedings, and when the data must be subject to legal scrutiny. Chain-of-
custody provides conclusive written proof that samples are taken, transferred,
prepared, and analyzed in an unbroken line as a means to maintain sample in-
tegrity. A sample is in custody if:
- It is in the possession of an authorized individual.
It is in the field of vision of an authorized individual.
It is in a designated secure area.
It has been placed in a locked container by an authorized
individual.
A typical chain-of-custody protocol contains the following elements:
1. Unique sample identification numbers.
38
-------
-------
9.0 DOCUMENTATION AND
Each EPA office is responsible for preparing and maintaining com-
plete records of the field sampling operations. A detailed documentation plan
should be prepared as a. part of the QAP , and should be strictly followed. The
following written records should be maintained for each field sampling opera-
tion:
Equipment preparation log book
Sample codes
Field log book
Site description forms
thai n~of -custody forms
Sample analysis request forms
Field trip report
9. 1 Equipment Preparation Log Book
A log book should be maintained which lists the sampling equipment
taken to each spill site. A detailed description of the cleaning and prepara-
tion procedures used for the sample collection equipment (templates, scoops,
glass bottle, etc.) should be recorded.
9. 2
Each sample should be assigned a unique sample code and labeled
accordingly when collected. The sample code should contain information on
the site and which sampling point the sample represents. This sample code
must, be used to identify all sample records,
Each sample must also be labeled with a yellow TSCA PCB mark as
described in 40 CFR 761.45 until it is determined to be PCB free.
40
-------
9.3 Field Log Book
The EPA inspector should maintain a field log book which contains
all information pertinent to the field sampling program. The notebook should
be bound and entries be made in ink by the field inspector. All entries
should be signed by the inspector.
At a minimum, the log book should include the following entries;
Owner of spi11 site
Location of spill site
Date(s) of sample collection
Exact times of sample collection
Type of samples taken and sample identification numbers
Number of samples taken
Description of sampling methodology
Field observations
Name and address of field contact
Cross-reference of sample identification numbers to grid sample
points (shown on site description forms)
Since sampling situations will vary widely, no specific guidelines
can be given as to the extent of information which should be entered into the
field log book. Enough information should be recorded, however, so that some-
one can reconstruct the sampling program in the absence of the field inspector.
The field log book should be maintained in a secure place.
9.4 Site Description Forms
Serialized site description forms should be used to record the con-
ditions of the site, provide sketches of the site, and show the location of
the grid sampling points. The grid sampling points should be shown on di-
mensioned drawings and numbered. These forms should be accompanied by
41
-------
photographs (preferably Polaroid-type photographs) of the site. Each form
and photograph should be signed and dated by the EPA inspector,
9, 5 Chain-of-C us tody Fo rms
Chain-of-custody forms should be completed and accompany the samples.
These forms should contain the following information:
Project site
Sample identification number
Date and time of sample collection
Location of sample site
Type of sample (soil, water, etc.)
Signature of sample collector
Signatures of those who relinquish and those who receive the
samples, and date and time that samples change possession
Inclusive dates of possession
9. 6 Sajti£j_e_and_Anii'j_ -\_^_^^< . j e__s tFgrtns
A sample analysis request form should accompany the samples de-
livered to the laboratory. The field inspector should enter the following
information on the form:
Project si te
Name of sample collector
Sample identification numbers
Types of samples (soil, water, etc.)
Location of sample site for each sample
Analysis requested [analyte (i.e., total PCBs), method, desired
method detection limit, etc.]
QC requirements (replicates, lab blanks, lab spikes, etc.)
Special handling and storage requirements
42
-------
The laboratory personnel receiving the samples should enter the
following information on the form:
Name of person receiving the samples
Laboratory sample numbers
Date of sample receipt
Sample allocation
Analyses to be performed
9.7 Field Trip Report
The EPA inspector should prepare a brief field trip report to be
maintained on file. The report should provide information such as the proj-
ect site, date(s) of sampling, types and number of samples collected, any
problems encountered, any notable events, and specific reference to the other
documents listed above.
