-------
u
c
1
1
1
1
o-l
"I
• 1
?!
1
\
I
\
\
\
\
\
1
n c
M P
1
1
1
*l
Ull
°l
\ *
\
\
\
\
\
\
\
\
1
5 U
4 r
I
\
\
\
\
[^
^
\\
\\
*
\
1
•) e
- »
LU
Q
<
CC
U
m
>
o
m
<
0.
LU
UJ
Q
in
CM
1
\
\
\
\
\
\
\
1
3. U
I
1
•J
00
CO
oc
o
UJ
H
co
UJ
G
<
CC
a
LLJ
CC
O
LL
LU
CO
X
a.
LU
Q
Q
Z
CC
O
u.
w
CC
o
00
o
UJ
oc
oc
o
u
CO
a
UJ
OC
D
a
D-6
-------
\
\
\
\
\
1
u
1
c
o
I
I
1
t a.
I uJ
\ o
*\ S
fi\
' t •
\\
*\
%,.
\
\
\
\
\
\
\
\
Q
00
1
3 C
i c
|
1 c
1 a
1 2
i :
' a.1 u
1 UJ| U
w| o
A 11 '
1 i
2 -51
ll I1
ccl cc|
T i
' •
\ '
\* i
1 <
» ;
i
\'
* 11
*
\ \
^ \
^
\ \
\ i
\
\
\i
1
•
i
y e
>i '
1 J
D '
i 1
S I
|
J a.!
J u|
" ^J
»
V)t
o1
II
^1
LLJ.
Si
sj ^
I •
H I
3 ' 1
71 1
§ 1 /
il ! '
i /
/ •
1 /
1 '
1 .'
1 f
1 .
f 1
• t
II
II
II
1
\
1
S C
- c
/
/
f
/
1
J
S1
/
'Ft*
U
1
5 C
M C
1
1
S
•i
00
CC
UJ
>
UJ
O
UJ
a
z
o
CC
<
O
O
o
O
a
z
UJ
>
01
U
<
U.
cc
CO
CC
O
LL
CO
CC
e
u
g
t-
u
LU
oc
CC
o
u
LU
CC
D
a
D-7
-------
The uncorrected migration distance from the solid waste limit can then be
read on the left for the appropriate site age and soil type.
If the soil along a given boundary is stratified and the variability extends from
the waste deposit to the property boundary, the most permeable unsaturated
thickness should be used in entering the charts. For example, if dry, clean sand
underliessurficial silty clays, the uncorrected migration distance should be obtained
using the sand line of the chart. If there are questions as to the extent of particular
soils along a boundary, helpful information might be obtained from Soil
Conservation Services (SCS) Soil Survey Maps or the landfill operator. Field
inspection, SCS maps, and permit boring information are sufficient. Additional
borings are not necessary as this is only a ranking procedure. Where there is doubt,
use the most permeable soil group present.
For the example landfill in Figure D-1, the uncorrected 5 percent methane
migration distances for a 10-year old landfill would be (Figure C-2):
Section A-A: East side, 10 years, sand = 165'
West side, 10 years, sand = 165'
Section B-B: South side, 10 years, sand = 165'
North side, 10 years, clay = 130'
The corresponding uncorrected distances for the 1.25 percent methane
migration would be:
Section A-A: East side, 10 years, sand = 225'
West side, 10 years, sand = 135'
Section B-B: South side, 10 years, sand = 255'
North side, 10 years, clay = 200'
The depth to corrective mulitpliers for the example sites would be:
Section A-A: East side, 10 years, 20'deep » 1.0
West side, 10 years, 20'deep = 1.0
D-8
-------
Section B-B: South side, 10 years, 10' deep = 0.95
North side, 10 years, 50'deep = 1.4
VENTING CONDITIONS CORRECTION
The corrective factors for the surrounding soil venting conditions are obtained
using the chart in Figure D-4. This chart is based on the assumption that the
surrounding surficial soil is impervious 100 percent of the time. Thus, the value read
from the chart must be adjusted, based on the percentage of time the surrounding
surficial soil is saturated or frozen and the percentage of land along the path of gas
migration from which gas venting to the atmosphere is blocked all year (asphalt or
concrete roads or parking lots, shallow perched ground water, surface water bodies
not interconnected to ground water). The totally impervious corrective factor is
only used when the landfill is entirely surrounded at all times by these conditions.
Both time and area adjustments are necessary, and the percentages are additive.
Estimates to the nearest 20 percent are sufficient. An adjusted corrective factor is
obtained by entering the chart with site age and obtaining the totally impervious
corrective factor for the appropriate depth and soil type and then entering this
value in the following equation :
Adjusted corrected factor = [(Impervious corrective factor)-1)]
x [5 of impervious time or area] + 1
When free venting conditions are prevalent most of the year, simply use 1.0
(no correction). For depths less than 25 feet deep, use the 25 foot value. For the
example site, the adjusted corrective factors for frozen or wet soil conditions 50
percent of the year are:
Section A-A: East side (ignore narrow = (2.1-1)(0.50) + 1 a 1.55
road, sand 20'deep,
10 years old)
West side (sand 20'deep, = (2.1-1)(0.50) + 1 = 1.55
10 years old)
D-9
-------
Section B-B: South side (sand, 10 deep, = (2.1-1)(0.50) + 1 = 1.55
10 years old)
North side (clay, 50'deep, = (1.4-1)(0.50) + 1 = 1.2
10 years old)
Once the surface venting factors have been tabulated as in Table D-1, the
corrective distance can be obtained by multiplying across the chart for each side of
the landfill. These values can then be plotted on the scale plan to describe contours
of the 5 percent and 1.25 percent methane concentrations or simply compared to
the distance from the waste deposit to structures of concern (Figure D-5).
D-10
-------
a:
O
Z
H-
<
CO
h-
LU
U
Z
t-
5
LU
Z
X
r-
LU
u
c
•M
I/I
Q
•o
^2
u
^
O
C °*
'•§ c
II
o ^**
'^3 C^
u
O
X
in
in
rsi
^
in
rsi
00
r—
rsi
II
rsi
X
^
X
o
*~
^
in
Z
VO
on
on
II
rsi
X
^.
X
o
0
rsi
SC3
in
rsi
c
O
o E
•^ 3
(U C
"•» E
C'x
0)
^
2 1,
nj O>
a;
O»
s
D-11
-------
cs
Houses
1.25%
NOTE: NOT TO SCALE
FIGURE D-5. EXAMPLE LANDFILL METHANE CONDITIONS
D-12
-------
APPENDIX E
ESTIMATION OF BASEMENT AIR CONTAMINANT CONCENTRATIONS DUE TO
VOLATILE COMPONENTS IN GROUND WATER SEEPED INTO THE BASEMENT
E-1
-------
APPENDIX E
ESTIMATION OF BASEMENT AIR CONTAMINANT CONCENTRATIONS DUE TO
VOLATILE COMPONENTS IN GROUND WATER SEEPED INTO THE BASEMENT
Ground water can reach the basement and the walls of a house in several
ways. If ground water is contaminated by volatile components, there are several
possibilities that the indoor ambient air can be affected by these constituents.
There are several methods which can be applied to estimating the ambient air
concentrations in the basement into which the contaminants are volatilized from
ground water. The manner in which and the extent to which the ground water
reaches the basement or the walls will dictate the choice of a method.
Two cases are considered as example scenarios: Case 1) Ground water is
seeped inside the basement completely wetting the basement, with a visual
indication of water on the floor. Case 2) The basement is partially wetted without
a visual indication of liquid on the floor. This latter case can be subdivided into two
subcases: Subcase 1) involving a damp floor evident on the surface; Subcase 2)
involving a floor without observable dampness on the floor surface but with ground
water underneath the concrete floor.
The way the emission rates are estimated will be different for the three cases.
If the emission flux rate per unit square area of the exposed surface is denoted by E
(g/m2 day), then in all cases the air concentration, C (ug/m3), in the basement can be
estimated from:
C(ug/m3) = E
where A a basement floor and wall area exposed to ground water, m2
VB = volume of the basement, m3, and
te - air exchange time for the basement, days.
E-2
-------
The air exchange time should be determined on a site-specific or situation-
specific basis. The tight room will have a longer time per air exchange in the room,
and the room with an exhaust fan will have a shorter time per air exchange. The
default value for a typical house could bete = 0.05 days.
The emission rates in Eq. (1) can be estimated for the various case scenarios
illustrated above.
Case 1. Wet basement with visible liquid.
The volatilization is a mass transfer phenomenon from the liquid phase of
ground water on the floor to the basement air. Emission flux rate can be estimated
from:
E = KOL(CL-CL*) (2)
where KQL = overall mass transfer coefficient in the liquid phase unit, m/day, C|_ =
concentration of contaminant in water, g/m3, and CL* = liquid phase concentration
in equilibrium concentration with the basement air, g/m3. The equilibrium
concentration C* could be assumed to be approaching a small value compared to
the ground water contaminant concentration when the air exchange rate is high, or
when the time per air exchange is small. But this assumption would not be valid at a
low air exchange rate or at a longer time for a room air exchange. In this case, the
emission flux rate should be estimated by a trial and error method using Equation
(2) in combination with Equation (1), and Henry's Law constant.
It is a well-established scientific principle to use the two-resistance theory to
obtain the overall mass transfer coefficient, KQL, as follows:
Hck
(3)
where kj. and kg = individual mass transfer coefficients in liquid and gas phases,
respectively, m/day, and HC = dimensionless Henry's Law constant obtained from
E-3
-------
concentration units for gas and liquid phase concentrations. The numerical value
for HC can be calculated from Henry's Law constant given in atm/g-mol.m3 by
multiplying by 41. Default values for the individual mass transfer coefficients can be
estimated from:
44 >
MW
i!i
i »
I 100
M
cm
kL = 3 cm I 44 \ « / 24 M hr
hr I MW | \ 100 cm day
1
t2
MW J I 100 cm day | (5)
cm / 18 \ 2 / 24 M hr
where MW = molecular weight of the contaminant.
Case 2. Basement partially wetted with no visual indication of liquid.
(a) Subcase 1. Dampness evident on the floor or wall surface. The
volatilization process can be treated as a diffusional process from the air at the
water-air interface through the air pores in the basement floor material and into
the basement air. The diffusional process can be solved using the approach
described in the EPA report Development of Advisory Levels for Polychlorinated
Biphenvls (PCBs) Cleanup (PB86-232774). The final result needed for emission flux
estimation would be:
2e Dej u
H
E-4
-------
where e = porosity of the floor material, Dej = effective diffusivity in the air pores
( = Dj e 1/3), m2/day, Dj = molecular diffusivity, m2/day, T = averaging time, days,
and a = Dei e/(e + (1-e))/Hc. If steady state conditions are achieved as a result of a
continuous supply of contaminated water to the floor surface, it may be more
appropriate to treat the emission rate problem using Eq. (2) rather than Eq. (6).
(b) Subcase 2. No dampness evident on the floor or wall surface but ground
water underneath the basement or wall material. Diffusion through the air space
of the floor or wall material will result in a slow release of volatile contaminants
from ground water to the basement air. The steady state flux rate can be estimated
from:
= D, ,4/3
where h = thickness of the barrier between the surface of ground water and the
air-basement floor interface, m. When the basement air concentration is small
compared to the HcCt term in Eq. (7), the C term can be ignored in estimating e from
Eq. (7). Otherwise Eq. (7) should be solved along with Eq. (1) requiring a trial and
error solution.
E-5
-------
APPENDIX F
METHOD 1312: SYNTHETIC PRECIPITATION
LEACH TEST FOR SOILS
-------
METHOD 1312
SYNTHETIC PRECIPITATION LEACH TEST FOR SOILS
1.0 SCOPE AND APPLICATION
1.1 Method 1312 is designed to determine the mobility of
both organic and inorganic contaminants present in soils.
1.2 If a total analysis of the soil demonstrates that in-
dividual contaminants are not present in the soil, or that they
are present but at such low concentrations that the appropriate
regulatory thresholds could not possibly be exceeded, Method
1312 need not be run.
2.0 SUMMARY OF METHOD
2.1 The particle size of the soil is reduced (if necessary)
and is extracted with an amount of extraction fluid egual to 20
times the weight of the soil. The extraction fluid employed is
a function of the region of the country where the soil site is
located. A special extractor vessel is used when testing for
volatiles. Following extraction, the liguid extract is separated
from the soil by 0.6-0.8 urn glass fiber filter.
3.0 INTERFERENCES
3.1 Potential interferences that may be encountered during
analysis are discussed in the individual analytical methods.
4.0 APPARATUS AND MATERIALS
4.1 Agitation apparatus - an acceptable agitation apparatus
is one which is capable of rotating the extraction vessel in an
end-over-end fashion at 30 + 2 rpm (see Figure 1). Suitable
devices known to EPA are identified in Table 2.
4.2 Extraction vessel - acceptable extraction vessels are
those that are listed below:
4.2.1 Zero Headspace Extraction Vessel (ZHE) - This
device is for use only when the soil is being tested for the
mobility of volatile constituents (see Table 1). The ZHE is an
extraction vessel that allows for liguid/solid separation within
the device and which effectively precludes headspace (as depicted
in Figure 3). This type of vessel allows for initial liguid/soli
separation, extraction, and final extract filtration without
having to open the vessel (see Step 4.3.1). These vessels shall
have an internal volume of 500 to 600 mL and be eguipped to
accommodate a 90-mm filter. Suitable ZHE devices known to EPA
are identified in Table 3. These devices contain viton 0-rings
which should be replaced frequently. For the ZHE to be acceptabl
for use, the piston within the ZHE should be able to be moved
1312-1 Revision 0
December 1988
-------
with approximately 15 psi or less. If it takes more pressure
to move the piston, the 0-rinqs in the device should be replaced.
If this does not solve the problem, the ZHE is unacceptable for
1312 analyses and the manufacturer should be contacted. The ZHE
should be checked after every extraction. If the device con-
tains a built-in pressure gauge, pressurize the device to
50 psi, allow it to stand unattended for 1 hour, and recheck
the pressure. If the device does not have a built-in pressure
gauge, pressurize the device to 50 psi, submerge it in water
and check for the presence of air bubbles escaping from any
of the fittings. If pressure is lost, check all fittings and
inspect and replace 0-rings, if necessary. Retest the device.
If leakage problems cannot be solved, the manufacturer should
be contacted.
4.2.2 When the soil is being evaluated for other than
volatile contaminants, an extraction vessel that does not pre-
clude headspace (e.g. a 2-liter bottle) is used. Suitable
extraction vessels include bottles made from various materials,
depending on the contaminants to be analyzed and the nature of the
waste (see Step 4.3.3). It is recommended that borosilicate
glass bottles be used over other types of glass, especially
when inorganics are of concern. Plastic bottles may be used
only if inorganics are to be investigated. Bottles are available
from a number of laboratory suppliers. When this type of ex-
traction vessel is used, the filtration device discussed in
Step 4.3.2 is used for initial liguid/solid separation and final
extract filtration.
4.2.3 Some ZHEs use gas pressure to actuate the ZHE piston,
while others use mechanical pressure (see Table 3). Whereas
the volatiles procedure (see Step 7.4) refers to pounds-per-
sguare inch (psi), for the mechanically actuated piston, the
pressure applied is measured in torque-inch-pounds. Refer to
the manufacturer's instuctions as to the proper conversion.
4.3 Filtration devices - It is recommended that all filtrations
be performed in a hood.
4.3.1 Zero-Headspace Extractor Vessel (see Figure 3) -
When the waste is being evaluated for volatiles, the zero-
headspace extraction vessel is used for filtration. The device
shall be capable of supporting and keeping in place the fiber
filter, and be able to withstand the pressure needed to accomplish
separation (.50 psi).
NOTE; When is it suspected that the glass fiber filter
has been ruptured, an in-line glass fiber filter may be
used to filter the material within the ZHE.
4.3.2 Filter holder - when the soil is being evaluated
for other than volatile compounds, a filter holder capable of
1312-2 Revision 0
December 1988
-------
supporting a glass fiber filter and able to withstand 50 psi
or more of pressure shall be used. These devices shall have a
minimum internal volume of 300 mL and be equipped to accomodate
a minimum filter size of 47 mm (filter holders having an
internal capacity of 1.5 liters or greater are recommended).
4.3.3 Materials of construction - filtration devices shall
be made of inert materials which will not leach or absorb soil
components. Glass, polytetrafluoroethylene (PTFE) or type 316
stainless steel equipment may be used when evaluating the mcbilit
of both organic and inorganic components. Devices made of nigh
density polyethylene (HOPE), polypropylene, or polyvinyl chloride
may be used only when evaluating the mobility of metals. Boro-
silicate glass bottles are recommended for use over other types
of glass bottles, especially when inorganics are constituents
of concern.
4.4 Filters - filters shall be made of borosilicate glass
fiber, shall have an effective pore size of 0.6 - 0.8 urn and
shall contain no binder materials. Filters known to EPA to meet
these requirements are identified in Table 5. When evaluating the
mobility of metals, filters should be acid-washed prior to use
by rinsing with 1.ON nitric acid followed by three consecutive rinses
with deionized distilled water (a minimum of 1-liter per rinse is
recommended). Glass fiber filters are fragile and should be handled
with care.
4.5 pH meters - any of the commmonly available pH meters are
acceptable.
4.6 ZHE extract collection devices - TEDLAR bags, glass, stain-
less steel or PTFE gas tight syringes are used to collect the volatili
extract.
4.7 Laboratory balance - any laboratory balance accurate to
within + 0.01 g may be used (all weight measurements are to be within
+_ 0.1 g).
4.8 ZHE extraction fluid transfer devices - any device capable
of transferring the extraction fluid into the ZHE without changing
the nature of the extraction fluid is recommended.
5.0 REAGENTS
5.1 Reagent water-- reagent water is defined as water in
which an interferent is not observed at or above the method
detection limit of the analyte(s) of interest. For non-volatile
extractions, ASTM Type II water, or equivalent meets the definition
of reagent water. For volatile extractions, it is recommended
that reagent water be generated by any of the following methods.
Reagent water should be monitored periodically for impurities.
1312-3 . Revision 0
December 1988
-------
5.1.1 Reagent water for volatile extractions may be
generated by passinq tap water through a carbon filter bed
containing about 500 g of activated carbon (Calgon Corp.,
Filtrasorb 300 or equivalent).
5.1.2 A water purification system (Millipore Super-Q or
equivalent) may also be used to generate reagent water for
volatile extractions.
5.1.3 Reagent water for volatile extractions may also
be prepared by boiling water for 15 minutes. Subsequently,
while maintaining the water temperature at 90 + 5°C, bubble
a contaminant-free inert gas (e.g. nitrogen) through the
water for 1 hour. While still hot, transfer the water to a
narrow-mouth screw-cap bottle under zero headspace and seal
with a Teflon lined septum and cap.
5.2 Sulfuric acid/nitric acid (60/40 weight percent mixture)
H2S04/HN03. Cautiously mix 60 g of concentrated sulfuric acid with
40 g of concentrated nitric acid.
5.3 Extraction fluids:
5.3.1 Extraction fluid #1 - this fluid is made by adding
the 60/40 weight percent mixture of sulfuric and nitric acids
to reagent water until the pH is 4.20 + 0.05.
5.3.2 Extraction fluid #2 - this fluid is made by adding
the 60/40 weight percent mixture of sulfuric and nitric acids
to reagent water until the pH is 5.00 + 0.05.
5.3.3 Extraction fluid #3 - this fluid is reagent water
(ASTM Type II water, or equivalent) used to determine cyanide
leachability.
Note; It is suggested that these extxraction fluids be moni-
tored frequently for impurities. The pH should be
checked prior to use to ensure that these fluids are
made up accurately.
5.4 Analytical standards shall be prepared according to the
appropriate analytical method.
6.0 SAMPLE COLLECTION, PRESERVATION, AND HANDLING
6.1 All samples shall be collected using an appropriate
sampling plan.
6.2 At least two separate representative samples of a soil
should be collected. The first sample is used to determine if the
soil requires particle-size reduction and, if desired, the percent
solids of the soil. The second sample is used for extraction
of volatiles and non-volatiles.
1312-4 Revision 0
December 1988
-------
6.3 Preservatives shall not be added to samples.
6.4 Samples shall be refrigerated to minimize loss of volatile
organics and to retard biological activity.
6.5 When the soil is to be evaluated for volatile contaminants,
care should be taken to minimize the loss of volatiles. Samples
shall be taken and stored in a manner to prevent the loss of
volatile contaminants. If possible, it is recommended that any
necessary particle-size reduction be conducted as the sample is
being taken.
6.6. 1312 extracts should be prepared for analysis and
analyzed as soon as possible following extraction. If they need
to be stored, even for a short period of time, storage shall be at
4°C, and samples for volatiles analysis shall not be allowed to
come into contact with the atmosphere (i.e. no headspace). See
Section 8.0 (Quality Control) for acceptable sample and extract
holding times.
7.0 PROCEDURE
7.1 The preliminary 1312 evaluations are performed on a mini-
mum 100 g representative sample of soil that will not actually under-
go 1312 extraction (designated as the first sample in Step 6.2).
7.1.1 Determine whether the soil requires particle-size
reduction. If the soil passes through a 9.5 mm (0.375-inch)
standard sieve, particle-size reduction is not required
(proceed to Step 7.2). If portions of the sample do not
pass through the sieve, then the oversize portion of the
soil will have to be prepared for extraction by crushing
the soil to pass the 9.5 mm sieve.
7.1.2 Determine the percent solids if desired.
7.2 Procedure when volatiles are not involved - Enough
solids should be generated for extraction such that the volume
of 1312 extract will be sufficient to support all of the analyses
required. However, a minimum sample size of 100 grams shall
be used. If the amount of extract generated by a single 1312
extract will not be sufficient to perform all of the analyses,
it is recommended that more than one extraction be performed and
the extracts be combined and then aliquoted for analysis.
7.2.1 Weigh out a representative subsample of the soil and
transfer to the filter holder extractor vessel.
«
7.2.2 Determine the appropriate extraction fluid to use.
If the soil is from a site that is east of the Mississippi
River, extraction fluid #1 should be used. If the soil is
from a site that is west of the Mississippi River, extraction
fluid #2 should be used. If the soil is to be tested for
cyanide leachability, extraction fluid #3 should be used.
1312-5 Revision 0
December 1988
-------
Note: Extraction fluid #3 (reagent water) must be used
when evaluating cyanide-containing soils because leaching
of cyanide-containing soils under acidic conditions may
result in the formation of hydrogen cyanide gas.
7.2.3 Determine the amount of extraction fluid to add
based on the following formula:
amount of extraction fluid (mL) = 20 x weight of soil (g)
Slowly add the amount of appropriate extraction fluid to the
extractor vessel. Close the extractor bottle tightly (it
is recommended that Teflon tape be used to ensure a tight
seal), secure in rotary extractor device, and rotate at 30
+_ 2 rpm for 18 ^+ 2 hours. Ambient temperature (i.e. temper-
ature of room in which extraction is to take place) shall
be maintained at 22 _+ 3°C during the extraction period.
Note; As agitation continues, pressure may build up within the
extractor bottle for some types of soil (e.g. limed or
calcium carbonate containing soil may evolve gases such as
carbon dioxide). To relieve excess pressure, the extractor
bottle may be periodically opened (e.g. after 15 minutes,
30 minutes, and 1 hour) and vented into a hood.
7.2.4 Following the 18 _+ 2 hour extraction, the material in
the extractor vessel is separated into its component liquid and
solid phases by filtering through a glass fiber filter.
7.2.5 Following collection of the 1312 extract it is re-
commended that the pH of the extract be recorded. The extract
should be immediately aliquoted for analysis and properly
preserved (metals aliquots must be acidified with nitric
acid to pH < 2; all other aliquots must be stored under
refrigeration (4°C) until analyzed). The 1312 extract
shall be prepared and analyzed according to appropriate
analytical methods. 1312 extracts to be analyzed for metals,
other than mercury, shall be acid digested.
7.2.6 The contaminant concentrations in the 1312 extract are
compared to thresholds in the clean closure guidance manual.
Refer to Section 8.0 for Quality Control requirements.
7.3 Procedure when volatiles are involved:
7.3.1 The ZHE device is used to obtain 1312 extracts for
volatile analysis only. Extract resulting from the use of the
ZHE shall not be used to evaluate the mobility of non-volatile
analytes (e.g. metals, pesticides, etc.). The ZHE device
has approximately a 500 mL internal capacity. Although a minimum
sample size of 100 g was required in the Step 7.2 procedure, the
ZHE can only accommodate a maximum of 25 g of solid , due to the
need to add an amount of extraction fluid equal to 20 times the
1312-6 Revision 0
December 1988
-------
weight of the soil. The ZHE is charged with sample only once and
the device is not opened until the final extract has been col-
lected. Although the following procedure allows for particle-
size reduction during the conduct of the procedure, this could
result in the loss of volatile compounds. If possible particle-
size reduction (see Step 7.1.1) should be conducted on the
sample as it is being taken (e.g./ particle-size may be reduced
by crumbling). If necessary particle-size reduction may be
conducted during the procedure. In carrying out the following
steps, do not allow the soil to be exposed to the atmosphere for
any more time than is absolutely necessary. Any manipulation of
these materials should be done when cold (4°C) to minimize the
loss of volatiles. Pre-weigh the evaculated container which
will receive the filtrate (see Step 4.6), and set aside. If
using a TEDLAR® bag, all air must be expressed from the device.
7.3.2 Place the ZHE piston within the body of the ZHE (it
may be helpful first to moisten the piston 0-rings slightly with
extraction fluid). Adjust the piston within the ZHE body to a
height that will minimize the distance the piston will have to
move once it is charged with sample. Secure the gas inlet/outlet
flange (bottom flange) onto the ZHE body in accordance with the
manufacturer's instructions. Secure the glass fiber filter
between the support screens and set aside. Set liquid inlet/out-
let flange (top flange) aside.
7.3.3 Quantitatively transfer 25 g of soil to the ZHE.
Secure the filter and support screens into the top flange of the
device and secure the top flange to the ZHE body in accordance
with the manufacturer's instructions. Tighten all ZHE fittings
and place the device in the vertical position (gas inlet/outlet
flange on the bottom). Do not attach the extraction collection
device to the top plate. Attach a gas line to the gas inlet/out-
let valve (bottom flange) and, with the liquid inlet/outlet
valve (top flange) open, begin applying gentle pressure of 1-10
psi to a maximum of 50 psi to force most of the headspace out of
the device.
7.3.4 With the ZHE in the vertical position, attach a
line from the extraction fluid reservoir to the liquid inlet/
outlet valve. The line used shall contain fresh extraction
fluid and should be preflushed with fluid to eliminate any air
pockets in the line. Release qas pressure on the ZHE piston
(from the gas inlet/outlet valve), open the liquid inlet/
outlet valve, and begin transferring extraction fluid (by
pumping or similar means) into the ZHE. Continue pumping
extraction fluid into the ZHE until the appropriate amount of
fluid has been introduced into the device.
7.3.5 After the extraction fluid has been added, immediate!;
close the inlet/outlet valve and disconnect the extraction fluid
line. Check the ZHE to ensure that all valves are in their clos<
positions. Physically rotate the device in an end-over-end fash
1312-7 Revision 0
December 1988
-------
2 or 3 times. Reposition the ZHE in the vertical position with
the liquid inlet/outlet valve on top. Put 5-10 psi behind the
piston (if nesessary) and slowly open the liquid inlet/outlet
valve to bleed out any headspace (into a hood) that may have
been introduced due to the addition of extraction fluid.
This bleedinq shall be done quickly and shall be stopped at the
first appearance of liquid from the valve. Re-pressurize the
ZHE with 5-10 psi and check all ZHE fittings to ensure that
they are closed.
7.3.6- Place the ZHE in the rotary extractor apparatus (if
it is not already there) and rotate the ZHE at 30 + 2 rpm for
18 4^ 2 hours. Ambient temperature (i.e. temperature of the room
in which extraction is to occur) shall be maintained at 22 +_ 3°C
during agitation.
7.3.7 Following the 18+2 hour agitation period, check
the pressure behind the ZHE piston by quickly opening and closing
the gas inlet/outlet valve and noting the escape of gas. If the
pressure has not been maintained (i.e. no gas release observed),
the device is leaking. Check the ZHE for leaking and redo the
extraction with a new sample of soil. If the pressure within
the device has been maintained, the material in the extractor
vessel is separated into its component liquid and solid phases.
7.3.8 Attach the evacuated pre-weighed filtrate collection
container to the liquid inlet/outlet valve and open the valve.
Begin applying gentle pressure of 1-10 psi to force the liquid
phase into the filtrate collection container. If no additional
liquid has passed through the filter in any 2 minute interval,
slowly increase the pressure in 10-psi increments to a maximum of
50 psi. After each incremental increase of 10 psi, if no additional
liquid has passed through the filter in any 2 minute interval,
proceed to the next 10 psi increment. When liquid flow has
ceased such that continued pressure filtration at 50 psi does
not result in any additional filtrate within any 2 minute period,
filtration is stopped. Close the inlet/outlet valve, discontinue
pressure to the piston, and disconnect the filtration collection
container.
NOTE; Instantaneous application of high pressure can
degrade the glass fiber filter and may cause
premature plugging.
7.3.9 Following collection of the 1312 extract, the extract
should be immediately aliquoted for analysis and stored with
minimal headspace at 4°C until analyzed. The 1312 extract will be
prepared and analyzed according to the appropriate analytical
ma t-hr>Hc .
8.0 QUALITY CONTROL
8.1 All data, including quality assurance data, should be
1312-8 Revision 0
December 1988
-------
maintained and available for reference or inspection.
8.2 A minimum of one blank (extraction fluid # 1) for every
10 extractions that have been conducted in an extraction vessel
shall be employed as a check to determine if any memory effects
from the extraction equipment are occurring.
8.3 For each analytical batch (up to twenty samples), it is
recommended that a matrix spike be performed. Addition of matrix
spikes should occur once the 1312 extract has been generated
(i.e. should not occur prior to performance of the 1312 procedure).
The purpose of the matrix spike is to monitor the adequacy of the
analytical methods used on the 1312 extract and for determining
if matrix interferences exist in analyte detection.
8.4 All quality control measures described in the appropriate
analytical methods shall be followed.
8.5 The method of standard addition shall be employed for
each analyte if: 1) recovery of the compound from the 1312
extract is not between 50 and 150%, or 2) if the concentration of
the constituent measured in the extract is within 20% of the
appropriate regulatory threshold. If more than one extraction is
being run on samples of the same waste (up to twenty samples),
the method of standard addition need be applied only once and the
percent recoveries applied on the remainder of the extractions.
8.6 Samples must undergo 1312 extraction within the following
time period after sample receipt: Volatiles, 14 days; Semi-
Volatiles, 40 days; Mercury, 28 days; and other Metals, 180 days.
1312 extracts shall be analyzed after generation and preservation
within the following periods: Volatiles, 14 days; Semi-Volatiles,
40 days; Mercury, 28 days; and other Metals, 180 days.
9.0 METHOD PERFORMANCE
9.1 None available.
10.0 REFERENCES
10. 1 None available.
1312-9 Revision 0
December 1988
-------
TABLE 1. — VOiJ^TILE CONTAMINANTS
Compounds
Ethyl ether.
CAS No.
67-64-1
107-13-1
71-43-2
71-36-6
75-15-0
56-23-5
108-90-7
67-66-3
107-06-2
75-35-4
141-78-6
100-41-4
60-29-7
78-83-1
67-56-1
75-09-2
78-93-3
108-10-1
630-20-6
79-34-5
127-18-4
108-88-3
71-55-6
79-00-5
79-01-6
75-69-4
76-13-1
75-01-7
1330-20-7
1312-10
Revision
December
0
1988
-------
TABLE 2. — SUITABLE ROTARY AGITATION APPARATUS
1
Company
Location
Model
Analytical Testing and
Consulting Services, Inc,
Associated Design and
Manufacturing Company
Environmental Machine
and Design, Inc.
IRA Machine Shop and
Laboratory
Lars Lande Manufacturing
Millipore Corp.
REXNORD
Warrington, PA
(215) 343-4490
Alexandria, VA
(703) 549-5999
Lynchburg, VA
(804) 845-6424
Santurce, PR
(809) 752-4004
Whitmore Lake, MI
(313) 449-4116
Bedford, MA
(800) 225-3384
Milwaukee, WI
(414) 643-2850
4-vessel device
4-vessel device,
6-vessel device
4-vessel device,
6-vessel device
16-vessel device
10-vessel device
5-vessel device
4-vessel ZHE devic
or 4-one litter
bottle extractor
device
6-vessel device
*Any device that rotates the extraction vessel in an end-over-end
fashion at 30 + 2 rpm is acceptable.
TABLE 3. — SUITABLE ZERO-HEADSPACE EXTRACTOR VESSELS
Company
Location
Model No.
Analytical Testing & Con-
sulting Services, Inc.
Associated Design & Manu-
facturing Co.
Lars Lande Mfg.
Millipore Corp.
Warrington, PA,
(215)~343-4490
Alexandria, VA
(703) 549-5999
Whitmore Lake, MI
(313) 449-4116
Bedford, MA,
(800) 225-3384
C102, Mechanical
Pressure Devio
3740-ZHB, Gas
Pressure Devici
Gas Pressure
Device
SD1 P581 C5, Ga
Pressure Devic
1312-11
Revision 0
December 1988
-------
TABLE 4. — SUITABLE ZHE FILTER HOLDERS1
Company
Micro Filtration Systems
Millipore Corp.
Nucleopore Corp.
Location
Dublin, CA
(415) 828-6010
Bedford, MA
(800) 225-3384
Pleasanton, CA
(800) 882-7711
Model
302400
YT30142HW
XX1004700
425910
410400
Size
142 mm
142 mm
47 mm
142 mm
47 mm
device capable of separating the liquid from the solid phase of
the soil is suitable, providing that it is chemically compatible with
the soil and the constitutents to be analyzed. Plastic devices (not
listed above) may be used when only inorganic contaminants are of con-
cern. The 142 mm size filter holder is recommended.
TABLE 5. — SUITABLE FILTER MEDIA
Company
Millipore Corp.
Nucleopore Corp.
Whatman Laboratory
Products, Inc.
Location
Bedford, MA
(800) 225-3384
Pleasanton, CA
(415) 463-2530
Clifton, NJ
(201) 773-5800
Model
AP40
211625
GFF
Sizei
0.7
0.7
0.7
^•Nominal pore size
1312-12
Revision 0
December 1988
-------
Figure 1. Rotary Agi tati
on
Motor , \
|L
(30 * 2 rpra),
" V
Extraction Vessel Holder
1312-13
Revision 0
December 1988
-------
Figure 2. Zero-Headspace Extraction Vessel
liquid inlet/outlet valve
t
* filter
top flange
waste and
extraction
fluid
I
piston
-V body
VI TON 0-rings
>• bottom flange
pressurizing gas inlet/outlet valve
1312-14
Revision 0
December 1986
-------
METHOD 1312
SYNTHETIC ACID PRECIPITATION LEACH TEST FOR SOILS
T.J-T.I Hrftn
"tract!** »f ittl
for It k.iri; ».
•rfimlct
7-*
••HMtl.t t>
ai (f«r ..Utll.)
•': 3) ftltMd..'
Revi sion
December
0
1988
-------
OSWER DIRECTIVE 9502.00-6D
INTERIM FINAL
RCRA FACILITY INVESTIGATION (RFI) GUIDANCE
VOLUME III OF IV
t
AIR AND SURFACE WATER RELEASES »
EPA 530/SW-89-031
MAY 1989
WASTE MANAGEMENT DIVISION
OFFICE OF SOLID WASTE
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
ABSTRACT
On November 8, 1984, Congress enacted the Hazardous and Solid Waste
Amendments (HSWA) to RCRA. Among the most significant provisions of HSWA are
§3004(u), which requires corrective action for releases of hazardous waste or
constituents from solid waste management units at hazardous waste treatment,
storage and disposal facilities seeking final RCRA permits; and §3004(v), which
compels corrective action for releases that have migrated beyond the facility
property boundary. EPA will be promulgating rules to implement the corrective
action provisions of HSWA, including requirements for release investigations and
corrective measures.
\. t
This document, which is presented in four volumes, provides guidance tp
regulatory agency personnel on overseeing owners or operators of hazardous wastp
management facilities in the conduct of the second phase of the RCRA Corrective
Action Program, the RCRA Facility Investigation (RFI). Guidance is provided for the
development and performance of an investigation by the facility owner or operator
based on determinations made by the regulatory agency as expressed in the
schedule of a permit or in an enforcement order issued under §3008(h), §7003,
and/or §3013. The purpose of the RFI is to obtain information to fully characterize
the nature, extent and rate of migration of releases of hazardous waste or
constituents and to interpret this information to determine whether interim
corrective measures and/or a Corrective Measures Study may be necessary.
-------
DISCLAIMER
This document is intended to assist Regional and State personnel in exercising
the discretion conferred by regulation in developing requirements for the conduct
of RCRA Facility Investigations (RFIs) pursuant to 40 CFR 264. Conformance with this
guidance is expected to result in the development of RFIs that meet the regulatory
standard of adequately detecting and characterizing the nature and extent of
releases. However, EPA will not necessarily limit acceptable RFIs to those that
comport with the guidance set forth herein. This document is not a regulation (i.e.,
it does not establish a standard of conduct which has the force of law) and should
not be used as such. Regional and State personnel must exercise their discretion in
using this guidance document as well as other relevant information in determining
whether an RFI meets the regulatory standard. t
?
Mention of company or product names in this document should not be
considered as an endorsement by the U.S. Environmental Protection Agency.
-------
RCRA FACILITY INVESTIATION (RFI) GUIDANCE
VOLUME III
AIR AND SURFACE WATER RELEASES
TABLE OF CONTENTS
SECTION PAGE
ABSTRACT i
DISCLAIMER ii
TABLE OF CONTENTS iii
TABLES xi
FIGURES xiii
LIST OF ACRONYMS xiy
in
-------
VOLUME III CONTENTS (Continued)
SECTION PAGE
12.0 AIR 12-1
12.1 OVERVIEW 12-1
12.2 APPROACH FOR CHARACTERIZING RELEASES TO AIR 12-2
12.2.1 General Approach 12-2
12.2.1.1 Initial Phase 12-13
12.2.1.1.1 Collect and Review Preliminary 12-13
Information
12.2.1.1.2 Conduct Screening Assessment 12-14
12.2.1.2 Subsequent Phases 12-15
* 12.2.1.2.1 Conduct Emission Monitoring 12-16
12.2.1.2.2 Confirmatory Air Monitoring 12-17 ^
12.3 CHARACTERIZATION OF THE CONTAMINANT 12-20
SOURCE AND THE ENVIRONMENTAL SETTING
12.3.1 Waste Characterization 12-21
12.3.1.1 Presence of Constituents 12-21
12.3.1.2 Physical/Chemical Properties 12-21
12.3.2 Unit Characterization 12-27
12.3.2.1 Type of Unit 12-27
12.3.2.2 Size of Unit 12-33
12.3.2.3 Control Devices 12-34
12.3.2.4 Operational Schedules 12-35
12.3.2.5 Temperature of Operation 12-35
K-
12.3.3 Characterization of the Environmental Setting 12-36
12.3.3.1 Climate 12-36
12.3.3.2 Soil Conditions 12-38
12.3.3.3 Terrain 12-38
12.3.3.4 Receptors 12-39
12.3.4 Review of Existing Information 12-39
IV
-------
VOLUME III CONTENTS (Continued)
SECTION PAGE
12.3.5 Determination of Reasonable Worst Case" 12-41
Exposure Period
12.4 AIR EMISSION MODELING 12-43
12.4.1 Modeling Applications 12-43
12.4.2 Model Selection 12-44
12.4.2.1 Organic Emissions 12-44
12.4.2.2 Paniculate Emissions 12-46
12.4.3 General Modeling Considerations 12-47
12.5 DISPERSION MODELING 12-48
12.5.1 Modeling Applications 12-48
12.5.2 Model Selection 12-50 ?
12.5.2.1 Suitability of Models 12-51
12.5.2.2 Classes of Models 12-52
12.5.2.3 Levels of Sophistication of Models 12-53
12.5.2.4 Preferred Models 12-54
12.5.3 General Modeling Considerations 12-56
12.6 DESIGN OF A MONITORING PROGRAM TO 12-58
CHARACTERIZE RELEASES
12.6.1 Objectives of the Monitoring Program 12-58
12.6.2 Monitoring Constituents and Sampling 12-59
Considerations
12.6.3 Meteorological Monitoring « 12-60
12.6.3.1 ^Meteorological Monitoring Parameters 12-60
12.6.3.2 "Meteorological Monitor Siting 12-62
12.6.4 Monitoring Schedule 12-64
12.6.4.1 Screening Sampling 12-64
12.6.4.2 Emission Monitoring 12-65
12.6.4.3 Air Monitoring 12-68
12.6.4.4 Subsequent Monitoring 12-69
-------
VOLUME III CONTENTS (Continued)
SECTION PAGE
12.6.5 Monitoring Approach 12-69
12.6.5.1 Source Emissions Monitoring 12-71
12.6.5.2 Air Monitoring 12-72
12.6.6 Monitoring Locations 12-73
12.6.6.1 Upwind/Downwind Monitoring Location 12-73
12.6.6.2 Stack/Vent Emission Monitoring 12-77
12.6.6.3 Isolation Flux Chambers 12-77
12.7 DATA PRESENTATION 12-78
12.7.1 Waste and Unit Characterization 12-78.
i
12.7.2 Environmental Setting Characterization 12-79t
12.7.3 Characterization of the Release 12-80*
12.8 FIELD METHODS 12-85
12.8.1 Meteorological Monitoring 12-86
12.8.2 Air Monitoring 12-86
12.8.2.1 Screening Methods 12-89
12.8.2.2 Quantitative Methods 12-93
12.8.2.2.1 Monitoring Organic Compounds in 12-93
Air
12.8.2.2.1.1 Vapor-Phase Organics 12-94
12.8.2.2.1.2 Particulate Organics 12-111
12.8.2.2.2 Monitoring Inorganic Compounds in 12-113
Air
SL
12,8.2.2.2.1 Particulate Metals 12-113
12.8.2.2.2.2 Vapor-Phase Metals 12-114
12.8.2.2.2.3 Monitoring Acids and Other 12-120
Compounds in Air
12.8.3 Stack/Vent Emission Sampling 12-121
12.8.3.1 Vapor Phase and Particulate Associated 12-122
Organics
VI
-------
VOLUME III CONTENTS (Continued)
SECTION PAGE
12.8.3.2 Metals 12-127
12.9 SITE REMEDIATION 12-129
12.10 CHECKLIST 12-131
12.11 REFERENCES 12-133
vti
-------
VOLUME III CONTENTS (Continued)
SECTION PAGE
13.0 SURFACE WATER 13-1
13.1 OVERVIEW 13-1
13.2 APPROACH FOR CHARACTERIZING RELEASES TO 13-2
SURFACE WATER
13.2.1 General Approach 13-2
13.2.2 Inter-media Transport 13-8
13.3 CHARACTERIZATION OF THE CONTAMINANT 13-8
SOURCE AND THE ENVIRONMENTAL SETTING
13.3.1 Waste Characterization 13-8
13.3.2 Unit Characterization 13-17 '
13.3.2.1 Unit Characteristics 13-17 *
13.3.2.2 Frequency of Release 13-18
13.3.2.3 Form of Release 13-19
13.3.3 Characterization of the Environmental Setting 13-19
13.3.3.1 Characterization of Surface Waters 13-20
13.3.3.1.1 Streams and Rivers 13-20
13.3.3.1.2 Lakes and Impoundments 13-22
13.3.3.1.3 Wetlands 13-24
13.3.3.1.4 Marine Environments 13-25
13.3.3.2 Climatic and Geographic Conditions 13-26
13.3.4 Sources of Existing Information 13-27
13.4 DESIGN QF A MONITORING PROGRAM TO 13-28
CHARACTERIZE RELEASES
13.4.1 Objectives of the Monitoring Program 13-29
13.4.1.1 Phased Characterization 13-30
13.4.1.2 Development of Conceptual Model 13-31
13.4.1.3 Contaminant Concentration vs 13-31
Contaminant Loading
13.4.1.4 Contaminant Dispersion Concepts 13-33
VIII
-------
VOLUME III CONTENTS (Continued)
SECTION PAGE
13.4.1.5 Conservative vs Non-Conservative Species 13-36
13.4.2 Monitoring Constituents and Indicator 13-36
Parameters
13.4.2.1 Hazardous Constituents 13-36
13.4.2.2 Indicator Parameters 13-37
13.4.3 Selection of Monitoring Locations 13-42
13.4.4 Monitoring Schedule 13-44
13.4.5 Hydrologic Monitoring 13-46
13.4.6 The Role of Biomonitoring 13-46
13.4.6.1 Community Ecology Studies 13-47{
13.4.6.2 Evaluation of Food Chain/Sensitive Species 13-48;
Impacts
13.4.6.3 Bioassay 13-49
13.5 DVTA MANAGEMENT AND PRESENTATION 13-50
13.5.1 Waste and Unit Characterization 13-50
13.5.2 Environmental Setting Characterization 13-51
13.5.3 Characterization of the Release 13-51
13.6 FIELD AND OTHER METHODS 13-53
13.6.1 Surface Water Hydrology 13-53
13.6.2 Sampling of Surface Water, Runoff, Sediment 13-55
and Biota
13.6.2.r Surface Water 13-55
13.6.2M.1 Streams and Rivers 13-55
13.6.2.1.2 Lakes and Impoundments 13-56
13.6.2.1.3 Additional Information 13-57
13.6.2.2 Runoff Sampling 13-58
13.6.2.3 Sediment 13-59
13.6.2.4 Biota 13-62
IX
-------
VOLUME III CONTENTS (Continued)
SECTION PAGE
13.6.3 Characterization of the Condition of the 13-63
Aquatic Community
13.6.4 Bioassay Methods 13-66
13.7 SITE REMEDIATION 13-67
13.8 CHECKLIST 13-68
13.9 REFERENCES 13-71
APPENDICES
Appendix G: Draft Air Release Screening Assessment
Methodology
Appendix H: Soil Loss Calculation
-------
TABLES (Volume ill)
NUMBER PAGE
12-1 Example Strategy for Characterizing Releases to Air 12-3
12-2 Release Characterization Tasks for Air 12-5
12-3 Parameters and Measures for Use in Evaluating Potential 12-22
Releases of Hazardous Waste Constituents to Air
12-4 Physical Parameters of Volatile Hazardous Constituents 12-25
12-5 Physical Parameters of PCB Mixtures 12-26
12-6 Summary of Typical Unit Source Type and Air Release Type 12-28
12-7 Typical Pathways for Area Emission Sources 12-49
12-8 Preferred Models for Selected Applications in Simple 12-55
Terrain
r»-
12-9 Recommended Siting Criteria to Avoid Terrain Effects 12-63
12-10 Applicable Air Sampling Strategies by Source Type 12-70
12-11 Typical Commercially Available Screening Techniques 12-90
forOrganicsin Air
12-12 Summary of Selected Onsite Organic Screening 12-92
Methodologies
12-13A Summary of Candidate Methodologies for Quantification of 12-95
Vapor Phase Organics
12-13B List of Compound Classes Referenced in Table 12-15A 12-97
12-14 Sampling and Analysis Techniques Applicable to Vapor 12-98
Phase Organics
12-15 Compounds Monitored Using EMSL-RTPTenax Sampling 12-102
Protocols
12-16 Summary Listing of Organic Compounds Suggested for 12-106
Collection With a Low Volume Polyurethane Foam Sampler
and Subsequent Analysis With an Electron Capture Detector
(GC/ECD)
XI
-------
TABLES (Volume III • Continued)
NUMBER PAGE
12-17 Summary Listing of Additional Organic Compounds 12-107
Suggested for Collection With a Low Volume Polyurethane
Foam Sampler
12-18 Sampling and Analysis Methods for Volatile Mercury 12-115
12-19 Sampling and Analysis of Vapor State Trace Metals 12-118
(Except Mercury)
12-20 Sampling Methods for Toxic and Hazardous Organic 12-123
Materials From Point Sources
12-21 RCRA Appendix VIII Hazardous Metals and Metal 12-128
Compounds
13-1 Example Strategy for Characterizing Releases to 13-3
Surf ace Water
13-2 Release Characterization Tasks for Surface Water 13-7
13-3 Important Waste and Constituent Properties Affecting 1 3-9
Fate and Transport in a Surface Water Environment
13-4 General Significance of Properties and Environmental 13-16
Processes for Classes of Organic Chemicals Under
Environmental Conditions
XII
-------
FIGURES (Volume III)
NUMBER PAGE
12-1 Release Characterization Strategy for Air-Overview 12-6
12-2 Conduct Screening Assessments-Overview 12-7
12-3 Conduct Emission Monitoring-Overview 12-8
12-4 Conduct Confirmatory Air Monitoring 12-9
12-5 Evaluation of Modeling/Monitoring Results 12-10
12-6 Example Air Monitoring Network 12-74
12-7 Example of Downwind Exposures at Air Monitoring Stations 12-84
13-1 Qualitative Relationship Between Various Partitioning 13-11
Parameters
13-2 Typical Lake Cross Section 13-23 t
f
xiii
-------
LIST OF ACRONYMS
AA
Al
ASCS
ASTM
BCF
BOD
CAG
CPF
CBI
CEC
CERCLA
CFR
CIR
CM
CMI
CMS
COD
COLIWASA
DNPH
DO
DOT
ECD
EM
EP
EPA
FEMA
FID
Foe
FWS
GC
GC/MS
GPR
HEA
HEEP
HPLC
HSWA
HWM
ICP
ID
Kd
Koc
Kow
LEL
MCL
MM5
MS/MS
NFIP
NIOSH
NPDES
OSHA
Atomic Absorption
Soil Adsorption Isotherm Test
Agricultural Stabilization and Conservation Service
American Society for Testing and Materials
Bioconcentration Factor
Biological Oxygen Demand
EPA Carcinogen Assessment Group
Carcinogen Potency Factor
Confidential Business Information
Cation Exchange Capacity
Comprehensive Environmental Response, Compensation, and
Lability Act
Code of Federal Regulations
Color Infrared
Corrective Measures
Corrective Measures Implementation
Corrective Measures Study
Chemical Oxygen Demand
Composite Liquid Waste Sampler
Dinitrophenyl Hydrazine (
Dissolved Oxygen ?
Department of Transportation
Electron Capture Detector
Electromagnetic
Extraction Procedure
Environmental Protection Agency
Federal Emergency Management Agency
Flame lonization Detector
Fraction organic carbon in soil
U.S. Fish and Wildlife Service
Gas Chromatography
Gas Chromatography/Mass Spectroscopy
Ground Penetrating Radar
Health and Environmental Assessment
Health and Environmental Effects Profile
High Pressure Liquid Chromatography
Hazardous and Solid Waste Amendments (to RCRA)
Hazardous Waste Management
Inductively Coupled (Argon) Plasma
Infrared Detector
Soil/Water Partition Coefficient
Organic Carbon Absorption Coefficient
Octanol/Water Partition Coefficient
Lower Explosive Limit
Maximum Contaminant Level
Modified Method 5
Mass Spectroscopy/Mass Spectroscopy
National Flood Insurance Program
National Institute for Occupational Safety and Health
National Pollutant Discharge Elimination System
Occupational Safety and Health Administration
XIV
-------
LIST OF ACRONYMS (Continued)
OVA
PID
pKa
ppb
ppm
PUF
PVC
QA/QC
RCRA
RFA
RfD
RFI
RMCL
RSD
SASS
SCBA
SCS
SOP
SWMU
TCLP
TEGD
TOC
TOT
TOX
uses
USLE
UV
VOST
VSP
WQC
Organic Vapor Analyzer
Photo lonization Detector
Acid Dissociation Constant
parts per billion
parts per mi 11 ion
Polyurethane Foam
Polyyinyl Chloride
Quality Assurance/Quality Control
Resource Conservation and Recovery Act
RCRA Facility Assessment
Reference Dose
RCRA Facility Investigation
Recommended Maximum Contaminant Level
Risk Specific Dose
Source Assessment Sampling System
Self Contained Breathing Apparatus
Soil Conservation Service
Standard Operating Procedure
Solid Waste Management Unit
Toxicity Characteristic Leaching Procedure
Technical Enforcement Guidance Document (EPA, 1986)
Total Organic Carbon
Time of travel
Total Organic Halogen
United States Geologic Survey
Universal Soil Loss Equation
Ultraviolet
Volatile Organic Sampling Train
Vertide Seismic Profiling
Water Quality Criteria
xv
-------
-------
SECTION 12
AIR
12.1 Overview
The objective of an investigation of a release to air is to characterize the
nature, extent, and rate of migration of the release of hazardous waste or
constituents to that medium. This is done by characterizing long-term air
concentrations (commensurate with the long-term exposures which are the basis for
the health and environmental criteria presented in Section 8) associated with unit
releases of hazardous wastes or constituents to air. This section provides:
• An example strategy for characterizing releases to air, which includes'
characterization of the source and the environmental setting of the*
release, and conducting a monitoring and/or modeling, program which
will characterize the release itself;
• Formats for data organization and presentation;
• Modeling and field methods which may be used in the investigation; and
• A checklist of information that may be needed for release
characterization.
The exact type and amount of information required for sufficient release
characterization will be site-specific and should be determined through interactions
between the regulatory agency and the facility owner or operator during the RFI
process. This guidance does not define the specific data needed in all Instances; it
identifies possible information necessary to perform release characterizations and
methods for obtaining this information. The RFI Checklist, presented at the end of
this section, provides a tool for planning and tracking information for release
characterization. This list is not a list of requirements for all releases to air. Some
release investigations will involve the collection of only a subset of the items listed,
while other releases may involve-the collection of additional data.
12-1
-------
Case studies 25 and 26 in Volume IV (Case Study Examples) illustrate several of
the air investigation concepts discussed in this section.
12.2 Approach for Characterizing Releases to Air
12.2.1 General Approach
The intent of the air release investigation is to determine actual or potential
effects at the facility property boundary. This differs from the other media
discussed in this Guidance. During the health and environmental assessment
process for the air medium (see Section 8), the decision as to whether interim
corrective measures or a Corrective Measures Study will be necessary is based on
actual or potential effects at the facility property boundary.
"- »
Characterization of releases from waste management units to air may be.
approached in a tiered or phased fashion as described in Section 3. The key
elements to this approach are shown in Table 12-1. Tasks for implementing the
release characterization strategy for releases to air are summarized in Table 12-2.
An overview of the release characterization strategy for air is illustrated in Figures
12-1 through 12-5.
Two major elements can be derived from this strategy:
• Collection and review of data to be used for characterization of the
source of the air release and the environmental setting for this source.
Source characterization will include obtaining information on the unit
operating conditions and configuration, and may entail a sampling and
analytical effort to characterize the waste material in the unit or the
incoming waste streams. This effort will lead to development of a
conceptual model of the release that provides a working hypothesis of
the release mechanism, transport pathway/mechanism, and exposure
route (if any), which can be used to guide the investigation.
• Development and implementation of modeling and/or monitoring
procedures to be used for characterization of the release (e.g., from a
12-2
-------
TABLE 12-1
EXAMPLE STRATEGY FOR CHARACTERIZING RELEASES TO AIR*
INITIAL PHASE
1. Collect and review existing information on:
2.
Waste
Unit
Environmental setting (e.g., climate, topography)
Contaminant releases, including inter-media transport
Receptors at and beyond the facility property boundary
Identify additional information necessary to fully characterize release:
Waste
Unit
Environmental setting (e.g., climate, topography)
Contaminant releases, including inter-media transport
Receptors at and beyond the facility property boundary
3. Conduct screening assessments:
Formulate conceptual model of release
Determine monitoring/modeling program objectives
Obtain source characterization data needed for modeling input
Select release constituent surrogates
Calculate emission estimates based on emission rate screening
modeling results
Calculate concentration estimates based on dispersion screening
modeling results
Compare results to health based criteria
Conduct screening monitoring at source (as warranted)
Perform sensitivity analysis of modeling input/output
Obtain additional waste/unit data as needed for refined modeling
Consider conduct of more refined emission/dispersion modeling
4. Collect, evaluate and report results:
Account for unit/waste temporal and spatial variability and modeling
input/output uncertainties
Determine completeness and adequacy of screening assessment
results
Evaluat^ potential for inter-media contaminant transfer
Summar4ze and present results in appropriate format
Determine if monitoring program objectives were met
Compare screening results to "health and environmental criteria and
identify and respond to emergency situations and identify priority
situations that may warrant interim corrective measures - Notify
regulatory agency
Determine whether the conduct of subsequent release charaterization
phases are necessary to obtain more refined concentration estimates
12-3
-------
TABLE 12-1 (continued)
EXAMPLE STRATEGY FOR CHARACTERIZING RELEASES TO AIR*
SUBSEQUENT PHASES (if necessary)
.1. Conduct emission monitoring and dispersion modeling if necessary:
Conduct onsite meteorological monitoring if representative data are
not available for dispersion modeling input
Conduct emission rate monitoring
Conduct dispersion modeling using emission rate monitoring data as
input
Evaluate results and determine need for confirmatory air monitoring
2. Conduct confirmatory air monitoring if necessary:
Develop monitoring procedures
Conduct initial monitoring
Conduct additional monitoring if additional information is necessary
to characterize the release
3. Collect, evaluate and report results:
Account for source and meteorological data variability during
modeling and monitoring program
Evaluate long-term representativeness of air monitoring data
Apply dispersion models as appropriate to aid in data evaluation and
to provide concentration estimates at the facility property boundary
Compare monitoring results to health and environmental criteria and
identify and respond to emergency situations and identify priority
situations that may warrant interim corrective measures - Notify
regulatory agency
Determine completeness and adequacy of collected data
Summarize and present data in appropriate format
- . Determine if modeling and monitoring locations, constituents, and
frequency were adequate to characterize release (nature, extent, and
rate)
Determine if monitoring/modeling program objectives were met
Identify additional information needs, if necessary
Determine need to expand modeling and monitoring program
Evaluate, potential role of inter-media transport
The potential for inter-media transport of contamination should be
evaluated continually throughout the investigation.
12-4
-------
TABLE 12-2
RELEASE CHARACTERIZATION TASKS FOR AIR
Investigatory Tasks
Investigatory Techniques
Data Presentation
Formats/Outputs
Waste/Unit Characterization
Identification of waste
constituents and properties
•Prioritization of air emission
constituents
Identification of unit
characteristics which may
promote an air release
See Section 3, 7 and Volume I,
Appendix B List 2; Section 12.3,
Section 12.4, Appendix F
Waste sampling and
characterization
See Section 7, Section 12.3,
Section 12.4, Appendix F
Listing of potential release
constituents
Listing of target air emission
constituents for monitoring
Description of the unit
Environmental Setting
Characterization
Definition of climate
Definition of site-specific
meteorological conditions
Definition of soil conditions
to characterize emission
potential for paniculate
emissions and for certain
units (e.g., landfills and land
treatment) for gaseous
emissions
Definition of site-specific
terrain
Identification of potential
air-pathway receptors
Climate summaries for regional
National Weather Service
stations (may require onsite
meteorological monitoring
survey)
Onsite meteorological
monitoring concurrent with air
monitoring
See Section 9
See Section 7. 9 and Appendix A
(Volume 1) of RFI and recent
aerial photographs and U.S.
Geoigoical Survey maps
Census data, area surveys, recent
aerial photographs and U.S.
Geological Survey topographic
maps '
Wind roses and statistical
tabulations for parameters of
interest
Wind roses and tabulations for
parameters of interest
Soil physical properties (e.gJ,
porosity, organic matter
content) "
Topographic map of site area
Map with identification of
nearby populations and
buildings
Release Characterization
Emission rate modeling
Dispersion modeling
Emission rate monitoring
Air monitoring
Air emission models as discussed
in Section 12.4
Atmospheric dispersion models
as discussed in Section 12.5
Direct emission source tests for
point sources, isolation flux
chamber for area sources or
onsite air monitoring (Section
12.8)
Upwind/downwind air
monitoring for "release
mapping"
Unit-specific and constituent-
specific emission rates
Air concentration estimates at
facility property boundary
(tabular summaries or graphical
presentations which may include
release concentration isopleths)
Listing of emission rate
monitoring results
Air concentration estimates at
facility property boundary
(tabular summaries or graphical
presentations which may include
release concentration isopleths)
12-5
-------
FIGURE 12-1
RELEASE CHARACTERIZATION STRATEGY FOR AIR-OVERVIEW
Collect and Review
Preliminary Information
Waste/Unit
Characteristics
Historical
Air Monitoring/Modeling
Data
Environmental
Characteristics
Develop Conceptual Model of Release
Evalute
Hazard Index/
RFI Decision
Points
INITIAL
PHASE
Conduct Screening Assessments
(Emphasis on Emission Modeling)
Evalute
Hazard Index/
RFI Decision
Points
Conduct Emission Monitoring
Evalute
Hazard Index/
RFI Decision
Points
Confirmatory Air Monitoring
Evalute
Hazard Index/
RFI Decision
Points
1
Information Sufficient
to Characterize Air
Release as Significant
Information Sufficient
to Characterize Air
Release as Insignificant
SUBSEQl
PHAS
Corrective Measures
Study/Interim Corrective
Measure
12-6
No Further Action
Required
-------
FIGURE 12-2
CONDUCT SCREENING ASSESSMENTS - OVERVIEW
Collect and Review
Preliminary Information
Consider Refined
Emission/Dispersion
Modeling
Conduct Screening Modeling
Obtain source characterization data
Select release constituent surrogates
Calculate emission estimates based on emission
modeling results
Calculate concentration estimates based on
dispersion modeling results
Compare results to health based criteria
Obtain Additional
Waste/Unit Data
Conduct Preliminary
Monitoring at Source
(discretionary)
Conduct Model Sensitivity Analysis, Evaluate Input
Data and Model Accuracy to Determine Uncertainty
Factor (U?)
No
(Optional steps)
Screening
Assessment
Results
dequate
Evaluate
Hazard Index/
RFI Decision
Points
orrective Measure Study/
:enm Corrective Measures
Conduct Emission Monitoring
(See Figure 12-3)
No Further
Action Required
12-7
-------
FIGURE 12-3
CONDUCT EMISSION MONITORING - OVERVIEW
Screening Assessments
Representative
Meteorological
Data Available
No
ves
Conduct
Meteorological
Monitoring
Conduct Emissions
Rate Monitoring
Direct Emissions
Source Testing
for Point Sources
Isolation Flux
Chamber
for Area Sources
t
Onsite
Air
Monitoring
Conduct
Dispersion Modeling
Corrective Measures Study/
Interim Corrective Measures
Evaluate
Hazard Index/
RFI Decision
Points
Conduct Confirmatory
Air Monitoring
(See Figure 12-4)
No Further
Action Required
12-8
-------
FIGURE 12-4
CONDUCT CONFIRMATORY AIR MONITORING
Emission Monitoring Results
Screening Develop Monitoring
Air Samples Procedures
*
Select Monitoring
Approach/Procedures
t
Monitor _,
Placement*
*
Conduct Initial Monitoring
*
1 I
Air Meteorological
Monitoring Monitoring
1 |
*
Ca
>
-*-
-^~
ndidate Air Emission
Constituents (see
Appendix B, List 2)
Site
Meteorological
Characterization
Dispersion
Modeling
*e As close to source as
possible to increase
potential for release
detection and
quantification
e At actual receptors' at or
beyond the facility*
property boundaivto
support health ana
environmental
assessment (if oracticaO
Collect and Evaluate Results
i
t
Waste/Unit
Characterization
Data Summaries
Summarize Data/
Perform Dispersion
Modeling**
t
t
Air/Meteorological
Monitoring Data
Summaries
1
conce
recep
beyor
prop*
neces
t
Modeling Data
Summaries
i.