10.0 VALIDATION OF THE MANUAL
A previous draft of this manual entitled "Field Manual for Verifi-
cation of PCB Spill Cleanup" (Draft Interim Report No, 3, Task 37, EPA Prime
Contract No. 68-02-3938, June 27, 1985) was used in a brief field validation
study. The primary purposes of the study were to: (1) determine the degree
of difficulty of understanding the grid sampling designs in the field manual;
(2) determine the amount of time and degree of difficulty required to lay out
the sampling grids on simulated PCB spill sites; and (3) identify any concerns
or problems that may arise in implementing the field manual. To achieve these
goals, simulated PCB spill sites were constructed for the exercise. Four per-
sons (Mr, David Phillippi and Mr. Robert Jackson of the EPA Region VII Office
and Ms. Joan Westbrook and Mr. Ted Harrison of MRI) were selected to lay out
the sampling grids on the spill sites after they had read the field manual.
These four persons had no prior association with developing the field manual.
Other persons from EPA and MRI acted as observers since they were intimately
familiar with the field manual.
43
-------
Four simulated spill sites having the following characteristics
were laid out:
A rectangle (3 ft x 6 ft)
A parallelogram (about 3 ft on a side)
A circle (about 12 ft diameter)
A square (6 ft on a side)
The first two sites required seven grid sample points, and the other two re-
quired 19 grid sample points.
Each of the four "inspectors" laid out the grid sample points on
two of the four sites after constructing the designs on graph paper. In all
cases the sample points were laid out correctly with little or no difficulty
in 30 min or less. Each inspector commented that there was little or no dif-
ficulty in performing the exercises.
As a final exercise, a "large irregular simulated PCB spill site
was constructed, and all attendees participated in laying out the 37 grid
sample points. The spill site was designed so that some sample points were
located on the floor and two adjacent walls to make the exercise relatively
difficult. The 37 grid sample points were laid out correctly with relative
ease in about 45 min. Some discussions were required to decide how to treat
sampling points which fell in the overlap where the two walls intersected.
It was concluded from the exercise and discussions which followed
that: (1) the field manual is easy to follow and understood by people un-
familiar with the manual prior to reading it; (2) the grid sample points are
never "perfectly" laid out (with the sample points precisely aligned) so that,
some degree of randomness is built into the sample designs; (3) the time re-
quired to lay out the grid sample points after the boundaries of the spill
site have been determined is relatively short (less than 1 h); and (4) using
this manual, the grid sample points can be correctly laid out by inexperienced
people.
44
-------
11.0
BA, Erickson MO, Swanson SA, Kelso GL, Cox DC, Schultz BO.
(August), Verification of PCB spill cleanup by sampling analysis (second
printing). Interim report. Washington, DC: Office of Toxic Substances, U.S.
Environmental Protection Agency. EPA-560/5-85-026.
Mason BJ. 1982 (October). Preparation of soil sampling protocol: tech-
niques and strategies. ETHURA, McLean, VA, under subcontract to Environ-
mental Research Center, University of Nevada, for U.S. Environmental
Protection Agency, Las Vegas.
USEPA, 1980. U.S. Environmental Protection Agency. Guidelines specifi-
cations for preparing quality assurance project plans. Office of Monitoring
Systems and Quality Assurance, QAMS-005/80.
USEPA, 1981 (March), U.S. Environmental Protection Agency. TSCA Inspection
Manual.
45
-------
APPENDIX
A-l
-------
APPENDIX
This appendix gives suggested strategies for compositing samples
taken from PCB spill sites which are sampled using the grid sampling methods
described in the text of the report. Compositing may result in a savings of
analysis time and cost. Sample compositing is not required and should be
used only if time or cost savings may result. The strategies for forming
composites are as follows:
1. Composite only samples of the same type (i.e., all soil or all
water). Since the composite must be thoroughly mixed to ensure homogeneity,
certain types of samples such as asphalt, wipe samples, wood samples and
other hard-to-mix matrices should not be composited.
2. Do not form a composite with more than 10 samples, since in
some situations compositing a greater number of samples may lead to such low
PCB levels in the composite that the recommended analytical method approaches
its limit of detection and becomes less reliable.
3. for each type of sample, determine the number of composites to
be formed using the table below.
Number of samples Number of composites
2-10 1
11-20 2
21-30 3
31-37 4
As much as possible, try to form composites of equal size. For
example, if 37 soil samples are taken, then four composites could be formed
using 9, 9, 9, and 10 samples apiece.
A-2
-------
4. To the extent possible, composite adjacent samples. If resi-
dual contamination is present, it is likely that, high PCB levels will be
found in some samples taken close together.
Because of the large number of situations that be encountered
in practice, it is not possible to specify compositing strategies more pre-
cisely. The laboratory field staff should exercise judgment in all case?
ft-3
------- |