** To Estimate
concentrations at actual
receptor locations at or
beyond the facility
property boundary (as
Additional Monitoring (if necessary)
f
rective Measures Study/Interim
Corrective Measures
Evaluate
Hazard Index/
RFI Decision
Points
No Further Action
Required
12-9
-------
FIGURE 12-5
EVALUATION OF MODELING/MONITORING RESULTS
Modeling/Monitoring Results
i
Compute
Hazard Index
(HI)
I
Determine
Modeling/Monitoring
Uncertainty Factors
(±UF)*
Evaluate
Hazard
Index/RFI
Decision
Poin
I HI.>UF** I |UF>HI.>1/UF*** I |HK1/UF****|
•••^•^•^•^••••J ^^,^^^^^,^^^l^^m^mlm^ ^^^^m*mi^mmm^l
Information is
sufficient to
characterize
release as
significant
Information is
not sufficient
to
characterize
the release
7
Information is
sufficient to
characterize
the release as
insignificant
7
Corrective Measures
Study/Interim
Corrective Measures
Additional Release
Characterization
Assessments
Necessary
No
Further
Action
Required
* UnctrtairTty Factor assumed to be J> 1.0
** Hl>1 Generally used for evaulation of confirmatory air monitoring
results.
*** This alternative is generally not used to evaluate confirmatory air
monitoring results. However, additional air monitoring may be
warranted if monitoring objectives were not acheived. Confirmatory air
monitoring will generally be conducted during worst-case long-term
emission/dispersion conditions. Therefore, this facilitates the use of
more rigorous evaluation criteria for this final air release
characterization step prior to RFI decisionmaking.
**** HK1 Criterion generally used for evaluation of confirmatory air
monitoring results.
12-10
-------
unit or contaminated soil). Utilizing a phased approach, the air release is
characterized in terms of the types and amounts of hazardous
constituents being emitted, leading to a determination of actual or
potential exposure at the facility property boundary. This may involve
emission modeling (to estimate unit-specific emission rates), air
monitoring (to determine concentrations at the facility property
boundary), emission monitoring (monitoring at the source to determine
emission rates), and dispersion modeling (to estimate concentrations at
the facility property boundary). A phased approach utilizing both
modeling and monitoring may not always be necessary to achieve
adequate release charterization.
As indicated in Section 1 of this Guidance (See Volume I), standards for the
control and monitoring of air emissions at hazardous waste treatment, storage and
disposal (ISO) facilities are being developed by the Agency pursuant to HSWA'
Section 3004(n). These standards will address specific methodologies and
regulatory requirements for the identification and control of air releases at TSD
facilities. The Guidance provided herein is intended to provide interim
methodologies and procedures for the identification and delineation of significant
air releases. In particular, the Guidance addresses those releases which may pose an
existing and significant hazard to human health and the environment, and thus,
should be addressed without delay, i.e., prior to the issuance of the Section 3004(n)
regulations.
The RFI release characterization strategy for air includes several decision points
during the characterization process to evaluate the adequacy of available
information and to determine an appropriate course of action from the following
alternatives (as illustrated in Figures 12-1 through 12-5).
SL
• Information is sufficient to characterize the air release as significant and a
Corrective Measures Study/Interim Corrective Measures is warranted.
• Information is sufficient to characterize the air release as insignificant,
therefore, no further air assessments are required.
12-11
-------
• Information is not sufficient to characterize the air release, therefore further
release characterization is warranted.
Criteria for decisionmaking involves consideration of the uncertainty
associated with release characterization results (modeling/monitoring), which is
facilitated by use of a Hazard Index as illustrated in Figure 12-5. The Hazard Index is
defined as the ratio of exposure concentration levels or estimates, to specific health
criteria for an individual constituent or a mixture of constituents with similar
potential health impacts. Further guidance on the computation and application of
the Hazard Index is provided in Section 8.
The uncertainty associated with concentration estimates based on air pathway
modeling and monitoring results is factored into the decision making effort
through use of uncertainty analyses. A primary component of the uncertainty
analysis is the accuracy of the modeling and/or monitoring approach utilized for the
release characterization. Model-specific and monitoring method-specific accuracies
should be used as available for the uncertainty analysis. The quality of the input
data to models is another important component of the uncertainty analysis that
should be accounted for. Generally, conduct of a model sensitivity analysis (i.e.,
varying the values of input parameters based on their uncertainty range to evaluate
the effect on model output), will provide a quantitative basis to characterize input
data quality. This step is particularly important for some unit-specific models. For
example, the spatial variability of wastes at a landfill and the uncertainty of other
input parameters (e.g., soil porosity) can significantly affect the overall uncertainty
associated with emission modeling results.
As concentration measurements or estimates at the facility property boundary
become available, both within and at the conclusion of discrete investigation
phases, they shoujd be reported to the regulatory agency as directed. The
regulatory agency will compare the concentrations with applicable health and
environmental criteria to determine the need for (1) interim corrective measures;
and/or (2) a Corrective Measures Study. In addition, the regulatory agency will
evaluate the data with respect to adequacy and completeness to determine the
need for any additional characterization efforts. The health and environmental
criteria and a general discussion of how the regulatory agency will apply them are
12-12
-------
provided in Section 8. A flow diagram illustrating RFI Decision Points is provided in
Section 3 (See Figure 3-2).
Notwithstanding the above process, the owner or operator has a continuing
responsibility to identify and respond to emergency situations and to define priority
situations that may warrant interim corrective measures. For these situations, the
owner or operator is advised to follow the RCRA Contingency Plan requirements
under 40 CFR Part 264, Subpart D and Part 265, Subpart D.
The strategy for characterizing releases to air consists of an initial phase and, if
necessary, subsequent phases, as illustrated in Table 12-1 and Figure 12-1.
Additional phases may not be needed depending on the site-specific
modeling/monitoring data available, and the nature and magnitude of the release.
A summary discussion of the initial phase is presented in Section 12.2.1.1 and the
subsequent phases in Section 12.2.1.2.
12.2.1.1 Initial Phase
The initial phase of the release characterization strategy for air involves the
collection and review of preliminary information and the conduct of a screening
assessment.
12.2.1.1.1 Collectand Review Preliminary Information
The first step is to collect, review and evaluate available waste, unit,
environmental setting and release (monitoring and modeling) data. The air
pathway data collection effort should be coordinated, as appropriate, with similar
efforts for other media investigations.
c_
Evaluation of these data may, at this point, clearly indicate that a Corrective
Measures Study and/or interim corrective measures are necessary or that no further
action is required. For example, the source may involve a large, active storage
surface impoundment containing volatile constituents located adjacent to
residential housing. Therefore, action instead of further studies may be
appropriate. Another case may involve a unit in an isolated location, where an
acceptable modeling/monitoring data base may be available which definitively
12-13
-------
indicates that the air release can be considered insignificant and therefore further
studies are not warranted. In most cases, however, further release characterization
will be necessary.
A conceptual model (as discussed in Volume I - Summary Section and Section
3.2) of the release should then be developed based on available information. This
model (not a computer or numerical simulation model) should provide a working
hypothesis of the release mechanism, transport pathway/mechanism, and exposure
route (if any). The model should be testable/verifiable and flexible enough to be
modified as new data become available. For example, transport pathway and
exposure modes for a contaminated surface area may involve air emissions due to
volatilization, wind erosion and mechanical disturbances. These air emissions are
expected to result in inhalation exposure for offsite receptors. In addition, the
deposition of air emissions on soil, water bodies and crops, and infiltration and
runoff from the onsite source, may contribute to overall exposures.
12.2.1.1.2 Conduct Screening Assessment
Following review of existing information and development of the conceptual
model, a screening assessment should be conducted to characterize the air release
(see Figure 12-2). The initial screening should be based on conservative (i.e., worst-
case assumptions). A screening assessment based on more realistic assumptions
should be conducted if initial air concentration predictions exceed health criteria.
The Draft Final Air Release Screening Assessment Methodology, presented in
Appendix G, describes the screening assessment in detail. It consists of emission rate
and dispersion models and involves the following steps:
• Obtain source characterization input data
• Select release (target) constituents which may be present in the waste
and have health criteria for the air pathway (see Section 8.0)
• Calculate emission estimates
• Calculate concentration estimates at facility property boundary
• Compare results to health based criteria
12-14
-------
In order to assure adequate source characterization input data, it may be
necessary to collect additional waste/unit data. This may involve field sampling of
the waste to identify waste constituents and determine concentration levels. At this
early RFI stage, it may be more effective and conclusive to sample the wastes (with
relatively higher concentration levels) instead of the release. In general, if
obtaining source-specific data is not practical, conservative source assumptions
should be used.
Preliminary monitoring at the source may also be conducted to aid in the
evaluation of the screening/modeling results. Preliminary monitoring may involve
the use of screening or quantitative methods, and is discussed in Section 12.6. The
preliminary monitoring period will generally be limited to a few days. Although
preliminary monitoring results may identify release constituents that were not
expected based on modeling, or vice versa, the limitations of modeling and
monitoring should be considered when comparing these data and determining
appropriate followup activities. .
A sensitivity analysis should also be conducted to evaluate model input data
quality. The results of the sensitivity analysis as well as consideration of model
accuracy should be used to compute the UF for the screening assessment. The
results of the screening assessment should then be compared to the health and
environmental assessment criteria (as previously discussed) to determine
appropriate followup actions. Collection of additional waste/unit data and/or
considering the application of more refined emission/dispersion models are also
possible options if initial results from the screening assessment are inconclusive.
12.2.1.2 Subsequent Phases
£_
Subsequent phases of the release characterization strategy for air may be
necessary if screening assessment results are not conclusive to characterize the air
release, and should involve the conduct of emission monitoring and confirmatory
air monitoring as indicated in Figure 12-1. These are discussed below.
12-15
-------
12.2.1.2.1 Conduct Emission Monitoring
Source monitoring should be used in conjunction with dispersion modeling to
further characterize the release, as indicated in Figure 12.3. Direct emission
sampling should be used for point sources such as vents and stacks. An isolation
ffux chamber may be used for area source emission measurements. Onsite air
monitoring (particularly near the emission source) is an alternative approach for
characterizing area source emissions if direct emission monitoring is not practical
(e.g., considering equipment availability). Guidance for the conduct of these field
programs is presented in Section 12.6 and 12.8.
The development of emission monitoring procedures should address selection
of target air emission constituents. One acceptable approach is to monitor for all
potential Appendix VIII air emission constituents (see Appendix B, List 3) applicable
to the unit or release of concern. An alternative approach is to use unit and waste-(
specific information to identify constituents that are expected to be present, thus^
reducing the number of target constituents (see Section 3.6). The target
constituents selected should be limited to those which may be present in the waste
and have health criteria for the air pathway (see Section 8).
Representative meteorological data as well as emission monitoring results
should be available as input data for dispersion modeling. Therefore, it may be
necessary to conduct an onsite meteorological monitoring survey. The
meteorological monitoring survey should be conducted, at a minimum, for a period
sufficient to identify and define wind and stability patterns for the season
associated with worst-case, long-term source emission/dispersion conditions.
However, it may ajso be desirable to obtain sufficient data to characterize annual
dispersion conditions at the site. The season associated with the highest long-term
air concentration Js determined by evaluating seasonal emission/dispersion
modeling results based on available meteorological data (e.g., National Weather
Service data). This modeling application accounts for the complex relationships
between meteorological conditions and emissions potential and dispersion
potential. For example, high average wind speeds may increase the long-term
emission potential of organics at a surface impoundment, but worst case long-term
dispersion conditions would be associated with low average wind speed conditions.
Seasonal temperature conditions would also affect the emission potential.
12-16
-------
Therefore, it would be necessary to compare seasonal air concentration results to
identify the season with worst case long term exposure conditions. This season
would be the candidate period to collect several months of onsite meteorological
data to support more refined modeling analyses (e.g., dispersion modeling using
emission rate monitoring data as input). Guidance on selection of the emission
monitoring period within this worst case season is presented in Section 12.6.4.2.
Guidance on the conduct of a meteorological monitoring program is provided in
Sections 12.6.3 and 12.8.1.
Dispersion models are used to estimate constituent concentrations based on
source and meteorological monitoring input data. Guidance on the selection and
application of dispersion models is presented in Section 12.5 and in Guidance on Air
Quality Models (U.S. EPA, July 1986) and Procedures for Conducting Air Pathway
Analyses for Superfund Applications (U.S. EPA, December 1988). The results of the
dispersion modeling assessment should then be compared to the health and'
environmental assessment criteria (as previously discussed) to determine
appropriate followup actions.
12.2.1.2.2 Confirmatory Air Monitoring
Confirmatory air monitoring (as outlined in Figure 12-4), may also be
appropriate to provide additional release characterization information for RFI
decisionmaking. Air monitoring data will provide a basis for release mapping and
for evaluation and confirmation of modeling estimates. The conduct of an air
monitoring program should include the following components:
• Develop monitoring procedures
• Condudfinitial monitoring
• Collect and evaluate results
• Conduct additional air monitoring (if necessary)
The development of monitoring procedures should address selection of target
air emission constituents. One acceptable approach is to monitor for all potential
Appendix VIII air emission constituents (See Appendix B, List 3) applicable to the
unit or release of concern. An alternative approach is to use unit and waste-specific
information to identify constituents that are expected to be present, thus reducing
12-17
-------
the number of target monitoring constituents (See Section 3.6). The target
constituents selected should be limited to those which may be present in the waste
and have health criteria for the air pathway (see Section 8.0).
The development of monitoring procedures should also include selection of
appropriate field and analytical methods for conducting the air monitoring
program. Candidate methods and criteria for monitoring program design (e.g.,
relevant to sampling schedule and monitor placement) should be limited to
standard published protocols (such as those available from EPA, NIOSH, and ASTM).
The selection of appropriate methods will be dependent on site and unit-specific
conditions, and is discussed further in Section 12.8.
A limited screening-type sampling program may be appropriate for
determining the design of the air monitoring program. The objective of this
screening sampling will be to verify a suspected release, if appropriate, and to
further assist in identifying and quantifying release constituents of concern.
Screening sampling at each unit for a multiple-unit facility, for example, can be used
to prioritize release sources. The emphasis during this screening will generally be
on obtaining air samples near the source, or collecting a limited number of source
emission samples. The availability of air monitoring data on units with a limited set
of air emission constituents may preclude the need for screening sampling during
the investigation.
An initial air monitoring program should be conducted, as necessary, to
characterize the magnitude and distribution of air concentration levels for the
target constituents selected. Initial monitoring should be conducted for a period
sufficient to characterize air concentrations at the facility property boundary, as
input to the health and environmental assessment (e.g., a 90-day period may be
appropriate for a flat terrain site with minimal variability of dispersion and source
conditions).
The basic approach for the initial air monitoring will consist of collection of
ambient air samples for four target zones: the first zone located upwind of the
source to define background concentration levels; the second zone located
downwind at the unit boundary; the third zone located downwind at the facility
property boundary for input into the health and environmental assessment; and a
12-18
-------
fourth zone offsite, as practical, to determine the need for interim corrective
measures. Multiple monitoring stations will generally be required for each of the
four target zones. It should be noted that offsite air monitoring may not always be
practical due to various problems (e.g., vandalism, public tampering with
equipment, public relations and legal access problems). Dispersion modeling can be
used to estimate offsite concentrations if monitoring data are not available for the
actual receptor locations of interest.
The location of air monitors within each zone should be based on site-specific
diurnal and seasonal wind patterns appropriate for the monitoring period. An
onsite meteorological monitoring survey (as previously discussed) may be necessary
to characterize local wind patterns. The objective of the air monitoring network
should be to provide adequate coverage for primary air flowpaths for each of the
zones enumerated above.
*
The conduct of the initial air monitoring program generally includes the
collection of meteorological data concurrent with air quality measurements. The
meteorological data are needed during the air monitoring program to characterize
emission potential and atmospheric dispersion conditions. This information is also
used to evaluate source/receptor relationships and to interpret and extrapolate the
air monitoring data.
Additional air monitoring may be warranted if initial monitoring program
objectives are not met (e.g., data recovery goals were not adequate) or results are
not adequate to characterize the release (e.g., additional monitoring stations are
needed).
The air monitoring program data should be evaluated, and a dispersion model
used, as needed, to estimate concentrations at the facility property boundary.
These results should then be compared to the health and environmental assessment
criteria (as previously discussed). Subsequent monitoring may also be conducted
during or after the implementation of corrective measures to characterize changes
in downwind release concentrations attributed to mitigation efforts.
12-19
-------
12.3 Characterization of the Contaminant Source and the Environmental Setting
Release investigations can be conducted in an efficient, effective and
representative manner if certain information is obtained prior to implementation
of the effort. This information consists of both waste/unit characterization and
characterization of the environmental setting. Review of information from existing
sources can be used to identify data gaps and to initiate data collection activities to
fill these data gaps. Waste/unit characterization and characterization of the
environmental setting are discussed below:
Waste and unit specific information: Data on the specific constituents
present in the unit that are likely to be released to the air can be used to
design sampling efforts and identify candidate constituents to be
monitored. This information can be obtained from either a review of the
existing information on the waste or from new sampling and analysis.{
6 The manner in which the wastes are treated, stored or disposed may have.
a bearing on the magnitude of air emissions from a unit. In many cases,
this information may be obtained from facility records, contact with the
manufacturer of any control devices, or, in some cases, from the facility's
RCRA permit application.
Environmental setting information: Environmental setting information,
particularly climatological data, is essential in characterizing an air
release. Climatological parameters such as wind speed and temperature
will have a significant impact on the distribution of a release and in
determining whether a particular constituent will be released.
Climatological and meteorological information for the area in which the
facility is located can be obtained either through an onsite monitoring
effort or from the National Climatic Data Center (Asheville, NC). The
climatological data should be evaluated considering site topography and
other local influences that can affect the data representatives.
Information pertaining to the waste, unit, and environmental setting can be
found in many readily available sources. General information concerning
waste/unit characterization is discussed in Section 7. Air specific information is
provided in the following discussions.
12-20
-------
12.3.1 Waste Characterization
Several waste characteristics contribute to the potential for a waste
constituent to be released via the air pathway. These characteristics, in conjunction
with the type of unit and its operation, will determine whether a release will be via
volatilization of the constituent or as particulate entrainment. Major factors
include the types and number of hazardous constituents present, the
concentrations of these constituents in the waste(s), and the chemical and physical
characteristics of the waste and its constituents. All of these factors should be
considered in the context of the specific unit operation involved. It is important to
recognize that the constituents of concern in a particulate release may involve
constituents that are either sorbed onto the particulate, or constituents which
actually comprise the particulate.
%
i
12.3.1.1 Presence of Constituents . ,
The composition of the wastes managed in the unit of concern will influence
the nature of a release to air. Previous studies may indicate that the constituents
are present in the unit or that there is a potential for the presence of these
constituents. In determining the nature of a release, it may be necessary to
determine the specific waste constituents in the unit if this has not already been
done. Guidance on selecting monitoring constituents is presented in Section 3 (and
Appendix B); waste characterization guidance is presented in Section 7.
12.3.1.2 Physical/Chemical Properties
The physical and chemical properties of the waste constituents will affect
whether they will be released, and if released, what form the release will take (i.e.,
vapor, particulate, or paniculate-associated). These parameters are identified in
Table 12-3 as a function of emission and waste type. Important parameters to
consider when assessing the volatilization of a constituent include the following:
• Water solubility. The solubility in water indicates the maximum
concentration at which a constituent can dissolve in water at a given
temperature. This value can help the investigator estimate the
12-21
-------
TABLE 12-3
PARAMETERS AND MEASURES FOR USE IN EVALUATING POTENTIAL
RELEASES OF HAZARDOUS WASTE CONSTITUENTS TO AIR
Emission and Waste Type
A. Vapor Phase Emissions
• Dilute Aqueous
Solution^
- Cone. Aqueous
Solution^
- Immiscible Liquid
-* Solid
B. Particulate Emissions
- Solid
Units of Concern!/
Surface Impoundments,
Tanks, Containers
Tanks, Containers, Surface
Impoundments
Containers, Tanks
Landfills, Waste Piles, Land
Treatment
Useful Parameters
and Measures
Solubility, Vapor Pressure,
Partial Pressure3/
Solubility, Vapor Pressure,
Partial Pressure, Raoults
Law
Vapor Pressure, Partial
Pressure
Vapor Pressure, Partial t
Pressure, Octanol/Water
Partition Coefficient, f
Porosity
Landfills, Waste Piles, Land Particle Size Distribution,
Treatment Unit Operations,
Management Methods
1/ Incinerators are not specifically listed on this table because of the unique issues concerning air emissions
from these units. Although incinerators can burn many forms of waste, the potential for release from
these units is primarily a function of incinerator operating conditions and emission controls, rather than
waste characteristics.
2> Although the octanol/water partition coefficient of a constituent is usually not an important
characteristic in these waste streams, there are conditions where it can be critical. Specifically, in waste
containing high concentrations of organic particulates, constituents with high octanol/water partition
coefficients will adsorb te the particulates. They will become part of the sludge or sediment matrix,
rather than volatilizing from the unit.
3/ Applicable to mixtures of volatile components.
12-22
-------
distribution of a constituent between the dissolved aqueous phase in the
unit and the undissolved solid or immiscible liquid phase. Considered in
combination with the constituent's vapor pressure, solubility can provide
a relative assessment of the potential for volatilization of a constituent
from an aqueous environment.
• Vapor pressure. This property is a measure of the pressure of vapor in
equilibrium with a pure liquid. It is best used in a relative sense;
constituents with high vapor pressures are more likely to be released
than those with low vapor pressures, depending on other factors such as
relative solubility and concentration (e.g., at high concentrations releases
can occur even though a constituent's vapor pressure is relatively low).
• Octanol/water partition coefficient. The octanol/water partition
coefficient indicates the tendency of an organic constituent to sorb to"
organic components of soil or waste matrices. Constituents with high
octanol/water partition coefficients tend to adsorb readily to organic
carbon, rather than volatilizing to the atmosphere. This is particularly
important in landfills and land treatment units, where high organic
carbon content in soils or cover material can significantly reduce the
release potential of volatile constituents.
• Partial pressure. For constituents in a mixture, particularly in a solid
matrix, the partial pressure of a constituent will be more significant than
pure vapor pressure. A partial pressure measures the pressure which
each component of a mixture of liquid or solid substances will exert in
order to enter the gaseous phase. The rate of volatilization of an organic
chemical when either dissolved in water or present in a solid mixture is
characterized by the partial pressure of that chemical. In general, the
greater the partial pressure, the greater the potential for release. Partial
pressure values are unique for any given chemical in any given mixture
and may be difficult to obtain. However when waste characterization
data are available, partial pressure can be estimated using methods
commonly found in engineering and environmental science handbooks.
12-23
-------
• Henry's Law constant. Henry's law constant is the ratio of the vapor
pressure of a constituent to its aqueous solubility (at equilibrium). This
constant can be used to assess the relative ease with which the
compound may vaporize from the aqueous phase. It is applicable only
for low concentration (i.e., less than 10 percent) wastes in aqueous
solution and will be most useful when the unit being assessed is a surface
impoundment or tank containing dilute wastewaters. The potential for
significant vaporization increases is the value for Henry's Law Constant
increases; when it is greater than 10E-3, rapid volatilization will generally
occur.
• Raoult's Law. Raoult's Law accurately predicts the behavior of most
concentrated mixtures of water and organic solvents (i.e., solutions over
10% solute). According to Raoult's Law, the rate of volatilization of each
3 chemical in a mixture is proportional to the product of its concentration
in the mixture and its vapor pressure. Therefore, Raoult's Law can be
used to characterize volatilization potential. This will be especially useful
when the unit of concern entails container storage, tank storage, or
treatment of concentrated waste streams.
A summary of some of these factors for several constituents is given in Tables
12-4 and 12-5. The following document contains a compilation of chemical-physical
properties for several hundred constituents. Additional references for these data
are provided in Section 7.
U.S. EPA. December 1987. Hazardous Waste Treatment Storage and Disposal
Facilities fTSDF) - Air Emission Models. EPA-450/3-87-026. Office of Air Quality
Planning and Standards. Research Triangle Park, N.C. 27711
For airborne particulates, the particle size distribution plays an important role
in both dispersion and actual inhalation exposure. Large particles tend to settle out
of the air more rapidly than small particles. Very small particles (i.e., those that are
less than 2.5 to 10 microns in diameter) are considered to be respirable and thus
present a greater health hazard than the larger particles. Therefore, the source of
the release should be examined to obtain information on particle size. Process
information may be sufficient to grossly characterize the-potential for particulate
12-24
-------
TABLE 12-4
PHYSICAL PARAMETERS OF VOLATILE HAZARDOUS CONSTITUENTS
Hazardous constituent
Acetaldehyde
Acrolein
Acrylonitrile
Allylchloride
Benzene
Benzyl chloride
Carbon tetrachloride
Chlorobenzene
Chloroform
Chloroprene
Cresols
Cumene (isopropyl benzene)
1,4-dichlorobenzene
1,2-dichloroethane
Oichloromethane
Dioxin
Epichlorohydrin
Ethylbenzene
Ethylene oxide
Formaldehyde
Hexachlorobutadiene
Hydrogen cyanide
Hydrogen flouride
Hydrogen sulfide
Hexachlorocyclopentadiene
Maleic anhydride
Methyl acetate
N-Dimethylnitrosamine
Naphthlene
Nitrobenzene
Nitrosomorpholine -
Phenol
Phosgene
Phthalic anhydride
Propylene oxide
1 , 1 ,2,2-tetrachloroethane
Tetrach I oroethy I ene
Toluene
1,1,1-trichloroethane
Trichloroethylene
Vinylchloride
Vinylidenechlonde
Xyienes
Molecular
weight
44
56
53
76.5
78
126.6
154
112
119
88.5
108
120
147
99
85
178
92.5
106
44
30
261
27
20
34
273
98
74
81
123
94
98
148
168
166
92
133
131
62.5
97
106
Vapor pressure
at25°C(mmHg)
915
244
114
340
95
1.21
109
12
192
215
0.4
4.6
1.4
62
360
7.6E-7
13
10
1,095
3,500
0.15
726
900
15,200
0.03
0.3
170
3.4
0.23
0.3
5.3
0.34
1,300
0.03
400
9
15
30
123
90
2,600
500
8.5
Solubility
at25°C(mg/l)
1.00E + 06
4.00E + 05
7.90E + 04
1.78E + 03
1.00
8.00E * 02
5.00E + 02
8.00E + 03
2.00E + 04
50.0
49.00
8.69E + 03
2.00E + 04
3.17E-04
6.00E + 04
152
1.35E + 05
3.00E + 05
1.63E + 05
3.19E + 05
1.90E + 03
9.30E + 04
6.17E + 03
2.90E + 03
200
534
720
1.10E + 03
6.00E + 03
1.00
Henry's Law
constant
(atm"3/mol)
9.50E-05
4.07E-05
8.80E-05
340E-01
5.50E-03
2.00E-02
2.00E-03
3 OOE-03
4.60E-07
2.00E-04
1.00E-04
2.00E-03
1.20E-03
3.08E-05 r
7.00E-03
1.00E-04
1.30E-05
1.02E-05
9.00E-07
2.00E-04
5.00E-03
2.15E-02
8.92E-03
1.90E-01
4 04E-04 .
12-25
-------
TABLE 12-5
PHYSICAL PARAMETERS OF PCB MIXTURES*
Arochlor
(PCB)
1242
1248
1254
1260
Vapor pressure
at 25°C (atm)
2.19E-07
1.02E-07
1.85E-08
5.17E-09
Solubility
at25°C(mg/l)
2400
520
120
30
Henry's Law
constant
(atm-m3/mol)
238E-08
1.02E-08
1.40E-08
6.46E-08
All values estimated based on calculations.
12-26
-------
formation. For example, the presence of ash materials and similar wastes would be
a case in which paniculate emissions would be of concern.
12.3.2 Unit Characterization
Different types of units may have differing release potentials. The particular
type of unit, its configuration, and its operating conditions will have a great effect
on the nature, extent, and rate of the release. These practices or parameters should
be determined and reasonable worst-case operating practices or conditions should
also be identified priorto initial sampling.
12.3.2.1 Type of Unit
The type of unit will affect its release potential and the types of releases
eapected. For the purpose of this guidance, units have been divided into three{
general types with regard to investigating releases to air. Theseare: I
• Area sources having solid surfaces, including land treatment facilities,
surfaces of landfills, and waste piles;
• Point sources, including vents, (e.g., breathing vents from tanks) and
ventilation outlets from enclosed units (e.g., container handling facilities
or stacks); and
• Area sources having liquid surfaces, including surface impoundments and
open-top tanks.
The following discussion provides examples for each of these unit types and
illustrates the kind of data that should be collected prior to establishing a sampling
plan. Table 12-6 indicates types of releases most likely to be observed from each of
these example unit types. It should also be recognized that releases to air can be
continuous or intermittent in nature.
Waste piles-Waste piles are primary sources of particulate releases due to
entrainment into the air of solid particles from the pile. Waste piles are generally
comprised of dry materials which may be released into the air by wind or
12-27
-------
TABLE 12-6
SUMMARY OF TYPICAL UNIT SOURCE TYPE AND AIR RELEASE TYPE
Typical
Unit Type
Waste Piles
Land Treatment
Units
Landfills
Drum Handling
Facilities
Tanks
Surface
Impoundments
Incinerators*
Source Type
Area Sources
with Liquid
Surface
X
X
Area Sources
with Solid
Surface
X
X
X
Point Sources
X
X
X
X
Potential Phase
of Release
Vapor
X
X
X
X
X
X
X
Paniculate
X
X
X
X
X
* Includes units (e.g., garbage incinerators) not covered by 40 CFR Part 264,
SubpartO which pertains to hazardous waste incinerators.
12-28
-------
operational activities. The major air contaminants of concern from waste piles will
be those compounds that are part of or have been adsorbed onto the particulates.
Additionally, volatilization of some constituents may occur. Important unit factors
include the waste pile dimensions (e.g., length, width, height, diameter and shape),
and the waste management practices (e.g., the frequency and manner in which the
wastes are applied to the pile and whether any dust suppression procedures are
employed). The pile dimensions determine the surface area available for wind
erosion. Disturbances to the pile can break down the surface crust and thus increase
the potential for particulate emissions. Dust suppression activities, however, can
help to reduce particulate emissions.
Land treatment units-Liquid or sludge wastes may be applied to tracts of soil
in various ways such as surface spreading of sludges, liquid spraying on the surface,
and subsurface liquid injection. These methods may also involve cultivation or
tilling of the soil. Vapor phase and particulate contaminant releases are influenced'
by the various application techniques. Particulate or volatile emission releases are
most likely to occur during initial application or during tilling, because tilling keeps
the soil unconsolidated and loose, and increases the air to waste surface area.
Important unit factors in assessing an air release from a land treatment unit
include:
• Waste application method - Liquid spraying applications tend to
minimize particulate releases while increasing potential volatile releases.
Subsurface applications generally reduce the potential for particulate
and volatile releases.
• Moisture content of the waste - Wastes with high moisture content will
be less Ifkely to be released as particulates; however, a potential vapor
phase refease may become more likely.
• Soil characteristics - Certain constituents, such as hydrophobic organics,
will be more likely to be bound to highly organic soils than non-organic
soils. Therefore, releases of these types of constituents are most likely to
be associated with particulate emissions.
12-29
-------
Landfills-Landfills can result in participate and vapor phase releases. This
process generally involves placement of waste in subsurface disposal cells and
subsequent covering of the waste with uncontaminated soil. Landfill characteristics
that can affect contaminant release include:
• Porosity and moisture content of the soil or clay covering can influence
the rate at which vapor phase releases move through the soil towards the
surface. Finer soils with lower porosities will generally slow movement of
vapors through the unit. The frequency of applying soil cover to the
open working face of a landfill will also affect the time of waste
exposure to the air.
• Co-disposal of hazardous and municipal wastes will often increase the
potential for vapor phase releases, because biodegradation of municipal
& wastes results in the formation of methane gas as well as other volatile*
organics. Methane gas may act as a driving force for release of other
volatile hazardous components that may be in the unit (See Section 11 -
Subsurface Gas.)
• Landfill gas vents, if present, can act as sources of vapor phase emissions
of contaminated landfill gases.
• Leachate collection systems can be sites of increased vapor phase
emissions due to the concentrated nature of the leachate collected.
Open trenches are more likely to be emission sources than underground
collection sumps due to the increased exposure to the atmosphere.
• Waste mixing or consolidation areas where bulk wastes are mixed with
soil or "other materials (e.g., fly ash) prior to landfilling can be
contributors to both particulate and vapor phase air releases. Practices
such as spreading materials on the ground to release moisture prior to
landfilling will also increase exposure to the atmosphere.
Drum handling facilities-Emissions from drum or container handling areas can
result from several types of basic operations. Frequently, emissions from these
operations are vented to the air through ducts or ventilation systems. Air sampling
12-30
-------
to assess emissions from these operations may include sampling of the control
device outlets, the workplace atmosphere at each operation, or the ambient air
downwind of the unit. Factors which effect emissions include:
• Filling operations can be a major source of either vapor or particulate
emissions due to agitation of the materials during the filling process.
Spillage which occurs during loading may also contribute to emissions.
Organic waste components with high volatility will readily vaporize into
the air. Similarly, particulate matter can be atmospherically entrained by
agitation and wind action. The emission potential of filling-operations
will be affected by exposure to ambient air. Generally, fugitive emissions
from an enclosed building will be less than emissions created during
loading in an open structure.
,* • Cleaning operations can have a high potential for emissions. These'
, emissions may be enhanced by the use of solvents or steam cleaning
equipment. The waste collection systems at these operations usually
provide for surface runoff to open or below ground sumps, which can
also contribute to air emissions.
• Volatilization of waste components can also occur at storage units. Since
it is common practice to segregate incompatible wastes during storage,
the potential for air releases may differ within a storage unit depending
on the nature of the wastes stored in any particular area. The most
common source of air emission releases from drum storage areas is spills
from drums ruptured during shipping and handling.
• For offsite facilities, storage areas frequently are located where drums
are sampled during the waste testing/acceptance process. This process
involves drum opening for sampling and could also include spillage of
waste materials on the ground or floor.
Important release information includes emission rates, and data to estimate
release rise (e.g., vent height and diameter as well as vent exit temperature and
velocity). Information pertaining to building dimension/orientation of the unit and
nearby structures is needed to assess the potential for aerodynamic behavior of the
12-31
-------
stack/vent release. These input data would be needed if atmospheric dispersion
modeling was necessary.
Tanks-Tanks can emit volatile waste components under various circumstances.
A major determinant of any air emission will be the type of tank being studied.
Closed or fixed roof storage tanks will most likely exhibit less potential for air
emissions than open topped tanks. Some tanks are equipped with vapor recovery
systems that are designed to reduce emissions. Important process variables for
understanding air emissions from tanks can be classified as descriptive and
operational variables:
• Descriptive variables include type, age, location, and configuration of the
tank.
, • Operational variables include aeration, agitation, filling techniques/
r surface area, throughput, operating pressure and temperature, sludge
removal technique and frequency, cleaning technique.and frequency,
waste retention and vent pipe dimensions and flow rate.
Important release information includes emission rates, and data to estimate
plume rise (e.g., height and diameter as well as exit temperature and velocity).
Information pertaining to building dimensions/orientation of the unit and nearby
structures is needed to assess the potential for aerodynamic behavior of the
stack/vent release. These input data would be needed if atmospheric dispersion
modeling was necessary.
Surface impoundments-Surface impoundments are similar in many ways to
tanks in the manner in which air emissions may be created. Surface impoundments
are generally larger, at least in terms of exposed surface areas, and are generally
open to the atmosphere. The process variables important for the evaluation of
releases to air from surface impoundments can also be classified as descriptive and
operational.
• Descriptive parameters include dimensions, including length, width, and
depth, berm design, construction and liner materials used, and the
location of the unit on the site.
12-32
-------
• Operational parameters include freeboard, filling techniques (in
particular, splash versus submerged inlet), depth of liquid and sludge
layers, presence of multiple liquid layers, operating temperature, sludge
removal techniques and frequency, cleaning technique and frequency,
presence of aerators or mixers, biological activity factors for
biotreatment, and the presence of baffles, oil layers, or other control
measures on the liquid surface. (These factors are relevant to some tanks
as well.)
Some surface impoundments are equipped with leak collection systems that
collect leaking liquids, usually into a sump. Air emissions can also occur from these
sumps. Sump operational characteristics and dimensions should be documented
and, if leaks occur, the volume of material entering the sump should be
documented. (These factors are relevant to some tanks as well.)
I
Incinerators - Stack emissions from incinerators (i.e., incinerator units not
addressed by RCRA in Part 264, Subpart 0, e.g., municipal refuse incinerators) can
contain both particulates and volatile constituents. The high temperatures of the
incineration process can also cause volatilization of low vapor pressure organics and
metals. Additional volatile releases can occur from malfunctioning valves during
incinerator charging. The potential for air emissions from these units is primarily a
function of incinerator operating conditions and emission controls. Important unit
release information includes emission rates, and data to estimate plume rise (e.g.,
height and diameter as well as exit temperature and velocity), as well as building
dimensions/orientation of the unit and nearby structures. This information is
needed to assess the aerodynamic behavior of the stack/vent release and for input
to atmospheric dispersion models.
*L
12.3.2.2 Size'ofUnit
The size of the unit(s) of concern will have an important impact on the
potential magnitude of a release to air. The release of hazardous constituents to
the air from an area source is often directly proportional to the surface area of the
unit, whether this surface area is a liquid (e.g., in a tank) or a solid surface (e.g., a
land treatment unit). The scope of the air investigation may be a function of the
12-33
-------
size of the unit. Generally, more sampling locations will be required as the unit
increases in size, due primarily to increased surface area. Also, as the total amount
of waste material present in a particular unit increases, it will represent a larger
potential reservoir or source of constituents which may be released.
Scaling factors, such as surface area to volume ratios should also be evaluated.
One large waste pile, for instance, can exhibit a lower ratio of surface area to total
volume than the sum of two smaller piles in which the total volume equals that of
the larger pile. Other units such as tanks may exhibit a similar economy of surface
area, based on the compact geometry of the unit.
Because releases to air generally occur at the waste/atmosphere interface,
surface area is generally a more important factor than total waste volume.
Consequently, operations that increase the atmosphere/waste interface', such as
agitation or aeration, splash filling, dumping or filling operations, and spreading*
operations will tend to increase the emission rate. Total emissions, however, will be
a function of the total mass of the waste constituent(s) and the.duration of the
release.
For point sources, the process or waste throughput rate will be the most
important unit information needed to evaluate the potential for air emissions (i.e.,
stack/vent releases).
12.3.2.3 Control Devices
The presence of air pollution control devices on units can have a major
influence on the nature and extent of releases. Control devices can include wet or
dry scrubbers, electrostatic precipitators, baghouses, filter systems, wetting
practices for solid materials, oil layers on surface impoundments, charcoal or resin
absorption systems, vapor flares, and vapor recovery systems. Many of these
controls systems can be installed on many of the unit types discussed in this section.
Due to the variety of types of devices and the range of operational differences, an in
depth discussion of individual control devices is not presented here. Additional
information on control technologies for hazardous air pollutants is available in the
following references:
12-34
-------
U.S. EPA. 1986. Handbook - Control Technologies for Hazardous Air
Pollutants. EPA/625/6-86/014. Office of Research and Development. Research
Triangle Park, N.C. 27711.
U.S. EPA. 1986. Evaluation of Control Technologies for Hazardous Air
Pollutants: Volume 1 - Technical Report. EPA/600/7-86/009a. NTIS PB 86-
167020. Volume 2 - Appendices. EPA/600/7-86/009b. NTIS PB 86-167038.
Off ice of Research and Development. Research Triangle Park, N.C. 27711.
If a control device is present on the unit of concern, descriptive and
operational characteristics of the unit/control device combination should be
reviewed and documented. In many cases, performance testing of these devices has
been conducted after their installation on the unit(s). Information from this testing
may help to quantify releases to air from the unit(s); however, this testing may not
have been performed under a "reasonable worst-case" situation. The conditions'
under which the testing was performed should be documented.
12.3.2.4 Operational Schedules
Another characteristic which can affect the magnitude of a release to air from
a unit is the unit's operational schedule. If the unit is operational on a part time or
batch basis, the emission or release rate should be measured during both
operational and non-operational periods. In contrast to batch operations, emission
or release rates from continuous waste management operations may be measured
at any time.
12.3.2.5 Temperature of Operation
Phase changes~bf liquids and solids to gases is directly related to temperature.
Therefore, vapor phase releases to air are directly proportional to process
temperature. Thus, it is important to document operational temperature (i.e.,
waste temperature) and fluctuations to enhance the understanding of releases to
air from units. Particular attention should be paid to this parameter in the review of
existing data or information regarding the operation of the unit.
12-35
-------
The release rate of volatile components also generally increases with
temperature. Frequently, the same effect is observed for particulates, because
entrainment is enhanced as materials are dried. Thus, the evaporation of any water
from solids, which generally increases as temperature increases, will likely increase
the emissions of many particulates in the waste streams. Evaporation of water may
also serve to concentrate wastes, leading to conditions more conducive to vapor
phase releases to air. It should also be noted that the destruction efficiency of
incinerators is also a function of temperature (i.e., higher temperatures are
generally associated with greater destruction efficiency).
12.3.3 Characterization of the Environmental Setting
Environmental factors can influence not only the rate of a release to air but
also the potential for exposure. Significant environmental factors include climate,
s«il conditions, terrain and location of receptors. These factors are discussed below.'
12.3.3.1 Climate
Wind, atmospheric stability and temperature conditions affect emission rates
from area sources as well as atmospheric dispersion conditions for both area and
point sources. Historical summaries of climatic factors can provide a basis to assess
the long-term potential for air emissions and to characterize long-term ambient
concentration patterns for the areNa. Short-term measurements of these conditions
during air monitoring will provide the meteorological data needed to interpret the
concurrent air quality data. Meteorological monitoring procedures are discussed in
Section 12.8. Available climatic information, on an annual and monthly or seasonal
basis, should be collected for the following parameters:
• Wind direction and roses (which affects atmospheric transport, and can
be used to determine the direction and dispersion of release migration);
• Mean wind speeds (which affects the potential for dilution of releases to
air);
• Atmospheric stability distributions (which affects dispersion conditions);
12-36
-------
• Temperature means and extremes (which affects the potential for
volatilization, release rise and wind erosion);
• Precipitation means (which affects the potential for wind erosion of
particulates);
• Atmospheric pressure means (which affects the potential for air
emissions from landfills); and
• Humidity means (which can affect the air collection efficiencies of some
adsorbents - see Section 12.8).
The primary source of climate information for the United States is the National
Climatic Data Center (Asheville, NC). The National Climatic Data Center can provide
climate summaries for the National Weather Service station nearest to the site of
interest. Standard references for climatic information include the following:
National Climatic Data Center. Local Climatoloqical Data - Annual Summaries
with Comparative Data, published annually. Asheville, NC 28801.
National Climatic Data Center. Climates of the States. 1973. Asheville, NC
28801.
National Climatic Data Center. Weather Atlas of the United States. 1968.
Asheville, NC 28801.
The climatological data should be evaluated considering the effects of
topography and other local influences that can affect data representativeness.
t
A meteorological monitoring survey may be conducted prior to ambient air
monitoring to establish the local wind flow patterns and for determining the
number and locations of sampling stations. The survey results will be used to
characterize local prevailing winds and diurnal wind flow patterns (e.g., daytime
upslope winds, nighttime downslope winds, sea breeze conditions) at the site. The
survey should be conducted for a one-month period and possibly longer to
adequately characterize anticipated wind patterns during the air monitoring
12-37
-------
program. Inland, flat terrain conditions may not necessitate an onsite
meteorological monitoring survey if representative data are available from previous
onsite studies or from National Weather Service stations.
The meteorological monitoring data collected during the initial monitoring
phase can serve as a basis for the placement of air sampling stations during any
subsequent monitoring phases.
12.3.3.2 Soil Conditions
Soil conditions (e.g., soil porosity) can affect air emissions from landfills and
the particulate wind erosion potential for contaminated surface soils. Soil
conditions pertinent to characterizing the potential for air emissions include the
following:
t
* t
• Soil porosity (which affects the rate of potential gaseous emissions);
• Particle size distribution (which affects the potential for particulate
emissions from contaminated soils); and
• Contaminant concentrations in soil (i.e., potential to act as air emission
sources).
Soil characterization information is presented in Section 9.
12.3.3.3 Terrain
Terrain features can significantly influence the atmospheric transport of air
emissions. Terrain heights relative to release heights will affect groundlevel
concentration. Terrain obstacles such as hills and mountains can divert regional
winds. Likewise, valleys can channel wind flows and also limit horizontal dispersion.
In addition, complex terrain can result in the development of local diurnal wind
circulations and affect wind speed, atmospheric turbulence and stability conditions.
Topographic maps of the facility and adjacent areas are needed to assess local and
regional terrain. Guidance on the appropriate format and sources of topographic
and other maps is presented in Section 7 and Appendix A.
12-38
-------
12.3.3.4 Receptors
Information concerning the locations of nearby buildings and the population
distribution in the vicinity of the site are needed to identify potential air-pathway
receptors. This receptor information provides a basis for determining the need for
interim corrective measures. Both environmental and human receptor information
is needed to assess potential air-pathway exposures. Such information may include:
• A site boundary map;
• Location of nearest buildings and residences for each of the sixteen 22.5
degree sectors which corresponds to major compass points (e.g., north,
north-northwest);
• Location of buildings and residences that correspond to the area or
maximum offsite groundlevel concentrations based. on preliminary
modeling estimates (these locations may not necessarily be near the site
boundary for elevated releases); and
• Identification of nearby sensitive receptors (e.g., nursing homes,
hospitals, schools, critical habitat of endangered or threatened species).
The above information should be considered in the planning of an air
monitoring program. Additional guidance on receptor information is provided in
Section 2.
12.3.4 Review of Existing Information
s_
The review of existing air modeling/monitoring data entails both summarizing
the reported air contaminant concentrations as well as evaluating the quality of
these data. Air data can be of many varieties and of varying utility to the RFI
process. Modeling data should be evaluated based on the applicability of the model
used, model accuracy, as well as the quality and representativeness of the input
data.
12-39
-------
One of the most basic parameters to review in any type of air monitoring data
should be the validity of the sampling locations used during the collection of the
monitoring data. The results of previous investigations should be assessed with
respect to the upwind-downwind pattern around the unit to determine the
likelihood that the sampling devices would have measured releases from the unit of
concern. For relatively simple sites (e.g., flat terrain, constant wind speed and
direction), this determination should be fairly straight-forward; however, for
complex sites (e.g., complex terrain, variable winds, multiple sources, etc.), assessing
the appropriateness of past sampling locations should consider such factors as
potential interferences that may not have been addressed by the sampling scheme.
The most useful monitoring data are compound-specific results which can be
associated with the unit being investigated, or, for point sources (such as vent stacks
or ventilation system outlets), direct measurements of the exhaust prior to its
release into the atmosphere. Because the hazardous properties and health andj
environmental criteria are compound-specific, general compound category or classj
data (e.g., hydrocarbon results) are less meaningful. Any existing air data should
also be described and documented as to the sampling and analysis methods utilized,
the associated detection limits, precision and accuracy, and the results of QA/QC
analyses conducted. Results reported as non-detected (i.e., not providing numerical
detection limits) are likely to be of no value.
In addition, available upwind and downwind air data should be evaluated to
determine if the contamination is due to releases from the unit. If background data
are available for the unit of concern, the data will be of much greater use in the
planning of additional air monitoring tasks. Upwind data (to characterize ambient
air background levels) are important for evaluating if downwind contamination can
be attributed to the unit of concern. If background data are not available, the
existing downwind air concentration data will be of less value in characterizing a
release; however, the lack of background data does not negate the utility of the
available monitoring data.
Data may also be available from air monitoring studies that did not focus
directly on releases from a unit of concern. Many facilities conduct onsite health
and safety programs, including routine monitoring of air quality for purposes of
evaluating worker exposure. This type of data may include personnel hygiene
12-40
-------
monitoring results from personal sampling systems worn by employees as they
perform their jobs, general area monitoring of zones at which hazardous
operations are conducted, or actual unit-emission monitoring. The detection limits
of these methods (generally in parts per million) are frequently higher than are
needed for RFI purposes. However, this type of industrial hygiene monitoring is
frequently compound-specific, and can be useful in qualitatively evaluating the air
emissions from particular sources.
Indoor air monitoring, generally only applicable to units that are enclosed in a
building (e.g., drum handling areas or tanks), often includes flow monitoring of the
ventilation system. Monitoring of hoods and ductwork systems may have been
conducted to determine exchange time and air circulation rates. These flow
determinations could prove to be useful in the evaluation of air emission
measurements during the RFI.
Another important aspect of the existing data review is to document any
changes in composition of the waste managed in the unit of concern since the air
data were collected. Also, changes in operating conditions or system configuration
for waste generation and/or unit functions could have major effects on the nature
or extent of releases to air. If such operational or waste changes have occurred,
they should be summarized and reviewed to determine their role in the evaluation
of existing data. This summary and review will not negate the need to take new
samples to characterize releases from the unit. However, such information can be
useful in the planning of the new air monitoring activities.
12.3.5 Determination of "Reasonable Worst-Case" Exposure Period
A "reasonable worst-case" exposure period over a 90 day period should be
identified if an airmonitoring program is to be conducted. Determination of
reasonable worst-case exposure conditions will aid in planning the air monitoring
program and is dependent on seasonal variations in emission rates and dispersion
conditions.
The selection of the "reasonable worst-case" 90-day exposure period for the
conduct of air monitoring should account for the following factors:
•12-41
-------
• For vapor phase releases, wind speed and temperature are the key
factors affecting releases from the unit. In general, the higher the
temperature and windspeed, the greater the rate of volatilization of
constituents of concern from the waste. This process is tempered,
however, by the fact that at higher windspeeds, dispersion of the release
is generally greater, resulting in lower downwind concentrations at
potential exposure points.
• For paniculate releases, wind speed is the key meteorological factor. The
amount of local precipitation contributing to the degree of moisture of
the waste may also be important. In general, the higher the windspeed,
and the drier the waste, the greater will be the potential for particulate
release. As with vapor phase releases, higher wind speeds may also lead
to greater dispersion of the release, resulting in lower downwind
v concentrations.
!
• For point source releases, increased wind speeds and unstable
atmospheric conditions (e.g., during cloudless days) enhance dispersion
but also tend to reduce plume height and can lead to relatively high
groundlevel concentrations.
• Constituent concentrations at any downwind sector will also be directly
affected by the wind direction and frequency.
Air emission release rate models and atmospheric dispersion models can be
used to identify reasonable worst-case exposure conditions (i.e., to quantitatively
account for the above factors). For this application, it is recommended that the
modeling effort be limited to a screening/sensitivity exercise with the objective of
obtaining "relative".results for a variety of source and meteorological scenarios. By
comparing results in a relative fashion, only those input meteorological parameters
of greatest significance (e.g., temperature, wind speed and stability) need to be
considered.
In general, the summer season will be the "reasonable worst-case" exposure
period at most sites because of relatively high temperatures and low windspeeds.
Spring and fall are also candidate monitoring seasons that should be evaluated on a
12-42
-------
site-specific basis. Winter is generally not a prime season for air monitoring due to
lower temperatures and higher wind speeds.
12.4 Air Emission Modeling
12.4.1 Modeling Applications
Air emission models can be used to estimate constituent-specific emission rates
abased on waste/unit input data for many types of waste management units. (An
emission rate is defined as the source release rate for the air pathway in terms of
mass per unit of time.)
An important application of emission models in the RFI release
characterization strategy for air is the conduct of screening assessments. For this
application, available waste/unit input data for emission models, in conjunction
with dispersion modeling results, are used to estimate concentrations at locations of
interest. These results can then be evaluated to determine if adequate information
is available for RFI decisionmaking or if monitorinc is needed to further reduce the
uncertainty associated with characterizing the release. Depending on the degree of
uncertainty in the estimated concentrations relative to the differences between the
estimated concentrations and the health based levels, modeling results may be
sufficient to characterize the release as significant (i.e., implementation of
corrective action would be appropriate) or as insignificant (i.e., no further action is
warranted).
Emission rate models can also be used to identify potential major air emission
sources at a facility (especially multiple-unit facilities). For this type of application,
modeling results are used to compare routine long-term emissions from various
units to prioritize the need for release characterization at each unit. For example,
modeling results may indicate that 90 percent of the volatile organic compound
emissions at a facility are attributable to surface impoundment units and only 10
percent to other sources. Therefore, emphasis should be on characterizing releases
from the surface impoundments.
Emission modeling is not available for all air-related phenomenon associated
with waste management. For example, anaerobic biological activity in surface
12-43
-------
impoundments may, in certain instances, contribute to air pollution by emitting
constituents not contained in the waste placed in the impoundment and which
available models do not adequately address. In such instances, source testing or
monitoring may be necessary; based on such monitoring, emission rates can be
developed.
12.4.2 Model Selection
The information gathered during the initial stage of the air investigation
should be used to select appropriate models and to estimate unit-specific and
constituent-specific emission rates. A thorough understanding of the available
models is needed before selecting a model for an atypical emission source. When
gathering information on any emission source, it would be useful to obtain a
perspective of the potential variability of the waste and unit input data. A
sensitivity analysis of this variability relevant to emission rate estimates would help
determine the level of confidence associated with the emission modeling results.
Air emission models can be classified into two categories; models which can be
used to estimate volatile organic releases, and models which can be used to
estimate particulate emissions. These are discussed below.
12.4.2.1 Organic Emissions
Comprehensive guidance on the application of air emission models for volatile
organic releases from various units is presented in the following references:
U.S. EPA. December 1987. Hazardous Waste Treatment. Storage, and Disposal
Facilities rTSDF) - Air Emission Models. EPA-450/3-87-026. Office of Air Quality
Planning and*Standards. Research Triangle Park, NC 27711.
U.S. EPA. December 1988 Draft. Procedures for Conducting Air Pathway
Analyses for Superfund Applications. Office of Air Quality Planning and
Standards. Research Triangle Park, NC 27711.
12-44
-------
These references provide modeling guidance for the following units:
• Surface impoundments
Storage impoundments
Disposal impoundments
Mechanically aerated impoundments
Diffused air systems
Oil film surfaces
• Land treatment
Waste application
Oil film surfaces
Tilling
. • Landfills
t
Closed landfills
Fixation pits
Open landfills
• Waste piles
• Transfer, storage and handling operations
Container loading
Container storage
Containercleaning
Stationary tank loading
Stationary tank storage
Spills
Fugitive emissions
Vacuum truck loading
Emission factors for various evaporation loss sources (e.g., storage and handling of
organic liquids) are provided in the following reference:
12-45
-------
U.S. EPA. 1985. (Fourth edition and subsequent supplements) Compilation of
Air Pollutant Emission Factors. EPA AP-42. NTIS PB 86-124906. Office of Air
Quality Planning and Standards. Research Triangle Park, NC 27711.
An emission factor is generally defined as an average value which relates the
quantity of a pollutant released to the atmosphere with the activity associated with
the release of the pollutant. However, for estimation of organic releases from
storage tanks, the emission factors are presented in terms of empirical formulae
which can relate emissions to such variables as tank diameter, liquid temperature,
etc.
Selection of an appropriate air emission model will be based primarily on
selection of a model which is appropriate for the unit of concern, has technical
credibility and is practical to use. Some of the models presented in Hazardous
Waste Treatment. Storage and Disposal Facilities (TSDF) - Air Emission Models (U.S.'
EPA, December 1987), are available on a diskette for use on a microcomputer.
Computer-compatible air emission models (referred to as CHEMDAT6 models) are
available for the following sources.
• Nonaerated impoundments
• Open tanks
• Aerated impoundments
• Land treatment
• Landfills
These models are prime candidates for RFI air release characterization applications.
12.4.2.2 Paniculate Emissions
s^
Guidance on "the selection and application of air emission models for
particulate releases is presented in the following references:
U.S. EPA. February 1985. Rapid Assessment of Exposure to Particulate
Emissions from Surface Contamination Sites. EPA/600-18-85/002. Office of
Health and Environmental Research. Washington, D.C. 20460.
12-46
-------
U.S. EPA. 1985. (Fourth edition and subsequent supplements) Compilation of
Air Pollutant Emission Factors. EPA, AP-42. Office of Air Quality Planning and
Standards. Research Triangle Park, NC 27711.
U.S. EPA. 1978. Fugitive Emissions from Integrated Iron and Steel Plants. EPA
600/2-78-050. Washington, D.C. 20460.
U.S. EPA. December 1988 Draft. Procedures for Conducting Air Pathway
Analysis for Superfund Applications. Office of Air Quality Planning and
Standards. Research Triangle Park, NC 27711.
These references provide modeling guidance for the following particulate
sources and associated operations and activities (e.g..vehicular traffic):
• Wastepiles
• Flat, open surfaces
The air emission models for both types of sources should account for both
wind erosion potential as well as releases due to mechanical disturbances.
The U.S. EPA-Office of Air Quality Planning and Standards is currently
developing guidance regarding particulate emissions from hazardous waste
transfer, storage and disposal facilities.
12.4.3 General Modeling Considerations
Organics in surface impoundments, land treatment facilities, landfills, and
wastepiles, can depart through a variety of pathways, including volatilization,
biological decomposition, adsorption, photochemical reaction, and hydrolysis. To
allow reasonable estimates of organic disappearance, it is necessary to determine
which pathways predominate for a given chemical, type of unit, and set of
meteorological conditions.
Source variability will significantly influence the relative importance of the
pathways. For highly variable sources it may be possible to exclude insignificantly
small pathways from consideration. The relative magnitude of these pathways then
12-47
-------
can be computed by applying the methodology to a model facility to determine
relative differences among various compounds. A summary of typical pathways for
air emission sources is presented in Table 12-7.
It is also necessary to consider the variation of waste composition as a function
of time as well as other potential variations in source conditions. These variable
conditions may necessitate multiple modeling scenarios to adequately characterize
representative waste/unit conditions.
12.5 Dispersion Modeling
12.5.1 Modeling Applications
Atmospheric dispersion models can be used to estimate constituent-specific
concentrations at locations of interest based on input emission rate and
meteorological input data. The major RFI dispersion modeling applications for
characterizing releases to air can be summarized as follows:
• Screening assessments: Dispersion models can be used to estimate
concentrations at locations of interest using input emission rate data
based on air emission modeling.
• Emission monitoring: Dispersion models can be used to estimate
concentrations at locations of interest using input emission rate data
based on emission rate monitoring.
• Confirmatory air monitoring: Dispersion modeling can be used to assist
in designing an air monitoring program (i.e., to determine appropriate
SI
monitoring locations and monitoring period) as well as for interpretation
and extrapolation of monitoring results.
Atmospheric dispersion models can be used for monitoring program design
applications to identify areas of high concentration relative to the facility property
boundary or actual receptor locations. High concentration areas which correspond
to actual receptors are priority locations for air monitoring stations.
12-48
-------
TABLE 12-7
TYPICAL PATHWAYS FOR AREA EMISSION SOURCESa
Pathway
Volatilization
Biodegradation
Photodecomposition
Hydrolysis
Oxidation/reduction
Adsorption
Hydroxyl radical reaction
Migrationb
Runoff b
Surface
Impoundments
I
I
S
s
N
N
N
N
N
Land
Treatment
I
I
N
N
N
N
N
N
N
Landfill
I
S
N
N
N
N
N
N
N
I = Important
S = Secondary
N « Negligible or not applicable
a Individual chemicals in a given site type may have dominant pathways
different from the ones shown here.
b Water migration and runoff are considered to have negligible effects on
ground and surface water in a properly sited, operated, and maintained
RCRA-permitted hazardous waste treatment, storage, and disposal facility.
12-49
-------
Dispersion models (with input emission rates based on emission models) can
also be used to provide seasonal air concentration "patterns" based on available
representative historical meteorological data (either onsite or offsite). Comparison
of seasonal air concentration patterns can be used to identify the "reasonable worst
case" period for monitoring. Air concentration patterns based on modeling results
can similarly be used to evaluate the representativeness of the actual data collection
period. Representativeness is determined by comparing the air concentration
patterns for the actual air monitoring period with historic seasonal air
concentration patterns.
The objective of the modeling applications discussed above involves the
estimation of long-term (i.e., several months to years) concentration patterns.
These long-term patterns do not have the variability associated with short-term
(i.e., hours to days, such as a 24-hour event) emission rate and dispersion conditions,*
and are more conducive to data extrapolation applications. For example, neart
source and fenceline air monitoring results can be used to back calculate an?
emission rate for the source. This estimated emission rate can be used as dispersion
modeling input to estimate offsite air concentrations for the same downwind sector
and exposure period as for the air monitoring period.
12.5.2 Model Selection
Guidance on the selection and application of dispersion models is provided in
the following references:
U.S. EPA. July 1986. Guidelines on Air Quality Models (Revised). EPA-450/12-
78-027R. NTtS PB86-245248. Office of Air Quality Planning and Standards.
Research Triangle Part, NC 27711.
U.S. EPA. December 1988 Draft. Procedures for Conducting Air Pathway
Analyses for Superfund Applications. Office of Air Quality Planning and
Standards. Research Triangle Park, NC 27711.
The following information is based primarily on guidance provided in these
references.
12-50
-------
12.5.2.1 Suitability of Models
The extent to which a specific air quality model is suitable for the evaluation of
source impact depends upon several factors. These include: (1) the meteorological
and topographic complexities of the area; (2) the level of detail and accuracy
needed for the analysis; (3) the technical competence of those undertaking such
simulation modeling; (4) the resources available; and (5) the detail and accuracy of
the data base, i.e., emissions inventory, meteorological data, and air quality data.
Appropriate data should be available before any attempt is made to apply a model.
A model that requires detailed, precise, input data should not be used when such
data are unavailable. However, assuming the data are adequate, the greater the
detail with which a model considers the spatial and temporal variations in emissions
and meteorological conditions, the greater the ability to evaluate the source impact
e
afid to distinguish the effects of various control strategies.
' ?
Air quality models have been applied with the most accuracy or the least
degree of uncertainty to simulations of long term averages in areas with relatively
simr'e topography. Areas subject to major topographic influences experience
meteorological complexities that are extremely difficult to simulate. Although
models are available for such circumstances, they are frequently site-specific and
resource intensive. In the absence of a model capable of simulating such
complexities, only a preliminary approximation may be feasible until such time as
better models and data bases become available.
Models are highly specialized tools. Competent and experienced personnel
are an essential prerequisite to the successful application of simulation models. The
need for specialists is critical when the more sophisticated models are used or the
e_
area being investigated has complicated meteorological or topographic features. A
model applied improperly, or with inappropriately chosen data, can lead to serious
misjudgments regarding the source impact or the effectiveness of a control
strategy.
The resource demands generated by use of air quality models vary widely
depending on the specific application. The resources required depend on the
nature of the model and its complexity, the detail of the data base, the difficulty of
12-51
-------
the application, and the amount and level of expertise required. The costs of
manpower and computational facilities may also be important factors in the
selection and use of a model for a specific analysis. However, it should be
recognized that under some sets of physical circumstances and accuracy
requirements, no present model may be appropriate. Thus, consideration of these
factors should not lead to selection of an inappropriate model.
12.5.2.2 Classes of Models
Dispersion models can be categorized into four generic classes: Gaussian,
numerical, statistical or empirical, and physical. Within these classes, especially
Gaussian and numerical models, a large number of individual "computational
algorithms" may exist, each with its own specific applications. While each of the
algorithms may have the same generic basis, e.g., Gaussian, it is accepted practice to
refer to them individually as models. In many cases the only real difference'
between models within the different classes is the degree of detail considered in
{
the input or output data.
Gaussian models are the most widely used techniques for estimating the
impact of nonreactive pollutants. Numerical models may be more appropriate than
Gaussian models for area source urban applications that involve reactive pollutants,
but they require much more extensive input data bases and resources and therefore
are not as widely applied. Statistical or empirical techniques are frequently
employed in situations where incomplete scientific understanding of the physical
and chemical processes or lack of the required data bases make the use of a
Gaussian or numerical model impractical.
Physical modeling, the fourth generic type, involves the use of wind tunnel or
other fluid modeling facilities. This class of modeling is a complex process requiring
a high level of technical expertise, as well as access to the necessary facilities.
Nevertheless, physical modeling may be useful for complex flow situations, such as
building, terrain or stack down-wash conditions, plume impact on elevated terrain,
diffusion in an urban environment, or diffusion in complex terrain. It is particularly
applicable to such situations for a source or group of sources in a geographic area
limited to a few square kilometers. The publication "Guideline for Fluid Modeling
12-52
-------
of Atmospheric Diffusion" provides information on fluid modeling applications and
the limitations of that method (U.S. EPA, 1981).
12.5.2.3 Levels of Sophistication of Models
In addition to the various classes of models, there are two levels of
sophistication. The first level consists of general, relatively simple estimation
techniques that provide conservative estimates of the air quality impact of a specific
source, or source category. These are screening techniques or screening models.
The purpose of such techniques is to eliminate the need for further more detailed
modeling for those sources that clearly can be characterized and evaluated based
on simple screening assessments.
The second level consists of those analytical techniques that provide more
detailed treatment of physical and chemical atmospheric processes, require more
detailed and precise input data, and provide more specialized concentration*
estimates. As a result they provide a more refined and, at least theoretically, a more
accurate estimate of source impact and the effectiveness of control strategies.
These are referred to as refined models.
The use of screening techniques followed by a more refined analysis is always
desirable, however, there are situations where the screening techniques are
practically and technically the only viable option for estimating source impact. In
such cases, an attempt should be made to acquire or improve the necessary data
bases and to develop appropriate analytical techniques.
12.5.2.4 Preferred Models
Guidance ontPA preferred models for screening and refined applications is
provided in the following references:
U.S. EPA. July 1986. Guidelines on Air Quality Models (Revised). EPA-450/2-78-
027R. NTIS PB86-245248. Office of Air Quality Planning and Standards.
Research Triangle Park, N.C. 27711.
12-53
-------
U.S. EPA. October 1977. Guidelines for Air Quality Maintenance Planning and
Analysis. Vol. 10 (Revised): Procedures for Evaluating Air Quality Impact of
New Stationary Sources. EPA-450/4-77-001. NTIS PB274-087. Office of Air
Quality Planning and Standards. Research Triangle Park, N.C. 27711.
U.S. EPA. December 1988 Draft. Procedures for Conducting Air Pathway
Analyses for Superfund Applications. Office of Air Quality Planning and
Standards. Research Triangle Park, NC 27711.
Appropriate dispersion models commensurate with the above guidance and
sutiable for mainframe computer use are included in the UNAMAP series available
from NTIS. Versions of the UNAMAP models suitable for use on a microcomputer
are also available from commercial sources.
Alternative screening approaches based on hand calculations are available for
point sources located in flat terrain based on the following guidance: l
t
Turner, D.B. 1969. Workbook of Atmospheric Dispersion Estimates. Public
Health Service. Cincinnati, OH.
U.S. EPA. March 1988 Draft. A Workbook of Screening Techniques for
Assessing Impacts of Toxic Air Pollutants. Office of Air Quality Planning and
Standards. Research Triangle Park, NC 27711.
Preferred models for selected applications in simple terrain are identified in
Table 12-8. Appropriate dispersion models for complex terrain applications
generally need to be determined on a case-by-case basis. Acceptable models may
not be available for many complex terrain applications.
t
The use of the Industrial Source Complex (ISC) Model is recommended as a
prime candidate for RFI atmospheric dispersion modeling applications. Applicable
ISC source types include stack area and volume sources. Concentration estimates
can be based on times of as short as one hour and as long as one year. The model
can be used for both flat and rolling terrain. The ISC model can also account for
atmospheric deposition (i.e., inter-media transport to soil). The ISC Model (See EPA
12-54
-------
TABLE 12-8
PREFERRED MODELS FOR SELECTED APPLICATIONS IN SIMPLE TERRAIN
Short Term (1-24 hours)
Single Source
Multiple Source
Complicated Sources**
Buoyant Industrial Line Sources
Long Term (monthly, seasonal or annual)
Single Source
Multiple Source
Complicated Sources**
Buoyant Industrial Line Sources
Land Use
Rural
Urban
Rural
Urban
Rural/Urban
Rural
Rural
Urban
Rural
Urban
Rural/Urban
Rural
Model*
CRSTER
RAM
MPTER
RAM
ISC*
BLP
CRSTER
RAM
MPTER
COM 2.0 or RAM***
ISC* .
BLP
* The long-term version of ISC (i.e., ISCLT) is recommended as the preferred dispersion model for
RFI applications.
** Complicated sources are sources with special problems such as aerodynamic downwash,
particle deposition, volume and area sources, etc.
***lf only a few sources in an urban area are to be modeled, RAM should be used.
12-55
-------
450/4-86-005a and b) is included in the UNAMAP series available through the NTIS
(U.S. EPA, June 1986).
Additional guidance on dispersion model selection and application is available
from EPA Regional Office and State modeling representatives as well as from the
EPA Model Clearinghouse.
If other than preferred models are selected for use, early discussions with the
regulatory agency is encouraged. Agreement on the data base to be used,
modeling techniques to be applied and the overall technical approach, prior to the
actual analyses, helps avoid misunderstandings concerning the final results and may
reduce the later need for additional analyses. The preparation (and submittal to
the appropriate regulatory agency) of a written modeling protocol is recommended
for all RFI atmospheric dispersion modeling applications.
12.5.3 General Modeling Considerations
Dispersion modeling results are limited by the amount, quality and
representativeness of the input data. In addition to meteorological and source data
modeling input, the following are also important modeling factors:
• Location of facility property boundary
• Dispersion coefficients
• Sta b i I ity categories
• Plume rise
• Chemical transformation
• Gravitational settling and deposition
• Urban/rural classification
ai
In designing a computational network for modeling, the emphasis should be
placed on location with respect to the facility property boundary. The selection of
sites should be a case-by-case determination taking into consideration the
topography, the climatology, monitor sites, and should be based on the results of
the initial screening procedure. Additional locations may be needed in the high
concentration location if greater resolution is indicated by terrain or source factors.
12-56
-------
Gaussian models used in most applications should employ dispersion
coefficients consistent with those contained in the preferred models available in
UNAMAP. Factors such as averaging time, urban/rural surroundings, and type of
source (point vs. line) may dictate the selection of specific coefficients.
The Pasquill approach to classifying stability is generally required in all
preferred models. The Pasquill method, as modified by Turner, was developed for
use with commonly observed meteorological data from the National Weather
Service (NWS) and is based on cloud cover, insolation and wind speed.
Procedures to determine Pasquill stability categories from other than NWS
data are presented in Guidelines on Air Quality Models (Revised) (U.S. EPA, July
1986). Any other method to determine Pasquill stability categories should be
justified on a case-by-case basis.
t
The plume rise methods incorporated in the EPA preferred models arer
recommended for use in all modeling applications. No provisions,in these models*
are made for fumigation or multi-stack plume rise enhancement or the handling of
such special plumes as flares; these problems should be considered on a c se-by-case
basis.
Where aerodynamic downwash occurs due to the adverse influence of nearby
structures, the algorithms included in the ISC model should be used.
Use of models incorporating complex chemical mechanisms should be
considered only on a case-by-case basis with proper demonstration of applicability.
These are generally regional models not designed for the evaluation of individual
sources but used primarily for region-wide evaluations.
s_
An "infinite ftalf-life* should be used for estimates of total suspended
particulate concentrations when Gaussian models containing only exponential
decay terms for treating settling and deposition are used. Gravitational settling and
deposition may be directly included in a model if either is a significant factor. At
least one preferred model (ISC) contains settling and deposition algorithms and is
recommended for use when particulate matter sources can be quantified and
settling and deposition are problems.
12-57
-------
The selection of either rural or urban dispersion coefficients in a specific
application should follow one of the procedures presented in Guidelines on Air
Quality Models (Revised) (U.S. EPA, July 1986). These include a land use
classification procedure or a population based procedure to determine whether the
character of an area is primarily urban or rural.
12.6 Design of a Monitoring Program to Characterize Releases
Monitoring procedures should be developed based on the information
previously described, including determination of reasonable worst-case scenarios as
discussed above. This section discusses the recommended monitoring approaches.
Primary elements in designing a monitoring system include:
i
• Establishing monitoring objectives; *
I
• Determining monitoring constituents of concern;
• Monitoring schedule;
• Monitoring approach; and
• Monitoring locations.
*
Each of these elements should be addressed to meet the objectives of the
initial monitoring phase, and any subsequent monitoring that may be necessary.
These elements are described in detail below.
s_
12.6.1 Objectives of the Monitoring Program
The primary goal of the air investigation is to determine concentrations at the
facility property boundary as input to the health and environmental assessment
process. As discussed previously, the monitoring program may be conducted in a
phased approach, using the results of initial monitoring and/or modeling to
determine the need for and scope of subsequent monitoring.
12-58
-------
Principal components of both the initial and subsequent monitoring phases
are:
• Identification or verification of constituents;
• Characterization of long-term air constituent concentrations (based on a
"reasonable worst case" exposure period) at:
the unit boundary to maximize the potential for release detection
the facility property boundary
actual offsite receptor locations (for determining the need for
interim corrective measures)
areas upwind of the release source (to characterize background
concentrations); and
• Collection of meteorological data during the monitoring period to aid in
evaluating the air monitoring data.
Atmospheric dispersion modeling may also be used to estimate
concentrations, if monitoring is not practical, as discussed previously.
Subsequent monitoring may be necessary if initial monitoring and modeling
data were not sufficient to characterize long-term ambient constituent
concentrations.
12.6.2 Monitoring Constituents and Sampling Considerations
Sampling and analysis may be conducted for all appropriate Appendix VIII
constituents that have an air pathway potential (See Section 3 and Appendix B). An
alternative approach is to use unit and waste-specific information to identify
constituents that are not expected to be present and thus, reduce the list of target
monitoring constituents. For example, the industry specific monitoring constituent
lists presented in Appendix B, List 4 can be used to identify appropriate air
monitoring constituents for many applications (especially for units that serve only a
limited number of industrial categories). The target constituents selected should be
12-59
-------
limited to those which may be present in the waste and have health criteria for the
air pathway (see Section 8).
Results from screening assessment, emission monitoring, and/or screening
sampling phase (as defined later in Section 12.6.4.1) may also be used as a basis for
selection of monitoring constituents. These results may confirm/identify
appropriate monitoring constituents for the unit of concern.
12.6.3 Meteorological Monitoring
Monitoring of onsite meteorological conditions should be performed in
concert with other emission rate and air monitoring activities. Meteorological
monitoring results can serve as input for dispersion models, can be used to assure
that the air monitoring effort is conducted during the appropriate meteorological
conditions (e.g., "reasonable worst case" period for initial monitoring), and to aid
in the interpretation of air monitoring data.
12.6.3.1 Meteorological Monitoring Parameters
The following meteorological parameters should be routinely monitored
while collecting ambient air samples:
• Horizontal wind speed and direction;
• Ambient temperature;
• Atmospheric stability (e.g., based on the standard deviation of horizontal
wind direction or alternative standard methodologies);
SL
• Precipitation measurements if representative National Weather Service
data are not available; and
• Atmospheric pressure (e.g., for landfill sites or contaminated soils) if
representative National Weather Service data are not available.
12-60
-------
It is recommended that horizontal wind speed and direction, and air
temperature be determined onsite with continuous recording equipment.
Estimates from offsite monitors are not likely to be representative for all of the
conditions at the site. Input parameters for dispersion models, if appropriate,
should be reviewed prior to conducting the meteorological data collection phase to
ensure that all necessary parameters are included.
Field equipment used to collect meteorological data can range in
sophistication from small, portable, battery-operated units with wind speed and
direction sensors, to large, permanently mounted, multiple sensor units at varying
heights. Individual sensors can collect data on horizontal wind speed and direction,
three-dimensional wind speed, air temperature, humidity, dew point, and mixing
height. From such data, variables for dispersion models such as wind variability and
atmospheric stability can be determined. Additional guidance on meteorological
measurements can be obtained from:
U.S. EPA. June 1987. On-Site Meteorological Program Guidance for
Regulatory Modeling Applications. EPA-450/4-87-013. Office of Air Quality
Planning and Standards. Research Triangle Park, N.C. 27711.
U.S. EPA. February 1983. Quality Assurance handbook for- Air Pollution
Measurements Systems: Volume IV. Meteorological Measurements. EPA-
600/4-82-060. Office of Research and Development. Research Triangle Park,
N.C. 27711.
U.S. EPA. July 1986. Guidelines on Air Quality Models (Revised). EPA-405/2-78-
027R. NTIS PB 86-245248. Office of Air Quality Planning and Standards.
Research Triangle Park, N.C. 27711.
Appropriate performance specifications for monitoring equipment are given in the
following document:
U.S. EPA. November 1980. Ambient Monitoring Guidelines for Prevention of
Significant Deterioration (PSD). EPA-450/4-80/012. NTIS PB 81-153231. Office
of Air Quality Planning and Standards. Research Triangle Park, N.C. 27711.
12-61
-------
12.6.3.2 Meteorological Monitor Siting
Careful placement of meteorological monitoring equipment (e.g., sensors) is
important in gathering relevant data. The objective of monitoring tower
placement is to position sensors to obtain measurements representative of the
conditions that determine atmospheric dispersion in the area of interest. The
convention for placement of meteorological monitoring equipment is:
• At or above a height of 10 meters above ground; and
• At a horizontal distance of 10 times the obstruction height from any
upwind obstructions.
In addition, the recommendations given in Table 12-9 should be followed to avoid
effects of terrain on meteorological monitors.
Depending on the complexity of the terrain in the area of interest and the
parameters being measured, more than one tower location may be necessary.
Complex terrain can greatly influence the transport and diffusion of a contaminant
release to air so that one tower may not able to account for these influences. The
monitoring station height may also vary depending on source characteristics and
logistics. Heights should be selected to minimize near-ground effects that are not
representative of conditions in the atmospheric layer into which a constituent of
concern is being released.
A tower designed specifically to mount meteorological instruments should be
used. Instruments should be mounted on booms projecting horizontally out from
the tower at a minimum distance of twice the tower diameter. Sound engineering
practice should be used to assure tower integrity during all meteorologic
conditions.
Further guidance on siting meteorological instruments and stations is
available in the following publications:
12-62
-------
TABLE 12-9
RECOMMENDED SITING CRITERIA TO AVOID TERRAIN EFFECTS
Distance from Tower
(meters)
0 -15
15-30
30-100
100-300
Maximum Acceptable Construction
or Vegetation Height
(meters)
0.3
0.5-1.0
3
10
12-63
-------
U.S. EPA. November 1980. Ambient Monitoring Guidelines for Prevention of
Significant Deterioration (PSD). EPA-450/4-80-012. NTIS PB 81-153231. Office
of AirQuality Planning and Standards. Research Triangle Park, N.C. 27711.
U.S. EPA. June 1987. On-Site Meteorological Program Guidance for
Regulatory Modeling Applications. EPA-450/4-87-013. Office of Air Quality
Planning and Standards. Research Triangle Park, N.C. 27711.
U.S. EPA. February 1983. Quality Assurance Handbook for Air Pollution
Measurement Systems: Volume IV. Meteorological Measurements. EPA-
600/4-82-060. Office of Research and Development. Research Triangle Park,
N.C. 27711.
12.6.4 Monitoring Schedule
t
Establishment of a monitoring schedule is an important consideration in
developing a monitoring plan. When appropriate, air monitoring'should coincide
with monitoring of other media (e.g., subsurface gas, soils, and surface water) that
have the potential for ai. emissions. As with all other aspects of the monitoring
program, the objectives of monitoring should be considered in establishing a
schedule. As indicated previously, monitoring generally consists of screening
sampling, emission monitoring, and air monitoring. The monitoring schedule
during each of these phases is discussed below.
12.6.4.1 Screening Sampling
A limited screening sampling effort may be necessary to focus the design of
additional monitoring phases. Therefore, screening samples may be warranted
during the screening assessment or prior to initiating emission monitoring or air
monitoring studies. This screening phase can also be used to supplement modeling
and emission monitoring results as available, to verify the existence of a release to
air, and to prioritize the major release sources at the facility.
Screening sampling should be used to characterize air emissions (e.g., by using
total hydrocarbon measurements as an indicator), and to confirm/identify the
presence of candidate constituents. Screening samples should generally consist of
12-64
-------
source emissions measurements or ambient air samples collected, at or in close
proximity to the source. This approach will provide the best opportunity for
detection of air emission constituents. (A discussion of available screening methods
is presented in Section 12.8.) An alternative screening approach involves collection
of a limited number of air samples to facilitate the analysis of a wide range of
constituents (e.g., collection via Tenax adsorption tubes or whole air sampling with
analysis by GGMS - see Section 12.8).
The screening study should generally involve collection of a limited number of
grab or time-integrated samples (several minutes to 24 hours) for a limited time
period (e.g., one to five .days). Sampling should be conducted during
emission/dispersion conditions that are expected to result in relatively high
concentrations, as discussed previously. Screening results should be interpreted
considering the representativeness of the waste and unit operations during the
sampling, and the detection capabilities of the screening methodology used.
I
12.6.4.2 Emission Monitoring
Emission rate monitoring may be necessary to characterize a release if
screening assessment results are not conclusive. This approach involves stack or vent
emission monitoring for point sources. Point source monitoring is not dependent
on meteorological conditions. However, emission rate monitoring for both point
and area sources should be conducted during typical or "reasonable worst case"
emission rate conditions. Therefore, emission monitoring should be conducted
when source conditions (e.g., unit operations and waste concentrations) as well as
meteorological conditions are conducive to "reasonable worst case" emission rate
conditions. Emission rate monitoring for area sources should not be conducted
during or immediately following precipitation or if hourly average wind speeds are
s^
greater than 15 miles per hour. It should also be noted that soil or cover material (if
present) should be allowed to dry prior to continuing monitoring operations, as
volatilization decreases under saturated soil conditions. In these cases, the
monitoring should be interrupted and resumed as soon as possible after the
unfavorable conditions pass. Similarly, operational interruptions such as unit
shutdown should also be factored into the source sampling schedule.
12-65
-------
Point source emission sampling generally requires only a few hours of
sampling and occurs during a more limited time (e.g., one to three days). Guidance
on point-source sampling schedules is presented in the following:
U.S. EPA. November 1985. Practical Guide - Trial Burns for Hazardous Waste
Incinerators. NTIS PB 86-190246. Office of Research and Development.
Cincinnati, OH 45268.
U.S. EPA. Code of Federal Regulations. 40 CFR Part 60: Appendix A:
Reference Methods. Off ice of the Federal Register. Washington, D.C.
U.S. EPA. 1978. Stack Sampling Technical Information. A Collection of
Monographs and Papers. Volumes Nil. EPA-450/2-78-042a,b,c. NTIS PB 80-
161672, 80-161680, 80-161698. Office of Air Quality Planning and Standards.
Research Triangle Park, NC 27711.
U.S. EPA. February 1985. Modified Method 5 Train and'Source Assessment
Sampling System Operators Manual. EPA-600/8-85-003. NTIS PB 85-169878.
Off ice of Research and Development. Research Triangle Park, NC 27711.
U.S. EPA. March 1984. Protocol for the Collection and Analysis of Volatile
POHCs Using VOST. EPA-600/8-84-007. NTIS PB 84-170042. Office of Research
and Development. Research Triangle Park, NC 27711.
U.S. EPA. February 1984. Sampling and Analysis Methods for Hazardous
Waste Combustion. EPA-600/8-84-002. NTIS PB 84-155845. Washington, D.C.
20460.
U.S. EPA. 198\ Source Sampling and Analysis of Gaseous Pollutants. EPA-
APTI Course Manual 468. Air Pollution Control Institute. Research Triangle
Park, NC 27711.
U.S. EPA. 1979. Source Sampling for Paniculate Pollutants. EPA-APTI Course
Manual 450. NTIS PB 80-188840, 80-182439, 80-174360. Air Pollution Control
Institute. Research Triangle Park, NC 27711.
12-66
-------
U.S. EPA. 1986. Test Methods for Evaluating Solid Waste. 3rd Edition. Office
of Solid Waste. EPA/SW-846. GPO No. 955-001-00000-1. Washington, D.C
20460.
Emission rate monitoring should be conducted during a 1 to 3 day period
representative of "reasonable worst case" source emission conditions. The worst
case short-term emission rate conditions should be determined by parametric
analyses (i.e., by modeling a wide range of source operational conditions and
associated waste concentrations as well as meteorological conditions for
parameters such as wind speed and temperature). Historical meteorological data
representative of the site should be reviewed to determine the season and time of
day associated with worst case emission conditions. These results should be used to
select and schedule (along with meteorological forecasts for local conditions and
expected source operational and waste concentration) the emission monitoring
period. t
I
Emission rate monitoring results based on measurements during worst-case
conditions should be initially used as dispersion modeling input. If these initial
results exceed health criteria then the emission monitoring results should be scaled
to represent long term (i.e., annual) conditions. The scaling factor should be based
on the ratio of emission rate modeling results (using meteorological conditions
during the monitoring period as input) compared to modeling results based on
typical (annual) meteorological conditions.
Guidance on area source emission rate monitoring is provided in the
following:
U.S. EPA. 19^6. Measurement of Gaseous Emission Rates from Land Surfaces
Using an Emission Isolation Flux Chamber: User's Guide. EPA/600/8-86/008.
NTISPB86-223161. Environmental Monitoring Systems Laboratory. Las Vegas,
NV 89114.
U.S. EPA. December 1988 Draft. Procedures for Conducting Air Pathway
Analyses for Superfund Applications. Office of Air Quality Planning and
Standards. Research Triangle Park, NC 27711.
12-67
-------
12.6.4.3 Air Monitoring
The primary objective of confirmatory monitoring is to characterize long-term
exposures that may be associated with air emissions from the unit under reasonable
worst-case conditions. A schedule should be proposed that will provide an
adequate degree of confidence that those compounds that may be released will be
detected (i.e., by sampling during the season associated with the highest air
concentrations as determined based on modeling). Laboratory analytical costs
typically range from $200 to over $1,000 per air monitoring station for one 24-hour
integrated sample (the actual cost depends on the number and type of target
constituents). Recent advances in applied technology have facilitated the use of
field gas chromatographs (GCs) to automatically obtain analytical results for many
organics (i.e., offsite laboratory analyses may not be necessary for some air
monitoring programs). The cost for this equipment typically range from $20,000 to
over $50,000 and one GC can generally service multiple sampling stations.
An example sampling schedule (e.g., for flat terrain sites with minimal
variability of dispersion and source conditions) for meeting this objective is given
below:
• Meteorological monitoring - 90 days continuous monitoring.
• Initial air monitoring (Alternative 1) -90 days:
Analysis of 24-hour time integrated samples for target constituents
every day during the 90-day period (total of 90 samples)
• Additional monitoring - as necessary to supplement initial air monitoring
results in order to adequately characterize the release.
Si
The 90-day monitoring program will facilitate collecting samples over a wide
range of emission and dispersion conditions. The 90-day period should be selected,
as previously discussed, to coincide with the expected season of highest ambient
concentrations. Meteorological monitoring should be continuous and concurrent
with this 90-day period to adequately characterize dispersion conditions at the site
and to provide meteorological data to support interpretation of the air-quality
monitoring data.
12-68
-------
The collection of a time-integrated sample based on continuous monitoring
for several days can result in technical difficulties (e.g., poor collection efficiencies
for volatile constituents or large sample volumes). The application of five-day
composite samples at each station, or intermittent sampling during the five days,
results in continuous monitoring coverage during the 90-day period and facilitates
the characterization of long-term exposure levels.
Although there are some limitations associated with composite/intermittent
sampling (e.g., the potential for sample degradation), the 24-hour samples
collected every sixth day will provide a second data set for characterizing ambient
concentrations. Although the results of the two data sets should not be directly
combined (because of the different sampling periods) they provide a
comprehensive technical basis by which to evaluate long-term exposure conditions.
12.6.4.4 Subsequent Monitoring
Subsequent monitoring may be necessary if initial monitoring data were not
sufficient to estimate "reasonable worst case" long-term concentrations (e.g., data
recovery was not sufficient or additional monitoring stations are needed).
The same schedule specified for the initial monitoring phase is also applicable
to subsequent monitoring. However, when evaluating the results of subsequent
monitoring and comparing them to previously collected data, potential differences
in emission/dispersion conditions and other data representativeness factors should
be accounted for.
12.6.5 Monitoring Approach
EL
The RFI air release characterization strategy may involve source emission
monitoring and/or air monitoring. The strategy which defines the process for
selection and application of these alternative monitoring approaches has been
discussed previously. A summary of applicable air monitoring strategies related to
source type is presented in Table 12-10.
12-69
-------
TABLE 12-10
APPLICABLE AIR RELEASE SAMPLING STRATEGIES BY SOURCE TYPE
Unit Type/Expected Emission
Air Release Sampling Strategy
Air
Monitoring
Emissions Monitoring
Vent/Stack
Sampling
Isolation
Flux
Chambers
AREA SOURCES WITH LIQUID SURFACES
Surface Impoundments
Vapor Phase
Particulates
Open Roof Storage/Treatment
Tanks
Vapor Phase
X
X
AREA SOURCES WITH SOLID SURFACES
Waste Piles
Vapor Phase
Particulates
Landfill Surface
Vapor Phase
Particulates
Land Treatment
Vapor Phase
Particulates
X
X
X
X
X
X
POINT SOURCES
Vents from container Handling
Units
Vapor Phase
Landfill Vents
Vapor Phase
Storage/Treatment Tank Vents
Vapor Phase
Incinerators ,
Vapor Phase
Particulates
X
X
X
X
X
X
X
X
X
X
12-70
-------
12.6.5.1 Source Emissions Monitoring
Monitoring at the source to measure a rate of emission for the constituents of
concern may, in many cases, offer a practical approach to characterizing air
emissions. Using this technique, the emission rate is then input into a mathematical
dispersion model for estimation of downwind concentrations. Monitoring
interferences from sources close to the unit are eliminated because the source is
isolated from the ambient atmosphere for monitoring purposes. Source monitoring
techniques are also advantageous because they do not require the level of
sensitivity required by air monitors. Concentrations of airborne constituents at the
source are generally higher than at downwind locations due to the lack of
dispersion of the constituent over a wide area. The concentrations expected in the
air (generally part-per-billion levels) may be at or near the limit of detectability of
the methods used. Methods for source emissions monitoring for various constituent
classes are discussed in Section 12.8. t
!
Area sources (such as landfills, land treatment units, and surface
impoundments) can be monitored using the isolation flux chamber approach. This
method involves isolating a small area of contamination under a flux chamber, and
passing a known amount of a zero hydrocarbon carrier gas through the chamber,
thereby picking up any organic emissions in the effluent gas stream from the flux
chamber. Samples of this effluent stream are collected in inert sampling containers,
usually stainless steel canisters under vacuum, and removed to the laboratory for
subsequent analysis. The analytical results of the identified analytes can be
converted through a series of calculations to direct emission rates from the source.
These emission rates can be used to evaluate downwind concentrations by
application of dispersion models. Multiple emission tests should be conducted to
account for temporal and spatial variability of source conditions. More information
on use of the isolation flux chamber and test design is provided in the following
references:
U.S. EPA. 1986. Measurement of Gaseous Emission Rates from Land Surfaces
Using an Emission Isolation Flux Chamber: User's Guide. EPA/600/8-86/008.
NTISPB 86-223161. Washington, D.C. 20460.
12-71
-------
U.S. EPA. December 1988 Draft. Procedures for Conducting Air Pathway
Analyses for Suoerfund Applications. Office of Air Quality Planning and
Standards. Research Triangle Park, NC 27711.
Some area source units may not be amenable to the source sampling
approach, however. A unit in which the source cannot be isolated and viable
measurements taken of the parameters of concern is one example. This includes
active areas of landfills and land-treatment areas, as well as aerated surface
impoundments. Also, area sources in which particulate emissions are of concern
cannot be measured using an isolation flux chamber due to technical limitations in
the technique. For these applications, only an upwind/downwind monitoring
approach should be used.
12.6.5.2 Air Monitoring
*
t
Use of an upwind/downwind network of monitors or sample collection devices
is the primary air monitoring approach recommended to determine release and
background concentrations of the constituents of concern. Upwind/downwind air
monitoring networks provide concentrations of the constituents of concern at the
point of monitoring, whether at the unit boundary, facility property boundary, or at
a receptor point. The upwind/downwind approach involves the placement of
monitors or sample collection devices at various points around the unit of concern.
Each air sample collected is classified as upwind or downwind based on the wind
conditions for the sampling period. Downwind concentrations are compared to
those measured at upwind points to determine the relative contribution of the unit
to air concentrations of toxic compounds. This is generally accomplished by
subtracting the upwind concentration (which represents background conditions)
from the concurrent downwind concentrations. Applicable field methods for air
t
monitoring are discussed in Section 12.8 as well as in Procedures for Conducting Air
Pathway Analyses for Superfund Applications (U.S. EPA, December 1988).
Downwind air concentrations at the facility can be extrapolated to other locations
by using dispersion modeling results. This is accomplished by obtaining initial
modeling results based on meteorological conditions for the monitoring period and
an arbitrary emission rate. These initial dispersion modeling results along with
monitoring results at the site perimeter are used to back calculate an emission rate
such that modeling results can be adjusted to be equivalent to monitoring results at
12-72
-------
the onsite monitoring station. This estimated emission rate is then used as
dispersion modeling input to predict offsite concentrations.
12.6.6 Monitoring Locations
As with other factors associated with air monitoring, siting of the monitors
should reflect the primary objective of characterizing concentrations at the facility
property boundary. This section discusses monitoring locations for both
upwind/downwind approaches and source monitoring techniques.
12.6.6.1 Upwind/Downwind Monitoring Locations
The air monitoring network design should provide adequate coverage to
characterize both upwind (background) and downwind concentrations. Therefore,
four air monitoring zones are generally necessary for initial monitoring. Multiple^
monitoring stations per zone will frequently be required to adequately characterize^
the release. An upwind zone is used to define background concentration levels.
Downwind zones at the unit boundary, at the facility property boundary and
beyond the facility property boundary, if appropriate, are used to define potential
offsite exposure.
The location of air monitoring stations should be based on local wind patterns.
Air monitoring stations should be placed at strategic locations, as illustrated in the
following example (see Figure 12-6).
• Upwind (based on the expected prevailing wind flow during the 90-day
monitorjng period) of the unit and near the facility property boundary to
characterize background air concentration levels. There should be no air
emission_source between the upwind monitoring station and the unit
boundary.
• Downwind (based on the expected prevailing wind flow during the 90-
day monitoring period) at the unit boundary plus stations at adjacent
sectors also at the unit boundary (the separation distance of air
monitoring stations at the unit boundary should be 30° or 50 feet,
whichever is greater).
12-73
-------
_C
>^ X "C
• ">
Q. U a» *"
W Q. 0)
\
1
2
s
o
2
O
QC
a.
i
Ul
ce
D
\
\
\
X -^
'x
0
4, cnu
ac c
x O O
ui-j VJ
X
\
12-74
-------
• Downwind (based on the expected prevailing wind flow during the 90-
day monitoring period) at the facility property boundary (this station
may not be required if the site perimeter is within 100 meters of the unit
boundary).
• Downwind (at the area expected to have the highest average
concentration levels during the 90-day monitoring period) at the facility
property boundary, if appropriate.
• Downwind at actual offsite receptor locations (if appropriate).
• Additional locations at complex terrain and coastal sites associated with
pronounced secondary air flow paths (e.g., downwind of the unit near
the facility property boundary for both primary daytime and nighttime
flow paths). .
The above locations should be selected prior to initial monitoring based on the
onsite meteorological survey and on evaluation of available representative offsite
meteorological data. This analysis should provide an estimate of expected wind
conditions during the 90-day initial monitoring period. If sufficient representative
data are available, dispersion modeling can be used to identify the area of
maximum long term concentration levels at the facility property boundary and, if
appropriate, at actual offsite receptors. If not, the facility property boundary sector
nearest to the unit of concern should be selected for initial monitoring.
The network design defined above will provide an adequate basis to define
long-term concentrations based on continuous monitoring during the 90-day initial
monitoring period.. The monitoring stations at the unit boundary should increase
the potential for release detection. The facility property boundary air monitoring
stations should provide data (with the aid of dispersion modeling, if appropriate) to
perform health and environmental assessment, and if appropriate, characterize
offsite concentrations.
Air monitoring at offsite receptors (if deemed to be appropriate) may be
impractical in many cases, because analytical detection limits may not be low
12-75
-------
enought at offsite receptor locations to measure the release. Also, a 90-day offsite
monitoring program can be problematic. Factors such as vandalism, erroneous
readings due to public tampering with the equipment, public relations problems in
setting up the equipment, and legal access problems may preclude the use of offsite
air monitoring stations. For these cases, dispersion models may be used to
extrapolate monitoring data collected at the facility to actual offsite receptor
locations. This is accomplished by obtaining initial modeling results based on
meteorological conditions for the monitoring period and an arbitrary emission rate.
These initial dispersion modeling results along with monitoring results at the site
perimeter are used to back calculate an emission rate such that modeling results can
be adjusted to be equivalent to monitoring results at the onsite monitoring station.
This estimated emission rate is then used as dispersion modeling input to predict
offsite concentrations for the same downwind sector and exposure period as for this
monitoring period.
If additional monitoring is required, a similar network design to that
illustrated in Figure 12-6 will generally be appropriate. Evaluation of the
meteorological monitoring data collected during the initial phase should provide an
improved basis to identify local prevailing and diurnal wind flow paths. Also, the
site meteorological data will provide dispersion modeling input. These modeling
results should provide dilution patterns that can be used to identify areas with
expected relatively high concentration levels. However, these results should
account for seasonal meteorological differences between initial and additional
monitoring periods.
Wind-directionally controlled air monitoring stations can also be used at sites
with highly variable wind directions. These wind-directionally controlled stations
should be collocated with the fixed monitoring stations. This approach facilitates
determination of the unit source contribution to total constituent levels in the local
area. These automated stations will only sample for a user-defined range of wind
directions (e.g., downwind stations would only sample if winds were blowing from
the source towards the station). Interpretation of results from wind-directionally
controlled air monitoring stations should account for the lower sampling volumes
(and therefore, the possibility that not enough sample would be collected for
analysis) generally associated with this approach.
12-76
-------
The inlet exposure height of the air monitors should be 2 to 15 meters to be
representative .of potential inhalation exposure but not unduly biased by road dust
and natural wind erosion phenomena. Further guidance on air monitoring network
design and station exposure criteria (e.g., sampling height and proximity to
structures and air emission sources) is provided in the following reference:
U.S. EPA. September 1984. Network Design and Site Exposure Criteria for
Selected Non-criteria Air Pollutants. EPA-450/4-84-022. Office of Air Quality
Planning and Standards. Research Triangle Park, N.C.
The above referenced document recommends the use of dispersion models to
identify potential relatively high concentration areas as a basis for network design.
This topic is also discussed in the following document:
U.S. EPA. July 1986. Guidelines on Air Quality Models (Revised). EPA-450/2-78-
027R. NTIS PB 86-245248. Office of Air Quality Planning and Standards.
Research Triangle Park, NC 27711.
Uniformity among the sampling sites should be achieved to the greatest
degree possible. Descriptions should be prepared for all sampling sites. The
description should include the type of ground surface, and the direction, distance,
and approximate height with respect to the source of the release. Location should
also be described on a facility map.
12.6.6.2 Stack/Vent Emission Monitoring
Point source measurements should be taken in the vent. Both the VOST and
Modified Method 5 methodologies describe the exact placement in the stack for the
sampler inlet. (See Section 12.8.3). If warranted, an upwind/downwind monitoring
network can be used to supplement the release rate data.
12.6.6.3 Isolation Flux Chambers
Monitor placement using flux chambers (discussed earlier) is similar to
conducting a characterization of any area source. Section 3 of this guidance
discusses establishment of a grid network for sampling. Such a grid should be
12-77
-------
established for an area source, with sampling points established within the grids, as
appropriate. It is suggested that a minimum of six points be chosen for each
monitoring effort. Once these areas are sampled, the results can be temporally and
spatially averaged to provide an overall compound specific emission rate for the
plot. Additional guidance on monitoring locations for isolation flux chambers is
presented in Section 3.6 and in the following references:
U.S. EPA. 1986. Measurement of Gaseous Emission Rates from Land Surfaces
Using an Emission Isolation Flux Chamber: User's Guide. EPA/600/8-86/008.
NTIS PB86-223161. Environmental Monitoring Systems Laboratory. Las Vegas,
NV89114.
U.S. EPA. December 1988 Draft. Procedures for Conducting Air Pathway
Analyses for Superfund Applications. Office of Air Quality Planning and
Standards. Research Triangle Park, NC 27711.
12.7 Data Presentation
As discussed in Section 5, progress reports will be required by the regulatory
agency at periodic intervals during the investigation. The following data
presentation formats are suggested for the various phases of the air investigation in
order to adequately characterize concentrations at actual offsite receptors.
12.7.1 Waste and Unit Characterization
Waste and unit characteristics should be presented as:
• Tables of waste constituents and concentrations;
• Tables of relevant physical/chemical properties for potential air emission
constituents;
• Tables and narratives describing unit dimensions and special operating
conditions and operating schedules concurrent with the air monitoring
program;
12-78
-------
• Narrative description of unit operations; and
• Identification of "reasonable worst case" emission conditions that
occurred during the monitoring period.
12.7.2 Environmental Setting Characterization
Environmental characteristics should be presented as follows:
• Climate (historical summaries from available onsite and offsite sources):
Annual and monthly or seasonal wind roses;
Annual and monthly or seasonal tabular summaries of mean wind
speeds and atmospheric stability distributions; and
Annual and monthly or seasonal tabular summaries of temperature
and precipitation.
• Meteorological survey results:
Hourly listing of all meteorological parameters for the entire
monitoring period;
Daytime wind rose (at coastal or complex terrain sites);
Nighttime wind rose (at coastal or complex terrain sites);
Summary wind rose for all hours;
Summary of dispersion conditions for the monitoring period (joint
frequency distributions of wind direction versus wind speed
category and stability class frequencies); and
Tabular summaries of means and extremes for temperature and
other meteorological parameters.
12-79
-------
• Definition of soil conditions (if appropriate):
Narrative of soil characteristics (e.g., temperature, porosity and
organic matter content); and
Characterization of soil contamination conditions (e.g., in land
treatment units, etc.).
• Definition of site-specific terrain and nearby receptors:
Topographic map of the site area with identification of the units,
meteorological and air monitoring stations, and facility property
boundary;
Topographic map of 10-kilometer radius from site (U.S. Geological
Survey 7.5 minute quadrangle sheets are acceptable); and
Maps which indicate location of nearest residenc: for each of
sixteen 22.5 degree sectors which correspond to major compass
points (e.g., north, north-northwest, etc.), nearest population
centers and sensitive receptors (e.g., schools, hospitals and nursing
homes).
• Maps showing the topography of the area, location of the unit(s) of
concern, and the location of meteorological monitoring equipment.
• A narrative description of the meteorological conditions during the air
sampling periods, including qualitative descriptions of weather events
and precipitation which are needed for data interpretation.
12.7.3 Characterization of the Release
Characteristics of the release should be presented as follows:
• Screening sampling:
12-80
-------
Identification of sampling and analytical methodology;
Map which identifies sampling locations;
Listing of measured concentrations indicating collection time
period and locations;
Prioritization of units as air release sources which warrant
monitoring based on screening results;
Discussion of QA/QC results; and
Listing and discussion of meteorological data during the sampling
period.
• Initial and additional monitoring results:
Identification of monitoring constituents;
Discussion of sampling and analytical methodology as well as
equipment and specifications;
Identification of monitoring zones as defined in Section 12.6.6.1;
*
Map which identifies monitoring locations relative to units;
Discussion of QA/QC results;
Listing of concentrations measured by station and monitoring
period indicating concentrations of all constituents for which
monitoring was conducted. Listings should indicate detection limits
if a constituent is not detected;
Summary tables of concentration measured indicating maximum
and mean concentration values for each monitoring station;
12-81
-------
Discussion of meteorological station locations selection, sensor
height, local terrain, nearby obstructions and equipment
specifications;
Listing of all meteorological parameters concurrent with the air
sampling periods;
Daytime wind rose (only for coastal or complex terrain areas);
Nighttime wind rose (only for coastal or complex terrain areas);
Summary wind rose based on all wind direction observations for the
sampling period;
Summary of dispersion conditions for the sampling period (joint
frequency distributions of wind direction versus wind speed
category and stability class frequencies based on guidance
presented in Guidelines on Air Quality Models (Revised). (U.S. EPA,
July 1986));
Tabular summaries of means and extremes for temperature and
other meteorological parameters;
A narrative discussion of sampling results, indicating problems
encountered, relationship of the sampling activity to unit operating
conditions and meteorological conditions, sampling periods and
times, background levels and identification of other air emission
sources and interferences which may complicate data
interpretation;
Presentation and discussion of models used (if any), modeling input
data and modeling output data (e.g., dilution or dispersion patterns
based on modeling results); and
12-82
-------
Concentrations based on monitoring and/or modeling for actual
offsite receptor locations.
Interpretation of air monitoring results should also account for additional
factors such as complex terrain, variable winds, multiple contaminant sources and
intermittent or irregular releases. The key to data interpretation for these cases is
to evaluate monitoring results as a function of wind direction.
Terrain factors can alter wind flow trajectories especially during stable
nighttime conditions. Therefore, straightline wind trajectories may not occur
during these conditions if there is intervening terrain between the source and the
air monitoring station. For these cases wind flows will be directed around large
obstacles (such as hills) or channeled (for flows within valleys). Therefore, it is
necessary to determine the representativeness of the data from the meteorological
stations as a function of wind direction, wind speed and stability conditions. Based
on this assessment, and results from the meteorological survey, upwind and
downwind sectors (i.e., a range of wind direction as measured at the meteorological
station) should be defined for each air monitoring station to aid in data
interpretation. Figure 12-7 illustrates an example which classifies a range of wind
directions during which the air monitoring stations will be downwind of an air
emission source. Therefore, concentrations measured during upwind conditions can
be used to characterize background conditions and concentrations measured
during downwind conditions can be used to evaluate the air-quality impact of the
release.
Complex terrain sites and coastal sites frequently have very pronounced
diurnal wind patterns. Therefore, as previously discussed, the air monitoring
network at these sites may involve coverage for multiple wind direction sectors and
use of wind-directionally controlled air samplers. This monitoring approach is also
appropriate for sites with highly variable wind conditions. Comparing results from
two collocated air monitoring stations (i.e., one station which samples continuously
and a second station at the same location which is wind-directionally controlled on
an automated basis), facilitates determination of source contributions to ambient
air concentrations.
12-83
-------
FIGURE 12-7
EXAMPLE OF DOWNWIND EXPOSURES AT AIR MONITORING STATIONS
MONITORING STATIONS
DOWNWIND SECTOR
UNIT SOURCE
12-84
-------
Comparison of results from collocated (continuous versus wind-directionally
controlled) air monitoring stations can also be used to assist in data interpretation
at sites with multiple air emission sources or with intermittent/irregular releases.
For some situations, the consistent appearance of certain air emission constituents
can be used to "fingerprint" the source. Therefore, the air monitoring results can
be classified based on these "fingerprint" patterns. These results can then be
summarized as two separate data sets to assess background versus source
contributions to ambient concentrations.
The use of collocated (continuous and wind-directionally controlled) air
monitoring stations is a preferred approach to data interpretation for complex
terrain, variable wind, multiple source and intermittent release sites. An alternative
data interpretation approach involves reviewing the hourly meteorological data for
each air sampling period. Based on this review, the results from each sampling
period (generally a 24-hour period) for each station are classified in terms of
downwind frequency. The downwind frequency is defined as the number of hours
winds were blowing from the source towards the air monitoring station divided by
the total number of hours in the sampling period. These data can then be processed
(by plotting scattergrams) to determine the relationship of downwind frequency to
measured concentrations.
Data interpretation should also take into account the potential for deposition,
degradation and transformation of the monitoring constituents. These mechanisms
can affect ambient concentrations as well as air sample chemistry (during storage).
Therefore, standard technical references on chemical.properties, as well as the
monitoring guidance previously cited, should be consulted to determine the
importance of degradation and transformation for the monitoring constituents of
concern.
12.8 Field Methods
This section describes field methods which can be used during initial or
subsequent monitoring phases. Methods are classified according to source type and
area. Guidance on meteorological monitoring methods is also provided in this
section.
12-85
-------
12.8.1 Meteorological Monitoring
Meteorological monitoring generally should employ a 10-meter tower
equipped with wind direction, wind speed, temperature and atmospheric stability
instrumentation. Wind direction and wind speed monitors should exhibit a starting
threshold of less than 0.5 meters per second (m/s). Wind speed monitors should be
accurate above the starting threshold to within 0.25 m/s at speeds less than or equal
to 5 m/s. At higher speeds the error should not exceed 5 percent of the wind speed.
Wind direction monitor errors should not exceed 5 degrees. Errors in temperature
should not exceed 0.5°Cduring normal operating conditions.
The meteorological station should be installed at a location which is
representative of overall site terrain and wind conditions. Multiple meteorological
station locations may be required at coastal and complex terrain sites.
Additional guidance on equipment performance specifications, station
location, sensor exposure criteria, and field methods for meteorological monitoring
are provided in the following references:
U.S. EPA. February 1983. Quality Assurance Handbook for Air Pollution
Measurement Systems: Volume IV. Meteorological Measurement. EPA-600-4-
82-060. Office of Research and Development. Research Triangle Park, NC
27711.
U.S. EPA. November 1980. Ambient Monitoring Guidelines for Prevention of
Significant Deterioration (PSD). EPA-450/4-80-012. NTIS PB 81-153231. Office
of Air Quality Planning and Standards. Research Triangle Park, NC 27711.
U.S. EPA. July 1986. Guidelines on Air Quality Models (Revised). EP-450/2-78-
027R. NTIS PB 86-245248. Office of Air Quality Planning and Standards.
Research Triangle Park, NC 27711.
12.8.2 Air Monitoring
Selection of methods for monitoring air contaminants should consider a
number of factors, including the compounds to be detected, the purpose of the
12-86
-------
method (e.g., screening or quantification), the detection limits, and sampling rates
and duration required for the investigation.
Organic and inorganic constituents require different analytical methods.
Within these two groups, different methods may also be required depending on the
constituent and its physical/chemical properties. Another condition that affects the
choice of monitoring technique is whetherthe compound is primarily in the gaseous
phase or is found adsorbed to solid particles or aerosols.
Screening for the presence of air constituents involves techniques and
equipment that are rapid, portable, and can provide "real-time" monitoring data.
Air contamination screening will generally be used to confirm the presence of a
release, or to establish the extent of contamination during the screening phase of
the investigation. Quantification of individual components is not as important
during screening as during initial and additional air monitoring, however the
technique must have sufficient specificity to differentiate hazardous constituents of
concern from potential interferences, even when the latter are present in higher
concentrations. Detection limits for screening devices are often higher than for
quantitative methods.
Laboratory analytical techniques must provide positive identification of the
components, and accurate and precise measurement of concentrations. This
generally means that preconcentration and/or storage of air samples will be
required. Therefore, methods chosen for quantification usually involve a longer
analytical time-period, more sophisticated equipment, and more rigorous quality
assurance procedures.
The following list of references provides guidance on air monitoring
methodologies:
U.S. EPA. June 1983. Technical Assistance Document for Sampling and
Analysis of Toxic Organic Compounds in Ambient Air. EPA-600/4-83-027. NTIS
PB 83-239020. Office of Research and Development. Research Triangle Park,
NC 27711.
12-87
-------
U.S. EPA. April 1984. Compendium of Methods for the Determination of Toxic
Organic Compounds in Ambient Air. EPA-600/4-84-041. Office of Research
and Development. Research Triangle Park, NC 27711.
NIOSH. February 1984. NIOSH Manual of Analytical Methods. NTIS PB 85-
179018. National Institute for Occupational Safety and Health. Cincinnati, OH.
U.S. EPA. September 1983. Characterizaiton of Hazardous Waste Sites - A
Methods Manual: Volume II. Available Sampling Methods. EPA-60G74-83-040.
NTIS PB 84-126929. Office of Solid Waste. Washington, D.C. 20460.
U.S. EPA. September 1983. Characterization of Hazardous Waste Sites - A
Methods Manual: Volume III, Available Laboratory Analytical Methods. EPA-
600/4-83-040. NTIS PB 84-126929. Office of Solid Waste. Washington, D.C.
20460.
U.S. EPA. 1986. Test Methods for Evaluating Solid Waste. 3rd Edition. EPA
SW-846. GPO No 955-001-00000-1. Office of Solid Waste. Washington, D.C.
20460.
ASTM. 1982. Toxic Materials in the Atmosphere. ASTM, STP 786.
Philadelphia, PA.
ASTM. 1980. Sampling and Analysis of Toxic Orqanics in the Atmosphere.
ASTM, STP 721. Philadelphia, PA.
ASTM. 1974. Instrumentation for Monitoring Air Quality. ASTM, STP 555.
Philadelphia, PA.
APHA. 1977. Methods of Air Sampling and Analysis. American Public Health
Association. Cincinnati, OH.
ACGIH. 1983. Air Sampling Instruments for Evaluation of Atmospheric
Contaminants. American Conference of Governmental Industrial Hygienists.
Washington, D.C.
12-i
-------
U.S. EPA. December 1988 Draft. Procedures for Conducting Air Pathway
Analyses for Superfund Applications. Office of Air Quality Planning and
Standards. Research Triangle Park, NC 27711.
12.8.2.1 Screening Methods
Screening techniques for vapor-phase constituents fall into two main
categories. (1) organic and non-organic compound-specific indicators, and (2)
general organic detectors. Table 12-11 presents a summary of commercially
available screening methods for these compounds.
Indicator tubes and other colorimetric methods-Indicator tubes, also known
as gas detector or Draeger tubes, are small glass tubes filled with a reagent-coated
material which changes color when exposed to a particular chemical. Air is pulled
through the tube with a low-volume pump. Tubes are available for 40 organic
gases, and for 8 hour or 15 minute exposure periods. Indicator tubes were designed
for use in occupational settings, where high levels of relatively pure gases are likely
to occur. Therefore, they have only limited usefulness for ambient air sampling,
where part-per-billion levels are often of concern. However, because they are
covenient to use and available for a wide range of compounds, detector tubes may
be useful in some screening/sampling situations.
Other colorimetric methods, such as continuous flow and tape monitor
techniques, were developed to provide real-time monitoring capability with
indicator methods. The disadvantages of these systems are similar to those of
indicator tubes.
Instrument detection screening methods-More commonly used for volatile
organic surveys, portable instrument detection methods include flame ionization
detectors (FID), photoionization detectors (PID), electron capture detectors (BCD),
and infrared detectors (ID). Also in use are detectors that respond to specific
chemical classes such as sulfur- and nitrogen-containing organics. These
instruments are used to indicate levels of total organic vapors and for identification
of " hot zones" downwind of the release source(s). They can be used as real-time
non-specific monitors or, by adding a gas chromatograph, can provide
concentration estimates and tentative identification of pollutants.
12-89
-------
00
Of
O
oe
LL.
DC
O
OS
O
LL.
VA
LU
D
HI
K
U
vn
LU
I
>
u
OS
O
u
_j
u
Comments
sl
J'i
x O
t§
3- a
< S
(•j
Compounds Detected
h.
01
W
W
10
*^
3
IQ
4*
cr
'c
£.
LV
Ol
0
+4
C
01
•a
01
Q.
01
"O
>s
cn
(If rtJ
Sensitivity and sel
component of inti
i
Q.
a.
o
*—
o
Various organics and
inorganics
C ig
I'll
o S *
ection Tubes
*^
Ol
O
0{
Ol «/i
"S s
-c 5
•= ,«
5 ""^
^^ 01
ies(H20,CO
lotential inl
S =•
Some inorganic g<
and therefore are
a
a
o
~
*-
Most organics
a)
==
5
1
X
LL.
15
c
1
(0
c
•8
c
3
o
o
u
'c
.
Q C
LU <
C
o
^
(^
E
J
"c
'~
>
'••.
3
'^
fl
cr
vt
m
Qualitative as wel
obtained.
1
a
in
d
Same as above except
that polar compounds
may not elute from the
column.
w
3
V
u
]i
LL. <
portable)
Q
U.
D
a
a.
E
T3
^_i
11
9 S
W Q;
«. ^
^ ^^
>--o
jQ rtj
-o 2
2 i
^ s
.2.2
•o t
S
.^
Response varies w
compound.
>
a
a
§
o
•~
d
Halogenated and nitro-
substituted compounds
vj
9"
<
(portable)
u
LU
D
o
3
5
-C
Q.
Q.
O
^
^«_
organic sul
detected.
~O Ol
Both inorganic an
compounds will b
.0
a.
a.
o
o
o
Sulfur or phosphorus-
containing compounds
Q
<
(portable)
Q
Q.
LU
O
-------
Of the available detectors, those that are the most applicable to an RFI are the
FID and PID. Table 12-12 summarizes four instruments (two FID and two PID
versions) which are adequate for the purposes of the screening phase.
Flame lonization Detectors-The Century OVA 100 series and AID Model 550
utilize a FID to determine the presence of vapor phase organics. The detector
responds to the total of all organics present in the air at any given moment. Flame
ionization detectors will respond to most organics, but are most sensitive to
hydrocarbons (i.e., those chemicals which contain only carbon and hydrogen
molecules such as benzene and propane). FIDs are somewhat less sensitive to
compounds containing chlorine, nitrogen, oxygen, and sulfur molecules. The
response is calibrated against a reference gas, usually methane. FID response is
often termed "total hydrocarbons"; however, this is misleading because particulate
hydrocarbons are not detected. FID detection without gas chromatography is not
useful for quantification of individual compounds, but provides a useful tool for
general assessment purposes. Detection limits using a FID detector alone are about
1 ppm. Addition of a gas chromatograph (GC) lowers the detection limit to ppb
levels, but increases the analysis time significantly.
Photoionization Detectors-Portable photoionization detectors such as the
HNU Model PI-101 and the Photovac 10A10 operate by applying UV ionizing
radiation to the contaminant molecules. Some selectivity over the types of organic
compounds detected can be obtained by varying energy of the ionizing beam. In
the screening mode this feature can be used to distinguish between aliphatic and
aromatic hydrocarbons and to exclude background gases from the instrument's
response. The HNU and Photovac can be used either in the survey mode (PID only),
or with GC. Sensitivity with PID alone is about 1 ppm, but can go down to as low as
0.1 ppb when a GC is used.
PI and Fl detectors used in the GC mode can be used for semiquantitative
analysis of compounds in ambient air. However, in areas where numerous
contaminants are present, identification of peaks in a complex matrix may be
tentative at best.
12-91
-------
TABLE 12-12
SUMMARY OF SELECTED ONSITE ORGANIC SCREENING METHODOLOGIES
Instrument
or detector
Measurable
parameters
Low range
of detection
Comments
Century Series 100 or
AID Model 550 (survey
mode)
Volatile organic
species
Low ppm Uses Flame lonization
Detector (FID)
HNU Model PI-101
Volatile organic
species
Low ppm Photo-ionization (PI)
detector-provides
especially good
sensitivity to low
molecular weight
aromatic compounds
(i.e., benzene, toluene)
Century Systems Volatile organic
OVA-128(GCmode) species
Low ppm
Uses GC column for
possible specific
compound
identification
Photo Vac 10A10
Volatile organic
species
Low ppm Uses PI detector.
Especially sensitive to
aromatic species. May
be used for compound
identification if
interferences are not
present
12-92
-------
Another method which can be used as a survey technique is mobile mass
spectrometry. Ambient air is drawn through a probe directly into the instrument,
which is usually mounted in a van. Particularly in the MS/MS configuration this is a
powerful technique which can provide positive identification and semiquantitative
measurement of an extrememly wide range of organic and inorganic gaseous
contaminants.
12.8.2.2 Quantitative Methods
Laboratory analysis of hazardous constituents in air includes the following
standard steps:
• Preconcentration of organics (as necessary to achieve detection limit
goals);
• Transfer to a gas chromatograph or HPLC (High Pressure Liquid
Chromatography); and
• Quantification and/or identification with a detector.
Broad-spectrum methods applicable to most common air contaminants are
discussed below.
12.8.2.2.1 Monitoring OrganicCompoundsin Air
Due to the large number of organic compounds that may be present in air, and
their wide range in chemical and physical properties, no single monitoring
technique is applicable to all organic air contaminants. Numerous techniques have
been developed, and continue to be developed, to monitor for specific compound
classes, individual chemicals, or to address a wide range of hazardous contaminants.
This last approach may be the most efficient approach to monitoring at units where
a wide range of chemicals are likely to be present. Therefore, methods that apply to
a broad range of compounds are recommended. In cases where specific compounds
of concern are not adequately measured by broad-spectrum methods, compound-
specific techniques are described or referenced.
12-93
-------
12.8.2.2.1.1 Vapor-Phase Organics
The majority of hazardous constituents of concern can be classified as gaseous
or (vapor-phase) organics. These constituents include most petroleum-related
hydrocarbons, organic solvents, and many pesticides, and other semivolatile organic
compounds. Methods to monitor these compounds generally include on-site
analysis (making use of onsite concentration techniques, where necessary), or
require storage in a tightly sealed non-reactive container.
Techniques for volatile and semivolatile organics measurement include:
• Adsorption of the sample on a solid sorbent with subsequent desorption
(thermal or chemical), followed by gas chromatographic analysis using a
variety of detectors.
• Collection of whole air (grab) samples in an evacuated flask or in Tedlar
or Teflon bags, with direct injection of the sample into a GC using high
sensitivity and/or constituent-specific detectors. This analysis may or may
not be preceded by a preconcentration step.
• Cryogenic trapping of samples in the field with subsequent instrumental
analysis.
• Bubbling ambient air through a liquid-filled impinger, containing a
chemical that will absorb or react with specific compounds to form more
stable products for GC analysis.
• Direct introduction of the air into a MS/MS or other detector.
Tables 12-13 (A and B), 12-14, and 12-15 summarize sampling and analytical
techniques that are applicable to a wide range of vapor phase organics, have been
widely tested and validated in the literature, and make use of equipment that is
readily available. A discussion of general types of techniques is given below.
12-94
-------
cc
O
X
a.
cr
O
a.
<
LJL
O
z
o
cr
O
LL.
t^>
UJ
O
o
o
Q
O
I
Q
O
Z
<
u
LL
O
>-
a:
<
(N
egative Aspects
z
Positive Aspects
.ti ^
.0 —
(0 «
•— *
^ *-
^^ 5
* ^
"<5 .2
_ W* ^"*
.2 E S S
.« O "^ — ~~"
— u. ^ *•* O
C rt ni "S >
10 _ C ? i-
<-> C *- C .C
c o a^ w
!jlf If
1 u 2 '5 2 o
E- • - ™ —
NIBS'S* 8
wide range of
applicability
> p/m3 detection limits
» practical for field use
o
c ^_
-is
1 ^ w
Ql S ®
K (?0
w
1 . Sorption onto Tenax-
GC or carbon molecula
sieve packed cartridge:
using low-volume
pump
"o
c
g
Q.
X c
S o
•o £ -
2 III
1 cS «
C 005
w Q. 5
2 ./« 2J S -o
*- "O 4> O V
5 § S « J
'I S ^ E £
5 2- E 5: U
? E "oS-2 E ^
c o S w 3 c
* ^ •£• m -c «U -^
-Q p £ ^ ^ u
W C ^^ ^pj O! ^^
Q. i^ JZ (0 ^ V
• large data base for
various compounds
> wide use in industrial
applications
> practical for field use
^
^~
•*~ £ ^O
> 9H fc
|i|
S > o
* 82"
_ 01
II. Sorption onto charcoal
packed cartridges usin
low-volume pump
^
(0
E
•o
c
3
O
C Q.
.2 E
w O
.£*
Hi
111
:E "o -a
'S * 3
S! S w
a o S
"O w^
» wide range of
applicability
> easy to preclean and
extract
i very low blanks
> excellent collection ai
retention efficiencies
» reusable up to 10 timi
2Z
^^
_
§^
"«3 .£
W '7
2 -£•
i i
** . .
C u. (/^
* ft. ^
o •*• ^
^00
01 "5 o>
*3i . ~ ^
3 -Q >«- w
CT ^! '-o 3
c a * o
EC ^^
^ o ^
C 5 ^ 'm
-C J^ ^ C
-o E o 5
4i 3 .. ,.
1 111
15 •- * £
10 -M X 'E
£ « c 0 =
c '" ^ *
^2 ^5 V ^^ 3
Q ^ W ^ ^
a2 § E 2
^^
1 S 1 J § o\ c
5 5 « i fl = .2
(oSSc^feoog1
1/1 a. a. c a. a. M_ u Q.
• • •
w
O
^
u
^^
Hi III
' v 4! * ^ ^3
< -5 5 -SiS o
M
IV Sorption on passive
dosimeters using Tena;
or charcoal as
adsorbing medium
12-95
-------
T3
• requires field use of liquid nitrogen
oxygen
• sample is totally used in one analysi
reanalysis possible
• samplerseasily clogged with water
• no large data base on precision or
recoveries
applicable to a wide
range of compounds
artifact formation
minimized
low blanks
• • •
E
0
'c
a
XJu
a o
o
oi-o
c o;
a. 01
2 6
C *^
^* ^5
s §
>
o
o> ^
• difficult to obtain integrated sampi
• low sensitivity if preconcentration i
used
^fl ^
c «
useful for grab sampi
large data base
excellent long-term
storage
wide applicability
allows multiple analy
• • • • •
="
?S5
'•53 | £>
« 5 .£ £ o
o>||o| £
O Q. q» O <9 •=
* 2 •?• «- •- c
u ^ .£ o o m
5^J
Ol
2 ^
_Oj $
9-"c
S 2
^ ^
•5 5
III
I.s5
>
- Ol
O Oi
c $
• long-term stability uncertain
• low sensitivity if preconcentration i
used
• adequate cleaning of containers be
samples may be difficult
Ol
ISI
grab or integrated
sampling
wide applicability
allows multiple analy
• • •
=
*^
W I) rti
2 3 *
•6 o 'is;
s els
w O .— '•*»-
= ? S s 2
Oi C O 2 o
o»'arj u -Q
o a Jj 0 js
c^ 2 •=• ». k
u i: .E o o
c
IQ
Ol
a
C Ol
§2
k ^
"o 01
^ t-
>
c
01 £
S 5
• fragile equipment
• sensitivity limited by reagent impur
• problems with solvent evaporation
long-term sampling is performed
c *^
m U
specific to aldehydes
ketones
good stability for
denvatized compoun
low detection limits
^
15
|
a.
r
o> I
•O.S 3
!'•= S i
^ o» r. o
o. c 5 — '
O N CT>
-------
TABLE 12-13B. LIST OF COMPOUND CLASSES REFERENCED IN TABLE 12-13A
Category
Types of Compound
Volatile, nonpolar organics (e.g., aromatic
hydrocarbons, chlorinated hydrocarbons) having boiling
points in the range of 80 to 200°C.
Highly volatile, nonpolar organics (e.g., vinyl chloride,
vinylidene chloride, benzene, toluene) having boiling
points in the range of -15 to +120°C
III
Semivplatile organic chemicals (e.g., organochlorine
pesticides and PCBs).
IV
Aldehydes and ketones.
12-97
-------
00
a.
a.
LU
13
g
z
i
u
LU
H
oo
Zui
< (/i
o>
E
E
0
u
TJJ Jp
J o _5
531
-o *
Z O> in
!!8
J E
"5 u. *?
JaJy
.— O*
c c
!» io
llfc. rtj 1 —
W k-
u >-
51" 01
Ol
c 32 *?
u u
01
||3
u
o»
0 .t
1 **
Compound
Name
X
X
aphenone
01
w
X
X
c
'Z
o
u
X
X
X
onitrile
>,
u
(N
i
(S
X
X
01
^
c
o
>
z
•o
l/^
3
TO"
o
0
o
en
lie and compounds
o>
X
X
X
X
Ol
0)
01
00
o
o
tn
ethyl hexyl)phalate
IN
ca
o
(N
in
(N
X
Q.
X
omethane
t
o
m
0
>
i
Tg
l/>
0)
•o
O
00
o
lium and compounds
c
u
o
0
lO
X
Q.
z
X
andisulfide
J3
U
ro
O
O
X
X
00
X
an tetrachloride
4
u
X
X
X
V
<0
o
VJ
73
0
£1
I
T3
wi
8
3
•o
O
^
IN
O
Is*
mium and compounds
o
u
o
i
T3
C^
s
3
•o
o
^
IN
O
er cyanide
a.
O
U
12-98
-------
c TJ
•?
Method 1003 ma
used
o
X
X
X
loropropane(1,2)
•5
o
X
n
z
X
loropropene(1,3)
u
Q
•o
01
| No method ident
lyl phthalate
oi
0
i
X
$
Yellow crystals, u
Vol
>trotoluene (2,4)
o
Q
(N
O
^
X
X
X
I
X
0
Q
•o
Oi
1 No method ident
^
lenylhydrazine (1.2
Q.
O
01
c
J£
«
I Syn: 1,2-dibromo
00
o
o
X
00
X
lene dibromide
£.
LU
Of
C
n
OJ
|Syn: 1,2-dichloroi
m
O
O
X
00
X
lene dichloride
Jz
ULJ
O
Std Hi-Vol for
paniculate fractn
(N
O
ON
$
•o
^
1
i
•o
Waxy solid, use Si
Vol
w
u
Q.
01
X
X
ichlorobutadiene
IQ
X
Oi
X
12-99
-------
m
Ol
£.
Comments/O
ifl
its
-1 E
"5 u. 't
•i Q. 2
'5 c «
V 'X i
g'aO
5* (^ h^
w ^
u >-
5 o»
c "o ,
2'S 0
•g t. t-
o»
X ^^
*• o
Ol
O .h
5 <
•n
Compound
Name
c
r.
n: perchloroel
w^
O
O
X
a.
z
| Hexachloroethane
o
o
w
3
O
C
o
X
Q.
Z
|lsobutanol
Ol
3
U
ostly particula
-Vol
5 f
CM
§
[lead and compounds
Oi
3
Oi
ostly particula
-Vol
5 I
o
0
•o
c
1 Mercury andcompou
X
rt
z
X
| Methacrylonitrile
n: 2-butanone
>,
Wl
o
o
in
(N
X
X
01
o
a;
£*
o!
"S
X
Q.
Z
X
| Methyl methacrylate
Ol
c
fO
.c
n: dichlorome
w»
o
o
X
X
CD
JMethylene chloride
VB
Oi
c
I '
2 x
in
in
in
X
X
1 Naphthalene
01
3
w
ostly particula
-Vol
§ f
o
o
no
Pv
v/t
1 Nickel and compound
in
O
o
> Oi
i/i t-
X
X
X
I Perchloroethylene
.
1/1
o
in
X
a.
Z
X
Ol
91
>.
12-100
-------
~o
« fO
Q.
Z
X
Ol
1
a
(Q
X
O
a.
z
Q.
X
| Trichlorobenzene
c
n: Methyl Chloroforr
>.
0
o
V
X
X
CO
X
[ Trichloroethane ( 1
x
x
x
x
| Trichloroethylene
X
x
x
IN"
^
[Trichloropropane 1
ostly particulate, use
-Vol
5 x
Ol
•a
[vanadium pentoxi
x
x
[vinyl acetate
n: 1.1-dichloroethen
vn
O
O
X
X
X
[vinyl chloride
x
X
X
aj
| Vinylidene chlorid*
[dichloroethylene)
n: dimethylbenzene
v^"
O
in
^
X
X
X
|xylene(m, o, p)
o
X
•o
V
T3
$
-o
no p».
in
0)
•5
0
w
M
O
u
5
Ol
«J
a- 5
l| i
|||
"2 .2 *? §
J » S'5
^ E 2 i
E^ o o
.. O 5> w
S > -o X
5 £ * ,2
<-> w w 41
3 m £ *
15|1
^ i
*n C
^ II II II
J2 o.
03 CO Z X
12-101
-------
TABLE 12-15
COMPOUNDS MONITORED USING EMSL-RTP
TENAX SAMPLING PROTOCOLS
2-Chloropropane
1,1-Dichloroethene
Bromoethane
1-Chloropropane
Bromochloromethane
Chloroform
Tetrahydrofuran
1,2-Dichloroethane
1,1,1-Trichloroethane
Benzene
Carbon tetrachloride
Dibromomethane
1,2-Dichloropropane
Trichloroethene
1,1,2-Trichloroethane
2,3-Dichlorobutane
Bromotrichloromethane
Toluene
1,3-Dichloropropane
1,2-Dibromomethane
Tetrachloroethene
Chlorobenzene
1,2-Dibromopropane
Nitrobenzene
Acetophenone
Benzonitrite
Isopropylbenzene
p-lsopropyltoluene
1-Bromo-3-chloropropane
Ethylbenzene
Bromoform
Ethenylbenzene
o-Xylene
1,1,2,2-Tetrachloroethane
Bromobenzene
Benzaldehyde
Pentachloroethane
4-Chlorostyrene
3-Chloro-1-propene
1,4-Dichlorobutane
1,2,3-Trichloropropane
1,1-Dichloroethane
2-Chlorobutane
2-Chloroethyl vinyl ether
1,1,1,2-Tetrachloroethane
p-Dioxane
Epichlorobutane
1,3-Dichlorobutane
p-Dichlorobenzene
cis-1,4-Dichloro-2-butene
n-Butyl benzene
3,4-Dichloro-1-butene
1,3,5-Trimethyl benzene
12-102
-------
Sprbent techniques--A very common technique used to sample vapor-phase
organics involves sorption onto a solid medium. Methods of this type usually
employ a low- or high-volume pump to pull air through a glass tube containing the
sorbent material. Organic compounds are trapped (removed from the air) by
chemical attraction to the surface of the adsorbent material. After a predetermined
volume of air has been pulled through the trap, the tube is capped and returned to
the laboratory for analysis. Adsorbed organics are then thermally or chemically
desorbed from the trap prior to GC or GC/MS analysis.
Thermal desorption is accomplished by rapidly heating the sorbent tube while
a stream of inert gas flushes desorbed organics directly onto the GC column.
Generally a secondary trap (either another sorbent or a cryogenically cooled loop) is
used to hold the organics until injection into the GC column, but this step precludes
multiple analyses of the sample.
Chemical desorption involves flushing the sorbent tube with an organic
solvent, and analysis of the desorbed organics by GC or GC/MS. Since only a portion
of the solvent is injected into the GC, sensitivity is lower than with thermal
adsorption. However, reanalysis of samples is possible. The most common
application of chemical desorption is for analysis of workplace air samples, where
relatively high concentrations of organics are expected.
The primary advantages of sorbent techniques are their ease of use and ability
to sample large volumes of air. Sorbent cartridges are commercially available for
many applications, and can easily be adapted to portable monitoring pumps or
personal samplers. A wide variety of sorbent materials are available, and sorbent
traps can be used singly or in series for maximum retention of airborne pollutants.
Sorbent methods are especially applicable to integrated or long-term sampling,
because large volumes of air can be passed through the sampling tube before
breakthrough occurs.
In choosing a sorbent method, the advantages and limitations of specific
methods should be considered along with general limitations of sorbents. Some
important considerations are discussed below.
12-103
-------
• Sorbents can be easily contaminated during manufacturing, shipping or
storage. Extensive preparation (cleaning) procedures are generally
needed to insure that the sorbent is free from interfering compounds
prior to sampling. Tenax, for example, is often contaminated with
benzene and toluene from the manufacturing process, requiring
extensive solvent extraction and thermal conditioning before it is used.
Once prepared, sampling cartridges must be protected from
contamination before and after sampling.
• No single adsorbent exists that will retain all vapor phase organics. The
efficiency of retention of a compound on a sorbent depends on the
chemical properties of both compound and sorbent. Generally, a sorbent
that works well for nonpolar organics such as benzene will perform
poorly with polar organics such as methanol, and vice versa. Highly
volatile compounds such as vinyl chloride will not be retained on weakly
adsorbing materials such as Tenax, while less volatile compounds will be
irreversibly retained on strong adsorbents such as charcoal. The optimal
approach involves use of a sorbent that will retain a wide range of
compounds with good efficiency, supplemented by techniques
specifically directed towards "problem" compounds.
• Tenax-GC is a synthetic polymeric resin which is highly effective for
volatile nonpolar organics such as aliphatic and aromatic hydrocarbons,
and chlorinated organic solvents. Table 12-15 lists compounds that have
*
been successfully monitored using a Tenax sorption protocol. Tenax has
the important advantage that it does not retain water. Large amounts of
water vapor condensing on a sorbent reduces collection efficiency and
interferes with GC and GC/MS analysis. Another advantage of this
material is the ease of thermal or chemical desorption.
The major limitation of Tenax is that certain highly volatile or polar
compounds are poorly retained (e.g., vinyl chloride, methanol).
Formation of artifacts (i.e., degradation products from the air
contaminant sample collected due to hydrolysis, oxidation, photolysis or
other processes) on Tenax has also been noted, especially the oxidation
12-104
-------
of amines to form nitrosamines, yielding false positive results for the
latter compounds.
Carbon sorbents include activated carbon, carbon molecular sieves, and
carbonaceous polymeric resins. The major advantage of these materials
is their strong affinity for volatile organics, making them useful for
highly volatile compounds such as vinyl chloride. The strength of their
sorptive properties is also the major disadvantage of carbon sorbents
because some organic compounds may become irreversibly adsorbed on
the carbon. Thermal desorption of compounds with boiling points above
approximately 80°C is not feasible due to the high temperature (400°C)
required. Carbon adsorbents will retain some water, and therefore may
not be useful in high humidity conditions.
In addition to the Tenax and carbon tube sampling methods shown
above, passive sorption devices for ambient monitoring can be used.
These passive samplers consist of a portion of Tenax or carbon held
within a stainless steel mesh holder. Organics diffuse into the sampler
and are retained on the sorbent material. The sampling device is
designed to fit within a specially constructed oven for thermal
desorption. Results from these passive samplers were reported to
compare favorably with pump-based sorbent techniques. Because of the
difficulty of determining the volume of air sampled via passive sampling,
these devices would appear to be mainly applicable for screening
purposes.
Polyurethane foam (PUF) has been used extensively and effectively for
collection of semivolatile organics from ambient air. Semivolatiles
include PCBs and pesticides. Such compounds are often of concern even
at verly low concentrations. A significant advantage of PUF is its ability
to perform at high flow rates, typically in excess of 500 liters per minute
(l/m). This minimizes sampling times.
PUF has been shown to be effective for collection of a wide range of
semivolatile compounds. Tables 12-16 and 12-17 list compounds that
have been successfully quantified in ambient air with PUF. Compounds
12-105
-------
s
>
D
^
*^
E
>
> ^
?• ^.
5i
J!^
LI O^
J^
•• ^J
!3<
-»Z
c<
pi-
-N ^
— H
U ^J
^O
5m
5 3
^ ^O
Cj
»r UKUANIV. V_VJIVlr-UUIMU:>
^NE FOAM SAMPLER AN
JRE DETECTOR (GOECD)a
3i=H
3a.
Z*<
_ ^ v^
3-^0
HQ.QC
si LU 0
E^ ^2 LU
c "l uu
•3 5 ^
^ ^ "
b
N
^
LI
_l
3
1
vn .O .O -O .O .O .O
O O O O O O O
c c c c c c c
rtj QJ (K QJ 4; *n i/*> '.p u
m no ro ^ ^ ^
(N (N (N (N rv( f»
owax 1001C
owax 1013'
orinated Benzene
<0 <0 .C
XX (J
$
;o
•^
Of
5 ^ 1
« 2 5
^O T
o -§ \
X o ]
< U i
•o -a -o
O O O
c c c
01 Ol Ol
: Q. Q. Q. •n
5229o
i « o o £
W W W Q
3 2 2 2 o
; o! o! ol o
- 1- t- (- 1=
• III <^~
- U"» VD ^D ^
• * » ffl
i or) m ^
•> PM IN IN Q.
,3-Trichlorobenzenea
,4-Trichlorobenzened
,5-Trichlorobenzened
,3,4-Tetrachlorobenzenea
6 5 0
,3,5-Tetrachlorobenzened
,4, 5-Tetrachlorobenzened
(N (N
fl
"a?
(0
|
w rr
x ¥
00
itachlorobenzenea
01
Q.
?.
VJ
"*•
•
(achlorobenzenea
o>
X
•o
••
Q
^^
d
itachloronitrobenzenea
y
Q.
n
UJ
Q
"o.
ol
>s <£
£ ^
•S E
?l
C w
O
Z w S -a
,_ * ~C 2
O^ rti
?•— ._ O
« -c u j:
> in the literature for either a Ic
il protocols suggested for the
med that sample collection effi
tion/retention efficiency data s
v nj o w
3 i '2 «
-^^5
'(5 10 w ^J
> C oi
_>. „ JJ "5
1 1 s 5
* oi ^ o"
0-5 -jo
IS of,
•0 JJ v £
75 -g 2.?
T3 2 C 4>
1 |l
10 C 3
•O Ol <0 w
^^ in ^C
I j'|l
•c 1 lo 1
•J > 1" ^
oi a 5 w
5i 2Z
n
C
V
Ol
oi
Ol
$
0
^
^
(O
-C
^rf
c
Ol
o
a.
o
w
Oi
3
2
5
'••
c
Oi
JD
'3
>
m
ling conditions not presently
a
C
2^
01 01
jl
o *o
^^
5 z)
"~ 01
.i1!
o "5
^ >
O^ £
oi 2
<0 O)
T3 35
3
!*
** <->
(0 <0
•0 3
'"i 5
-T ^
> a>
i)
"a
E
TJ
o>
E g
— LU
o r^
> ^
5
o g^
j_ •—
O> Q
"?•§.
C u.
3 E
0 *
•I 1
? i
•gi
co S
3; n}
3 '*-
fl3 ^
^ ^w
4^ C
*^ to
Q
« J
J2 c
Tfl •_
> E
low volume PUt- sampler not a
isomerj listed. Component is a
?«
C tT
11
Q- ?
I!
?!
o> o
E o
2 t!
*S1 t»
«2 §
<0 "3!
To ^
-o -°
c 2
O 3
— O
<0 Si
"O «•
;^ <«
(0 C
> o
1 ^
- c
LJ 3r
-------
•z.
o
u
o
u
{£.
O
(/I
UJ
o
Q.
Pyrethrin 1
O
i 'is '= *
f 1 f § S |
k- i. jr w w ^ c
^ ^ < -6 S 5 tS
LU
>r>
°ci
5$
DO
Wt -J
r<
«N
LU
_J
CD
<
u
^L
wi
w
U
«
LU
Herbicide
w
Q.
i/f
Jj
LU
o
IN"
c
^
iq
w
o
w
^
>
u
'^
fO
c
5
Polynuclear Arc
Napthalene
|
2
0)
ts
LU
Q
fN*
">.
1
Q.
m
^
J3
O
v^
LU
Q
i
PM~
Fluorene
^
|
s
LU
O
1
IN"
41
^
in
Dibenzothiophu
Phenanthrene
f
0*
Q.
i/i
3
O
o
Q
0
Orqanoph
-e,
Mevinpho
«
c
V
u
Anthracene
Carbazole
2-Methylanthra
w
Dichlorvos
«»
21
£.
«*
C
1 -Methyl phena
^J
u w
§ ft 3
* f, '? 1
O f^ ^w ^^
2j ^ ^^ ^
*w C Q. ^ ^
— >* O — IS P
g a- ! £ .? 1
§ _O N -S >. £
—
.!£ (J;
^'i 1
-C QJ O
w 'Q C
** <^ AJ
5^8
^ § -
« .2 c
.2=5
*^ *•> *^
2 01 w
fO u Q,
C 1C
w QJ w
> ^ ^
1^>
w c .c
C 0 w
'K '»- C
3 <0 Si
oS E
!= ^5 3
1 * °
2 -o T>
S i £
««5 *
* «c 5
55 = ^
.£«!}•!
3 S'= s
C w v> c-
o « a» •-
a^; 5 _*
i?11
H'-e i&
?i?s
5. II
^ «» e c
These component
provided solely a
employing the sa
document may be
3;
.a
h/l
(U
JZ.
a*"
>
0)
§
£.
<.
01
C
o
a
£
o
VJ
Oi
w
3
2
5
—
c
0>
£>
2.
5
us
c
Of
J
Q.
O
c
I/I
c
o
'•5
o
u
Ol
"a
E
dl 4^
|1
|s
mploying low
gh volume PUF
Validation data e
evaluated using hi
QJ
3
2
5
«
cu
w
•o
(U
c
5
1
w
^™
V
a
E
2
ix.
a.
IV
£
3
O
>
5
o
^™
01
3
a!
S
~
c
1
o
\J
8
^
0
**—
-------
that have shown poor retention or storage behavior with PDF include
hexachlorocyclohexane, dimethyl and diethylphthalates, mono- and
dichlorophenols, and trichloro- and tetrachlorobenzenes. These
compounds have higher vapor pressures, and may be collected more
effectively with Tenax or with resin sorbents such as XAD-2.
PDF is easy to handle, pre-treat, and extract. Blanks with very low
contaminant concentrations can be obtained, as long as precautions are
taken against contamination after pretreatment. Samples have been
shown to remain stable on PUF during holding times of up to 30 days.
PUF concentration methods have shown excellent collection efficiency
and recovery of sorbed compounds from the material.
Most PUF methods specify the use of a filter ahead of the PUF cartridge,
to retain particulates. The filter prevents plugging of the PUF which
would reduce air flow through the sorbent. Some methods reo -nmend
extracting the filter separately to obtain a value for particulate organics.
However, because most semivolatile compounds have sufficient vapor
pressure to volatilize from the filter during the collection period,
particulate measurements may not be representative of true particulate
concentrations. Therefore, results from the PUF analyses may
overestimate gaseous concentrations of semi-volatile compounds due to
volatilization of semi-volatiles originally collected on the sampler inlet
filter and subsequently collected by the PUF cartridge.
• Cryogenic methods for capturing and collecting volatile organics involve
pulling air through a stainless steel or nickle U-tube immersed in liquid
oxygen or liquid argon. After sampling, the tube is sealed, stored in a
coolant, and returned to the laboratory for anlaysis. The trap is
connected to a GC, rapidly heated, and flushed into a GC or GCMS for
analysis.
The major advantage of cryogenic concentration is that all vapor phase
organics, except the most volatile, are concentrated. This is a distinct advantage
over sorbent concentration, which is especially selective for particular chemical
12-108
-------
classes. Contamination problems are minimal with cryogenic methods because a
collection media is not required.
Several disadvantages limit the current usefulness of cryogenic methods,
including:
• Samplers rapidly become plugged with ice in high humidity conditions.
This limits the volume of air that can be sampled.
• The entire sample is analyzed at once, enhancing sensitivity but making
multiple analyses of a sample impossible.
• The necessity of handling and transporting cryogenic liquids makes this
method cumbersome for many sampling applications.
• There is a possibility of chemical reactions between compounds in the
cryogenic trap.
Whole air sampling-Air may be collected without preconcentration for later
use in direct GC analysis or for other treatment. Samples may be collected in glass or
stainless steel containers, or in inert flexible containers such as Tedler bags. Rigid
containers are generally used for collection of grab samples, while flexible
containers or rigid containers may be used to obtain integrated samples. Using a
flexible container to collect whole air samples requires the use of a sampling pump
with flow rate controls. Sampling with rigid containers is performed either by
evacuating the container and allowing ambient air to enter, or by having both inlet
and outlet valves remain open while pumping air through the container until
equilibrium is achieved.
Whole air sampling is generally simple and efficient. Multiple analyses are
possible on samples, allowing for good quality control. This method also has the
ability to be used for widely differing analyses on a single sample. The method has
been widely used, and a substantial data base has been developed.
Problems may occur using this method due to decomposition of compounds
during storage and loss of some organics by adsorption to the container walls.
12-109
-------
Sample stability is generally much greater in stainless steel containers than in glass
or plastic. Whole-air sampling is limited to relatively small volumes of air (generally
up to 20 liters due to the impracticality of handling larger sample collection
containers), and has higher detection limits than some sorbent techniques.
Impinqer collection-lmpinqer collection involves passing the air stream
through an organic solvent. Organics in the air are dissolved in the solvent, which
can then be analyzed by GC/MS. Large volumes of air sampled cause the collection
solvent to evaporate. In addition, collection efficiency is dependent on flow rate of
the gas, and on the gas-liquid partition coefficients of the individual compounds.
However, there are certain specialized applications of impinger sampling that have
been found to be preferable to alternate collection techniques (e.g., sampling for
aldehydes and ketones).
Certain compounds of interest are highly unstable or reactive, and will
decompose during collection or storage. To concentrate and analyze these
compounds, they must be chemically altered (derivatized) to more stable forms.
Another common reason for derivatization is to improve the chromatographic
behavior of certain classes of compounds (e.g., phenols). Addition of the
derivatization reagent to impinger solvent is a convenient way to accomplish the
necessary reaction.
A widely used method for analysis of aldehydes and ketones is a DNPH
(dinitrophenylhydrazine) impinger technique. Easily oxidized aldehydes and
ketones react with DNPH to form more stable hydrazone derivatives, which are
analyzed by high performance liquid chromatography (HPLC) with a UV detector.
This method is applicable to formaldehyde as well as less volatile aldehydes and
ketones.
Direct analvsis--A method not requiring preconcentration or separation of air
components is highly desirable, because it avoids component degradation or loss
during storage. Air is drawn through an inert tube or probe directly into the
instrument detector. Several portable instruments exist that can provide direct air
analysis, including infrared spectrophotometers, mobile MS instruments, and
portable FID detectors. Some of these instruments have been discussed in the
section on screening methods.
12-110
-------
Mobile mass spectrometry has been used to compare upwind and downwind
concentrations of organic pollutants at hazardous waste management facilities.
The advantage of the multiple mass spectrometer configuration (MS/MS or triple
MS) over a single MS system is that multiple systems can identify compounds in
complex mixtures without pre-separation by gas chromatography. Major
limitations of MS/MS methods are low sensitivity and high instrument cost.
In summary, of the methods described in this subsection, the majority of
vapor-phase organics can be monitored by use of the following sampling methods:
• Concentration on Tenax or carbon adsorbents, followed by chemical or
thermal desorption onto GC or GGMS.
• Sorption on polyurethane foam (PDF) cartridges, followed by solvent
extraction.
• Cryogenic trapping in the field.
• Whole-airsampling.
12.8.2.2.1.2 Paniculate Organics
Certain hazardous organic compounds of concern in ambient air are primarily
associated with airborne particles, rather than in the vapor phase. Such compounds
include dioxins, organochlorine pesticides, and polyaromatic hydrocarbons.
Therefore, to measure these compounds accurately, it is necessary to monitor
participate emissions from units of concern.
Measurement of particulate organics is complicated because even relatively
nonvolatile organics exhibit some vapor pressure, and will volatilize to a certain
extent during sampling. The partitioning of a compound between solid and
gaseous phases is highly dependent on the sampling conditions (e.g., sampling flow
rate, temperature). Particulate sampling methods generally include a gas phase
collection device after the particulate collector to trap those organics that become
desorbed during sampling.
12-111
-------
The most common methods used for collection of particles from ambient air
are:
• Filtration
Cellulose Fiber
Glass or Quartz Fiber
Teflon Coated Glass Fiber
Membranes
• Centrifugal Collection (e.g., cyclones)
• Impaction
• Electrostatic Preciptation
The standard sampling method for particulates is filtration. Teflon-coated
glass membranec generally give the best retention without problems with
separating the particulates sampled from the filter. Problems, however, may be
caused by desorption of organics from the filter, by chemical transformation of
organics collected on the filter, and with chemical transformation of organics due
to reaction with atmospheric gases such as oxides of nitrogen and ozone. These
problems are magnified by the large volumes of air that must be sampled to obtain
sufficient particulate material to meet analytical requirements. For example, to
obtain 50 milligrams of particulates from a typical air sample, 1000 cubic meters of
air must be sampled, involving about 20 hours of sampling time with a high-volume
sampling pump.
Despite the drawbacks mentioned above, filtration is currently the simplest
and most thoroughly tested method of collecting particulates for organic analysis.
Other methods, such as electrostatic precipitation, make use of electrical charge or
mechanical acceleration of the particles. The effect of these procedures on
compound stability is poorly understood.
12-112
-------
12.8.2.2.2 Monitoring Inorganic Compounds in Ambient Air
12.8.2.2.2.1 Participate Metals
Metals in ambient air can occur as participates or can be adsorbed on other
paniculate material. Metals associated with particulate releases are effectively
collected by use of filter media allowing for the collection of adequate samples for
analysis of a number of particulate contaminants.
Collection on filter media-Sampling methods for particulate metals are
generally based on capture of the particulate on filter media. For the most part,
glass fiber filters are used; however, organic and membrane filters such as cellulose
ester and Teflon can also be used. These membrane filters demonstrate greater
uniformity of pore size and, in many cases, lower contamination levels of trace
metals than are found in glass fiber filters. Analytical procedures described in the
following reference can be utilized to analyze particulate samples.
U.S. EPA. 1986. Test Methods for Evaluating Solid Waste. 3rd Edition. EPA
SW-846. GPO No. 955-001-00000-1. Office of Solid Waste. Washington, D.C.
20460.
Hi-Vol collection devices--The basic ambient air sampler is the high volume
sampler which can collect a 2000 cubic meter sample over a 24-hour period and
capture particulates on an 8 x 10 inch filter (glass fiber) as described in 40 CFR Part
50. It has a nominal cut point of 100um for the maximum diameter particle size
captured. A recent modification involves the addition of a cyclone ahead of the
filter to separate respirable and non-respirable particulate matter. Health criteria
for particulate air contaminants are based on respirable particulate matter.
Personnel samplers-Another particulate sampling method involves the use of
personnel samplers according to NIOSH methods (NIOSH, 1984). The NIOSH
methods are intended to measure worker exposure to particulate metals for
comparison to OSHA standards. A 500-liter air volume is sampled at approximately
2 liters per minute. This method is most efficient when less than 2 mg total
particuiate weight are captured. Capture of more than 2 mg may lead to sample
12-113
-------
u
ee.
O
o>
<3
4-1
C
ITJ
1
Procedures summary
-o
0)
W
3
Ol
«
w
1
01
0)
5
"ai
oe
1
£
01
C
? ^
5 | § 8 o
^ fl •— 3 _ C
3SfljSP^^»
O '«z O ^ •" ,* O
•- 9- * 2 = ° •- "<5
iii i
= o
"O " o»
® "o *
1 go?
•o § a. >, * >,
1 i i i i 1 i
£ 1 I 5 E 1 E
^ >» 2£
. o o» *-
Sampling train consists of
membrane filter to capture
particulate Hg, followed by
Carbosieve B to trap organic Hg
and then silver coated Chromos
P (CP) to collect elemental Hg.
Each section is analyzed separat
by thermal desorption into a
flameless AA. Filters are acid
digested, reduced to Hg and
amalgamated on Ag CP prior to
AA analysis step.
•o
c
(Q
W ^
c ?
fw *
0> w
o 1
**•!
Jv ^^
| i
CL U
1
2
j^
aB
Q.
I/I
g
*
Requires complex thermal
desorption unit
i
k.
~o 5 *
O rf —
- u u •£ § ^ £
E °- S ° o £ —
^g W ^t '^ -™ ^
w *"^ ^ ^H ^ ^ 2
•S .1 r= 3 8 ot'g'5.
i/>O<~ E w oe 3.1/1
^
o> ^
Filter to separate particulate;
adsorb organic Hg on Carbosiev
B; thermally desorb into flamel<
AAunit
|-
w
V
^
'c
0»
O
"* ^
O tQ ?
f i * c o £
£ — Ji * *• "T
"O § (0 " 01 °-
llit«cf
2 * o>'5 (^ j o
\fl LJ 3- LL. >/l .U >
5
<
Collection in acidified 0. 1 N HCI
impinger solution; analysis by N
or optionally by cold vapor AA
•o 3
tn ">
JS 3
3 O
ii
(Q (Q
a. >
0
^
o
f
Q.
LU
O ft. T-I
KMn04 reagent must be
prepared within 12 hours
use
Short sample holding tim<
Reagent can be easily
expended in oxidizing an<
organic matrices
i i i
>.
-o g ?
o •— o
w JS 5 0> nj
E 1 "o * f- - -5
•go S^-ji-*
|l|f |ll|i
^Alisl"^
^— ^3
o ^
Collection in impinger solution
10% H2SC>4/2% KMnO4; analysi
cold vapor AA
?!
s s
^_ 3
3 O
- 0
C a
£ 2
1
£|
LU 2
S "S
"5 4?
1 §
U L^
12-115
-------
losses during handling of the sample. The preferred filter medium is cellulose ester
(47 mm diameter) which will dissolve during the standard acid digestion.
The NIOSH method, however, is not recommended for the RFI for several
reasons. The NIOSH analytical methods (and good QA/QC practices) require several
aliquots of the sample to be prepared for best analytical results. The 47 mm filter is
too small for aliquoting; therefore, use of the NIOSH method would require the
simultaneous operation of several sampling systems. More importantly, the 500
liter sample volume generally does not provide sufficient particulate matter for the
analytical methods to detect trace ambient levels of metals. The method is best
suited for industrial hygiene applications.
Dichotomous Samplers-Dichotomous samplers (virtual impactors) have been
developed for particle sizing with various limit cutpoints for use in EPA ambient
monitoring programs. These samplers collect two particulate fractions on separate
37 mm diameter filters from a total air volume of about 20 cubic meters. The
standard sampling period is 24 hours. Teflon filters are generally recommended by
sampler manufacturers because they exhibit negligible particle penetration and
result in a low pressure drop during the sampling period. However, glass fiber and
cellulose filters are also acceptable.
The need for multiple extractions would require multiple sampling trains. If
the two filters are combined to form one aliquot and extracted together, they will
provide sufficient sensitivity for some but not all analytical procedures and defeat
the purpose of fractioning the sample. The use of the dichotomous sampler is,
therefore, limited.
12.8.2.2.2.2 Vapor Phase Metals
Most metallic elements and compounds have very low volatilites at ambient
temperatures. Those that are relatively volatile, however, require a different
sampling method than used for collection of particulate forms, although analytical
techniques may be similar. For the purpose of ambient monitoring, vapor-phase
metals are defined as all elements or compounds that are not effectively captured
by standard filter sampling procedures. Available methods for the measurement of
vapor phase metals are presented in Tables 12-18 and 12-19. These available
12-114
-------
c
o
u
30
^
N
Disadvantages
Advantages
w_
Procedures summai
1
3
*/»
$)
E
.5
4>
i/l
41
C
OJ
41
"aJ
ce
O
JZ
^^
41
2
^
Complex
desorption/amalgamatio
unit
i
^3 ^**
O *^^
11 fill
ii|l||
^ ^ .Q
oi 4i ** T5 S
** w a — u
Vaporous mercury is collec
amalgamation on silver.
Paniculate is collected on
microquanz filters. Both a
analyzed by thermal desor
and/or pyrolysis with re-
amalgamation; then them
desorption for determinati
UV absorption at 253 7
5 *
^ w
w 3
y At
| E
Vt *^
o "3
w n
I'S
> Q.
•o
(fl
O
C
c
W
o
.i
c
LU
u
01 •£ o>
1 1 {{
ti o ^ £
> U nj fl
> >/> > ^
"* "O V Si w
Jlli 111
rsj i£ W 2i c 2r t.
_-* *JZ — tZ flj — O
u v X E h- "O w
i i i
•3 >• *
5s ^ ri* rt ^
•g « I'S .i 2 1 1
>>u£54i^5Jc'|IJQ-
>EE523o5So
"o
.2
OI >.
I m
Passive device-diffusion of
through membrane,
amalgamation on gold, an
badges performed by 3M
rt
i
>
X
2
C
4*
LU
41
•n*
fO
00
i
c
.2 _ E
tn >s C G^
Large coefficient of varii
Quality of results are ver
much operator dependei
Only works well at 200 M
. . ,
c
imple equipme
;quirements
ange of 50-200
g/m3 tested
v/S w cc a.
c
41 ai
U KJ
Adsorb mercury on iodine
impregnated charcoal; pla
tantalum boat and volatili,
u
'c
fl
OT
0
•o
"3
c ?
01 3
E Sf
rti (]|
S E
•§
^
oJ
^
^
. "O
Does not provide for ana
of paniculate mercury
Insufficient performance
in available literature
.
- 4)
C 0)
imple equipme
squirement
valuated in ran
f 50-200 pg/m3
1/1 Z LU O
Jf
^
> o"<
Adsorb on hopcalite; disso
sorbent and mercury in HN
HCI; analyze by cold vapor
w
'c
?
O
•a
(Q
15
c ?
V w
iS E
•§
^
OI
4)
.ti
-------
C
O
u
00
*•••
PM
Disadvantages
VI
0*
cn
fl
*^
(0
>
•o
<
summary
>
V
3
-o
41
u
O
£
•o
£
3
)ri
3
E
4!
'0
a
V
..£ -a ai
flj « — — Q O ^»
_JJ *w ^O ^^ 21 rtj
u o O 2 2! ^
i i
m
a; -o £ w i/>
** ^ 5 BI o ^ o
| gl 5 .2 2 ~ilr|
^5 O ^ V* S C gi C *^ "**»
w '£! . „ § -5 jj ^ 3 c a325
^" ^^ Ji" ^ C ® fli ^ ^ fli ^ /^
2 §^"t:^ « i'a'"^! a"5-S0
-O>S^;''O'2!n»E^>niEc'^"
^ls.la2<5i* v< "0
w 0> £
a* -o —
— "3 c
c § ^2
o o> "*•
If-?
3 &£c
£=15.2
"5 o> -c o
£ * -^ jS
< '7i 5 <0
15
c
0>
E
41
V
E
1
c
o
'^
E
| g z
17 E ^
TJ
0)
Dithizone method suffers
from high blanks,
interference from SC>2 and
interference from several
other metals
Mercury compounds colled
in HCI are unstable
i i
"o
S'o >>
fO ^i 4f
*-" aj
c j jj
y— *- J2
•*- = O
,
— 01
• £.
2 ^.-o
C 41 •£
O N rti
3t e
ns-s
w • * t
3 3
u 0
§ 1
» ^
Is
fl O
Ol Q.
6 3
0)
o
M
5
Q
C
'5.
E
E
Monitor costs $3500-$4000
May suffer interference fro
oxidants as noted for 3M
badges
. i
(jj m
°- 1
- £ 1 1 I 2
.O "^ ^, (Q *ji C
* -*. 2 w "C i3 Qr^
SO *^ C "^ ^
w w « .- S- J. 2 g,]?
— oj2E£iflC^E
JlES^isiuSS
i t >
c
— o
O '•? 4;
c 2 u
o -fc c
1§3
IS s
15 « '5
E =^
^ rfi —
w (j^ O
O c O^
•5 -.s
c "5 O»
E o> c
•^ o "5
C Ol i.
0 Z 1"
^
^
^
15
c
01
Uj
| = l
E 5T >
2-8 w
- 1 E 5
P r'11 >
o 2-215
lua <
1/1
o
.c
•»^
0)
0)
c
o»
o
u
(U
DC
12-117
-------
a:
Z)
U
cr
u
x
i^i
<
u
<
cr
wo
QC
O
a.
<
>
u_
O
£
uo
_l
<
Z
<
Q
Z
<
O
z
_l
a.
<
ui
w
i
(M
LU
_l
00
<
Disadvantages
vantages
•o
<
Procedures summary
•o
*~ 2
W
~ 2
Possible breakthrough
high concentrations
Earlier version of P&CA
265
S
o'S
£ 8,
*•
01 D TJ ^ a*
E S ^ 3 >
S. i | 2 S
c 2 £ 2-5
O a.E -^ te.
,
Absorb in dilute NaOH solution;
analytical procedure not specified
but it may be suitable to use arsine
generation or furnace AA
•o
(O
o £
*f -^
< o
c
o
1 2 >,
S 1 '£
-H >- «
? u 3
° ?
a-o -j
C ^K ^
1 c 6
U 0 '^
"o o ^ ^ o t;
_c •;; « "O ^ ?
||i|Sgus s
m §5"° ?J[c"c E2"^
ii ii
Adsorb on XAD-2; desorb with
pentane; analysis by GC
•o $
V flj
^^ £
£ fO
a! |j
(Q *^
5 "c "S
H- (o .2!
•i
00
w^
X
O flft
z 3?
•a
_i
c
Very little information
literature
Dithizone method may
have same problems
noted elsewhere for
other elements
>> c .
2 2 r
.5 u m E
w 4» c —
S * -S, S E
. 0 0> -0 ? .2 -5
d* f aj "^ •* O*
il|2lo
!= 0* ;? *- "O ,
I!ll52
i i i
Collect in HCI/NiCI impinger
solution; analyze by dithizone
colonmetric method when 8-hour
sampling period or by AA for 24
hour sample
^
_^ ^
<0 ^
- i
— Q.
^ E
< S
12-118
-------
'C 3
M
v
c
2
V
c
E
LU
No data available
Dithizone method may
have interferences as
noted above
i i
e
c —
O f
Good collecti
efficiency
Low detectiol
possible
,
^
rti
Adsorb on activated carbon; dig«
with HN03 + HCI04 ; analyze by
dithizone method
.1 0
— a
-^ b
< 3
Sorbent capacity limits
upper concentration
,
v 1
Standard met
AA specific fo
Nickel
Range 2-60 u
i i i
5
^l
Adsorb on charcoal; desorb with
dilute HNOa; analyze by furnace
c
_||
U M •"""
z 25
u
oa
Q.
X
I/I
Z m
U
w
Z
Not a standard method
Interference may occur
from other Nickel
compounds, Cu, Pb, Cr,
Se and V
i i
. —
Detection lim
0 001 ppm
i
O_
Absorb in 3% HCI impinger solut
analyze by colorimetric method
which color development in
chloroform phase is measured
^
o
|3
W M ^-*
z 2S
«N
*•
o"
«N
*~
oe.
No data to support this
method
,
Only method
suggested in
literature for
volatile Se
.
c
^ T aj
Collect in impinger with aqueou
solution of Na2S03, Na2S, or NaOl
analyze by NAA, AA. GC,
colorimetry. fluorimetry. ring ov
techniques, or catalytic methods
m
X*
IN
O
X
E
3
'c
—
-------
methods are generally developed for industrial hygiene applications by NIOSH.
The methods for measuring vapor-phase metals presented in Tables 12-18 and
12-19 have undergone limited testing for precision and accuracy and have had
matrix interferences documented. Therefore, they should be used in lieu of any
methods which have no supporting data.
Several methods are suitable for quantification of vapor-phase mercury. If
elemental mercury is to be measured, the silver amalgamation technique with
thermal desorption and flameless AA (atomic absorption) analysis is recommended.
This technique is presented in American Public Health Association (APHA) Method
317, which can achieve nanogram per cubic meter detection limits. If organic and/or
particulate mercury are also to be determined, NIOSH methods (NIOSH, 1984) are
recommended. These methods can measure all three airborne mercury species, but
require a complex two stage thermal desorption apparatus.
12.8.2.2.2.3 Monitoring Acidsand Other Compounds in Air
Monitoring for acids and other inorganic/non-metal compounds (e.g.,
hydrogen sulfide) in the ambient air will generally require application of industrial
hygiene technologies. Applicable methods have been compiled in the following
references:
NIOSH. February 1984. NIOSH Manual of Analytical Methods. NTIS PB 85-
179108. National Institute for Occupational Safety and Health. Cincinnati, OH.
ASTM. 1981. Toxic Materials in the Atmosphere. ASTM, STP 786.
Philadelphia, PA.
APHA. 1977. Methods of Air Sampling and Analysis. American Public Health
Association.
ACGIH. 1983. Air Sampling Instruments for Evaluation of Atmospheric
Contamination. American Conference of Governmental Industrial Hygienists.
Cincinnati, OH.
12-120
-------
12.8.3 Stack/Vent Emission Sampling
EPA methods for source-sampling and analysis are documented in the
following reference:
Code of Federal Regulations. 40 CFR Part 60, Appendix A: Reference
Methods. Office of the Federal Register, Washington, D.C.
Additional guidance is available in the following references:
U.S. EPA. 1978. Stack Sampling Technical Information, A Collection of
Monographs and Papers, Volumes Nil. EPA-450/2-78-042 a, b, c. NTIS PB 80-
161672,80-1616680,80-161698. Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711.
U.S. EPA. February 1985. Modified Method 5 Train and Source Assessment
Sampling System Operators Manual. EPA-600/8-85-003. NTIS PB 85-169878.
Office of Research and Development. Research Triangle Park, NC 27711.
U.S. EPA March 1984. Protocol for the Collection and Analysis of Volatile
POHC's Using VOST. EPA-600/8-84-007. NTIS PB 84-177799. Office of Research
and Development. Research Triangle Park, NC 27711.
U.S. EPA. February 1984. Sampling and Analysis Methods for Hazardous
Waste Combustion. EPA-600/8-84-002. NTIS PB 84-155845. Washington, D.C.
20460.
U.S. EPA. November 1985. Practical Guide - Trial Burns for Hazardous Waste
Incinerators. NTIS PB 86-190246. Office of Research and Development.
Cincinnati, OH 45268.
U.S. EPA. 1981. Source Sampling and Analysis of Gaseous Pollutants. EPA-
APTI Course Manual 468. Air Pollution Control Institute. Research Triangle
Park.NC 27711.
12-121
-------
U.S. EPA. 1979. Source Sampling for Participate Pollutants. EPA-APTI Course
Manual 450. NTIS PB 80-188840, 80-174360, 80-182439. Air Pollution Control
Institute. Research Triangle Park, NC 27711.
U.S. EPA. 1986. Test Methods for Evaluating Solid Waste. 3rd Edition.
EPA/SW-846. GPO No. 955-001-00000-1. Office of Solid Waste. Washington,
D.C. 20460.
12.8.3.1 Vapor-Phase and Particulate Associated Organics
Generally, point source vapor-phase samples are obtained from the process
vents and effluent streams either by a grab sample technique or by an integrated
sampling train. Careful planning is necessary to insure that sampling and analytical
techniques provide accurate quantitative and qualitative data for measurement of
vapor-phase organics. Considerations such as need for real-time (continuous) versus
instantaneous or short-term data, compatibility with other compounds/parameters
to be measured, and the need for onsite versus offsite analysis may all be important
in the selection process.
Monitoring for complex organic compounds generally requires detailed
methods and procedures for the collection, recovery, identification, and
quantification of these compounds. The selection of appropriate sampling and
analytical methods depends on a number of important considerations, including
source type and the compounds/parameters of interest. Table 12-20 lists several
sampling methods for various applications and compound classess (applicable to
combustion sources). The first three methods listed are fixed-volume, grab-
sampling methods. Grab sampling is generally the simplest technique to obtain
organic emission samples.
Sample collection by the bag and canister sampling methods can be used to
collect time-integrated samples. These methods also allow for a choice of sample
volumes due to a range of available bag sized (6, 12, and 20 liter capacities are
typical). Bags of various materials are available, including relatively inert and
noncontaminating materials such as Teflon, Tedlar, and Mylar. All sample collection
bag types may have some sample loss due to adsorption of the contaminants
collected to container walls. The bag sample is collected by inserting the bag into
12-122
-------
I/I
LU
u
cr
)-
"Z
o
a.
5
O
cc
LL.
UJ
<
U
3)
O
Q
a:
<
IN
X
o
O
LL.
«/l
a
o
x
a.
o
«ll
u TJ^ O
Q. C &
< < 5
V
a.
— h—
"S ~v
"5. o
a a
< E
o
u
« a
J3 >.
itj t-
w q,
IS
5
w
a, a> ^ rt !^
•2 * u o
V c V O.
w '-5" E
5'| 3 S
T3 ffl 4) 41
4/ ° TT Q. &f
i!r. h
C C * CT v"
^ 0 .N 2 0
o
•9 a w
j_ ja ]lfc
Q v/» Q
iZ 5 Q.
U U U
(3 (O (3
i i ^_ i
U jj (J tj
u^>i7
c
o
1 1 1 q, -^ I C tf
^ *^ O ^ jj */* .^ *^ ^j
o 1 5 5 *. | c 1 1 g" 2
illillllllll
It
3. 3 w w
Q Q W wl
c c "2 "S
10
S S £ £
^
f w|
w .C 2 w J3 S
C* i — 3 *» ^ rf
*C S *^ <0 *C "^3 a"
>..2 o > fl 2 o
iX
•O a;
c .-=
V ^
~ > 10
115
c
o
3 e ««
i o "
C fli *^
W _O w 0* _Q 'v* ^
| "i 1 | | I 1
Z *0 C w w <1* rtj
.£ "o
tu 5
w 2
** o
23
01 ^.J<
•9 §• c
1=5
r >^ "O
.E o «
^
w o) to
W C 3
is ts *^
o '5 «
J 0 _>>
in
(N
«^
o»
E
<
0.
UJ
So) 01 c
S. -o i- c —
T ™ ^ (u
? •= .2 5 is
Sample size is limited to 20 lite
pair of sorbent tubes. Sorbent
are susceptable to contaminat
from organics in ambient air d
installation and removal from
£u9
2 LU Q.
O 0 u
V3 (3 (3
TD a! ^
c -S u
10 5 '
«-is?
•^ > ^o
115
§ - S"
'•3 !2 3 i;
1 1 2 1 o» i
n C O S (O '0 *j
u 1^5 I .£ a)
Q.
"S- -A
*rt C j*
*" ^ i^
O ^^ g J3 g
V .£ J- K "S i
a> ^ i1 «i 5 2
** — O Q.5 w
$ .£ 1 S 5 .0
>/<
o
1
12-123
-------
o
II
o>2
0) — •**•
•s S-a
a " r
zf *- a;
^ 5 «•
41
a
;= 3
a o
a a
< E
o
u
a >.
IB t-
u 4,
i-ti
a 3
<$
73 *
111
O u. —
C . _>>
v/» .C 4;
SOI «•*
3 «
•o o E
a
-
a.o
IB •_
*2
V ^
•p
*/•» u
.0
U U U
U
3 O
•° 's |-
olS
U 41 >
a--
£ -a a
c .. <«
.= « u
5 3 a
"^ *" C
3 ^ C
a!
J 3
0.0-
S.S 8.
<0 TJ js
| ci CT
_o c c
5.11
z a 2
.Q
u u u
O 15 O
" ? IB
•5 ^ c
> O 0)
•- «f ?
EcS|
^ V^ *w
t/^ &. ^
c 2
av
o u"
I £
3 O
JD '3
o E
s
o
il
01
"5 « -o
> *: o
i ^ •£
01 O 4*
^
w t/S 4)
C t/1 S
O
•S c t U
. ^ •- O I
T3 rg »j
_ 41
9* Q. E •* >"
£ 01 ^ c- Q- *
8 E ~ o 2 S
.a 8 2 '5 3 S
HUM
^ W **- SJ -C nj
, Q
U U U
(j (J O
••- 0^ (X
.J. O Ot c
E^M* O ^
W —• ^J)
v/1 IB ^ O
Uf
«r o
I I S" I « i
llllll
= 9 =
O < fO
•= X ^
•S c 'c
* 10 .- ^
l*sl
a?
gas
trap.
nstruc
Sample
trap fol
sorben
steel c
C
2
V/1
•o
I
o
o
o
I *
2 w o>
E
O)
IB
C
1 E i 5-=
222«°
^M ^C ^H ' ^
-c =
u ••S
rfl >« .2-O
10 <0 O 3
01 en >-^
9^9^.5
^2a. £.Q
u u u O g
-------
an airtight, rigid container (lung) and evacuating the container. The sample is
drawn into the bag because reduced pressure in the container provides adequate
suction to fill the bag. This procedure is presented in detail in 40 CFR Part 60,
Appendix A (Method 3).
Evacuated canisters are conventionally constructed of high grade polished
stainless steel. There are many versions available ranging from units with torque
limiting needle valves, purge free assemblies, internal electropolished surfaces and
versions utilizing stainless steel beakers with custom designed tops and fittings.
Also, different container materials may react differently with the sample.
Therefore, sample storage time or sample recovery studies to determine or verify
inertness of the sampling canister should be considered.
Canisters are generally used to collect samples by slowly opening the sample
valve, allowing the vacuum to draw in the sample gas. In less than a minute, the
container should equilibrate with the ambient atmospheric pressure. At that time,
the sample valve is closed to retain the sample. To collect composite samples over
longer intervals, small calibrated orifices can be inserted before the inlet valve to
extend the time required for equilibration of pressure once the sample valve is
opened.
The sample collection procedure for EPA Method 5 (U.S. EPA, 1981) is similar in
principle to that for the evacuated canister. The train consists of a polished stainless
steel canister with a cold condensate trap in series and prior to the canister to collect
a higher boiling point organic fraction. This two fraction apparatus provides for
separate collection of two concentration ranges of volatile organic compounds
based on boiling point.
The following four sampling methods utilize sample concentration techniques
using one or more sorbent traps. The advantages of these methods is an enhanced
limit of detection for many toxic and hazardous organic compounds. These
techniques are preferred due to their lower detection limit. The Modified Method 5
(MM5) sampling train (U.S. EPA, 1981) is used to sample gaseous effluents for vapor-
phase organic compounds that exhibit vapor pressures of less than 2 mm Hg (at
20°C). This system is a modification of the conventional EPA Method 5 paniculate
sampling train. The modified system consists of a probe, a high efficiency glass or
12-125
-------
quartz fiber filter, a sorbent module, impingers, and related control hardware. The
sample gas is passed through a single sorbent trap, containing XAD-2. The M1V15
train is limited due to the single sorbent trap design that does not provide a backup
for breakthrough. This is especially important when large volumes of sample are
collected.
To minimize the potential for breakthrough, the MM5 train can be modified
to provide a backup trap. However, this dual trap modification increases the
pressure drop across the train, reducing the range of flow rates possible for sample
collection. To overcome this pressure drop and maintain the desired flow rate, the
high-volume MM5 train utilizes a much larger capacity pump.
The Source Assessment Sampling System (SASS) train is another comprehensive
sampling train, consisting of a probe that connects to three cyclones and a filter in a
heated oven module, a gas treatment section, and a series of impingers to provide
large collection capacities for paniculate matter, semivolatiles, and other lower
volatility organics. The materials of construction are all stainless steel making the
system very heavy and cumbersome. The stainless steel construction is also very
susceptible to corrosion. This system can, however, be used to collect and
concentrate large sample volumes, providing for a much lower detection limit.
Because of the sorbents used (generally XAD-2), its use is limited to the same class of
lower volatility organics and metals as the MM5 train.
The Volatile Organic Sampling Train (VOST) has proven to be a reliable and
accurate method for collection of the broad range of organic compounds. By using
a dual sorbent and dual in-series trap design, the VOST train can supplement either
the MM5 or SASS methods allowing for collection of more volatile species.
However, VOST has several limitations, including a maximum sample flow rate of
1.0 liter/minute, and a total sample volume of 20 liters per trap pair. Therefore,
frequent changes of the trap pairs are required for test periods that exceed 20
minutes. The frequent change of traps makes the samples more susceptible to
contamination.
Any of the point source monitoring techniques described above can be
adapted for use with the isolation flux chamber techniques described previously.
For point sources where paniculate emissions are of concern, the Modified Method
12-126
-------
5 or SASS train (originally designed to measure particle emissions from combustion
effluents) are also applicable and proven technologies.
Analytical methodologies for the techniques discussed above will vary with the
technique used. While certain techniques will offer advantages over others in the
measurement of specific contaminants, the investigator is advised to utilize
standard methodologies whenever possible in performing the RFI. For example, use
of the VOST and/or the MM5 train, and their associated analytical methodologies is
recommended for point source monitoring of the applicable compounds.
Descriptions for both of these methods are included in the 3rd Edition of "Test
Methods for Evaluating Solid Waste" (EPA SW-846), 1986 (GPO No. 955-001-00000-
1). Although these methods are designed for the evaluation of incinerator
efficiencies, they are essentially point-source monitoring methods which can be
adapted to most point sources.
12.8.3.2 Metals
Although the emission of metallic contaminants is primarily associated with
particulate emission from area sources caused by the transfer of material to and
from different locations, wind erosion, or general maintenance and traffic activities
at the unit, point source emission of particulate or vapor-phase metals can exist.
Metallic constituents may exist in the atmosphere as solid particulate matter, as
dissolved or suspended constituents of liquid droplets (mists), and as vapors.
Metals specified as hazardous constituents in 40 CFR Part 261, Appendix VIII
are generally noted as the element and compounds "not otherwise specified
(NOS)", as shown in Table 12-21, indicating that measurement of the total content
of that element in the sample is required.
Vapor phase metals-Forthe purpose of point-source monitoring, vapor-phase
metals will be defined as all elements or compounds thereof, that are not
quantitatively captured by standard filter sampling procedures. These include
volatile forms of metals such as elemental and alkyl mercury, arsine, antimony, alkyl
lead compounds, and nickel carybonyl.
12-127
-------
Table 12-21.
RCRA APPENDIX VIII HAZARDOUS METALS AND
METAL COMPOUNDS
Antimony and compounds
Arsenic and compounds NOSb
Barium and compounds NOSb
Beryllium and compounds NOS
Cadmium and compounds NOS
Chromium and compounds NOS
Lead and compounds NOS
Mercury and compounds NOSb
Nickel and compounds NOSb
Selenium and compounds NOSb
Silver and compounds NOSb
Thallium and compounds NOSb
a NOS = not otherwise specified.
b Additional specific compound(s) listed for this
element.
12-128
-------
The sampling of point sources for vapor phase metals has not been a common
or frequent activity for the investigation of air releases from solid waste
management units. If a point source of vapor-phase metals is identified, the
sampling approach should identify the best available monitoring techniques,
considering that many have been developed which are specific to single species
rather than multiple species of many different metal elements. The primary
references for identifying available techniques include National Institute of
Occupational Safety and Health (NIOSH, 1984) methods, EPA methods such as those
presented in SW-846 and in the Federal Register under the National Emissions
Standards for Hazardous Air Pollutants (NESHAPs), and American Public Health
Association (APHA, 1977) methods. The basic monitoring techniques include
collection on sorbents and in impinger solutions. The particular sorbent or impinger
solution utilized should be selected based on the specific metal species under
investigation.
Particulate Metals-Point-source releases to air could also require investigation
of particulate metals. Source sampling particulate procedures such as the Modified
Method 5 or SASS methods previously discussed are appropriate for this activity.
EPA Modified Method 5 is the recommended approach. Modification of this basic
technique involving the collection of particulate material on a filter with
subsequent analysis of the collected particulate materal on a filter for the metals of
concern, could include higher or lower flow rates and the use of alternate filter
media. Such modificaitons may be proposed when standard techniques prove to be
inadequate. Several important particulate metal sampling methods are available in
the NIOSH methods manuals (NIOSH, 1984); however, these methods were designed
for ambient or indoor applications and may require modification if used on point
sources.
12.9 Site Remediation
Although the RFI Guidance is not intended to provide detailed guidance on
site remediation, it should be recognized that certain data collection activities that
may be necessary for a Corrective Measures Study may be collected during the RFI.
EPA has developed a practical guide for assessing and remediating contaminated
sites that directs users toward technical support, potential data requirements and
12-129
-------
technologies that may be applicable to EPA programs such as RCRA and CERCLA.
The reference forthis guide is provided below.
U.S. EPA. 1988. Practical Guide for Assessing and Remediating Contaminated
Sites. Office of Solid Waste and Emergency Response. Washington, D.C.
20460.
The guide is designed to address releases to ground water as well as soil,
surface water and air. A short description of the guide is provided in Section 1.2
(Overall RCRA Corrective Action Process), under the discussion of Corrective
Measures Study.
12-130
-------
12.10 Checklist
RFI CHECKLIST-AIR
Site Name/Location
Type of Unit
1. Does waste characterization include the following information? (Y/N)
• Physical form of the waste
• Identification of waste components
• Concentrations of constituents of concern
• Chemical and physical properties of constituents
of concern
2. Does unit characterization include the following information? (Y/N)
• Type of unit
• Types and efficiencies of control devices
• Operational schedules
• Operating logs
• Dimensions of the unit
• Quantities of waste managed
• Locations and spatial distribution/
variation of waste in the unit
• Past odor complaints from neighbors
• Existing air monitoring data
• Flow rates from vents
3. Does environmental setting characterization include
the following information? (Y/N)
• Definition of regional climate
• Definiation of site-specific meteorological conditions
• Definition of soil conditions
12-131
-------
• Definition of site-specific terrain
• Identification of potential release receptors
4. Have the following data on the initial phase of the release
characterization been collected? (Y/N)
• Conceptual model of release developed
• Concentrations of released constituent at unit,
facility property boundary and, if appropriate,
at nearby offsite receptors (based on
screening assessment or available
modeling/monitoring data)
• Screening monitoring data (as warranted)
• Additional waste/unit data (as warranted)
5. Have the following data on the subsequent phase(s) of the
release characterization been collected? (Y/N)
• Identification of "reasonable worst case"
conditions
• Meteorological conditions during monitoring
• Release source conditions during monitoring
• Basis for selection of monitoring constituents
• Concentrations of released constituents at unit,
facility property boundary and, if appropriate,
at nearby offsite receptors (based on
monitoring or modeling and representative
of reasonable "worst case" conditions)
12-132
-------
12.11 References
ACGIH. 1983. Air Sampling Instruments for Evaluation of Atmospheric
Contamination. American Conference of Governmental Industrial Hygienists.
Washington, D.C.
APHA. 1977. Methods of Air Sampling and Analysis. American Public Health
Association. Cincinnati, OH.
ASTM. 1982. Toxic Materials in the Atmosphere. ASTM, STP 786. Philadelphia, PA.
ASTM. 1981. Toxic Materials in the Atmosphere. ASTM, STP 786. Philadelphia, PA.
ASTM. 1980. Sampling and Analysis of Toxic Orqanics in the Atmosphere. ASTM,
STP 721. Philadelphia, PA.
ASTM. 1974. Instrumentation for Monitoring Air Quality. ASTM, STP 555.
Philadelphia, PA.
National Climatic Data Center. Climates of the United States. Asheville, NC 28801.
National Climatic Data Center. Local Climatoloqical Data - Annual Summaries with
Comparative Data, published annually. Asheville, NC 28801.
National Climatic Data Center. Weather Atlas of the United States. Asheville,
NC 28801.
National Institute for Occupational Safety and Health (NIOSH). 1985. NIOSH
Manual of Analytical Methods. NTISPB85-179018.
Turner, D.B. 1969. Workbook of Atmospheric Dispersion Estimates. Public Health
Service. Cincinnati, OH.
U.S. EPA. December 1988 Draft. Procedures for Conducting Air Pathway Analyses
for Superfund Applications. Office of Air Quality Planning and Standards.
Research Triangle Park, NC 27711.
12-133
-------
U.S. EPA. March 1988 Draft. A Workbook of Screening Techniques for Assessing
Impacts of Toxic Air Pollutants. Office of Air Quality Planning and Standards.
Research Triangle Park, NC 27711.
U.S. EPA. June 1987. On-Site Meteorological Program Guidance for Regulatory
Modeling Applications. EPA-450/4-87-013. Office of Air Quality Planning and
Standards. Research Triangle Park, NC 27711.
>
U.S. EPA. December 1987. Hazardous Waste Treatment Storage and Disposal
Facilities (TSDF) Air Emission Models. EPA-450/3-87-026. Office of Air Quality
Planning and Standards. Research Triangle Park, NC 27711.
U.S. EPA. 1986. Evaluation of Control Technologies for Hazardous Air Pollutants:
Volume 1-Technical Report. EPA/60077-86/009a. NTIS PB 86-167020. Volume
2 - Appendices. EPA/600/7-86/009b. NTIS PB 86-167038. Office of Research and
Development. Research Triangle Park, NC 27711.
U.S. EPA. September 1986. Handbook-Control Technologies for Hazardous Air
Pollutants. EPA/625/6-86/014. Office of Research and Development. Research
Triangle Park, NC 27711.
U.S. EPA. February 1986. Measurement of Gaseous Emission Rates from Land
Surfaces Using an Emission Isolation Flux Chamber: User's Guide. 1986.
EPA/600/8-86/008. NTIS PB 86-223161. Environmental Monitoring Systems
Laboratory. Las Vegas, NV 89114.
U.S. EPA. 1986. Test Methods for Evaluating Solid Waste. 3rd Edition. EPA SW-846.
GPO No. 955-001-00000-1. Washington, D.C 20460.
U.S. EPA. July 1986. Guideline on Air Quality Models (Revised). EPA-450/2-78-027R.
NTIS PB 86-245248. Office of Air Quality Planning and Standards. Research
Triangle Park, NC 27711.
U.S. EPA. June 1986. Industrial Source Complex (ISC) Model User's Guide-Second
12-134
-------
Edition. EPA-450/4-86-005a and b. Office of Air Quality Planning and
Standards. Research Triangle Park, NC 27711.
U.S. EPA. November 1985. Practical Guide - Trial Burns for Hazardous Waste
Incinerators. NTIS PB 86-190246. Office of Air Quality Planning and Standards.
Research Triangle Park, NC 27711.
U.S. EPA. February 1985. Rapid Assessment of Exposure to Paniculate Emissions
from Surface Contamination Sites. EPA/600/8-85/002. NTIS PB 85-192219.
Office of Health and Environmental Assessment. Washington, D.C. 20460.
U.S. EPA. February 1985 (Fourth Edition and subsequent supplements). Modified
Method 5 Train and Source Assessment Sampling System Operators Manual.
EPA/600/8-85/003. NTIS PB 85-169878. Office of Research and Development.
Research Triangle Park, NC 27711.
U.S. EPA. 1985. Compilation of Air Pollutant Emission Factors. EPAAP-42. NTIS PB
86-124906. Office of Air Quality Planning and Standards. Research Triangle
Park.NC 27711.
U.S. EPA. 1984. Evaluation and Selection of Models for Estimating Air Emissons
from Hazardous Waste Treatment. Storage, and Disposal Facilities. EPA-450/3-
84-020. NTIS PB 85-156115. Office of Air Quality Planning and Standards.
Research Triangle Park, NC 27711.
U.S. EPA. September 1984. Network Design and Site Exposure Criteria for Selected
Noncriteria Air Pollutants. EPA-450/4-84-022. Office of Air Quality Planning
and Standards. Research Triangle Park, NC 27711.
U.S. EPA. June 1984. Evaluation of Air Emissions from Hazardous Waste
Treatment. Storage and Disposal Facilities. EPA 600/2-85/057. NTIS PB 85-
203792. Office of Research and Development. Cincinnati, OH 45268.
U.S. EPA. April 1984. Compendium of Methods for the Determination of Toxic
Organic Compounds in Ambient Air. EPA-600/4-84-041. Office of Research
and Development. Research Triangle Park, NC 27711.
12-135
-------
U.S. EPA. March 1984. Protocol for the Collection and Analysis of Volatile POHCs
Using VQST. EPA-600/8-84-007. NTIS PB 84-170042. Office of Research and
Development. Research Triangle Park, NC 27711.
U.S. EPA. February 1984. Sampling and Analysis Methods for Hazardous Waste
Combustion. EPA-600/8-84-002. NTIS PB 84-155845. Washington, D.C. 20460.
U.S. EPA. September 1983. Characterization of Hazardous Waste Sites - A Methods
Manual: Volume II. Available Sampling Methods. EPA-600/4-83-040. NTIS PB
83-014799. Off ice of Sol id Waste. Washington, D.C. 20460.
U.S. EPA. July 1983. Guidance Manual for Hazardous Waste Incinerator Permits.
NTIS PB 84-100577. Office of Solid Waste. Washington, D.C. 20460.
U.S. EPA. June 1983. Technical Assistance Document for Sampling and Analysis of
Toxic Organic Compounds in Ambient Air. EPA-600/4-83-027. NTIS PB 83-
239020. Office of Research and Development. Research Triangle Park, NC
27711.
U.S. EPA. February 1983. Quality Assurance Handbook for Air Pollution
Measurement Systems: Volume IV. Meteorological Measurement. February
1983. EPA-600-4-82-060. Office of Research and Development. Research
Triangle Park, NC 27711.
U.S. EPA. November 1980. Ambient Monitoring Guidelines for Prevention of
Significant Deterioration (PSD). EPA-450/4-80-012. NTIS PB 81-153231. Office
of Air Quality Planning and Standards. Research Triangle Park, NC 27711.
U.S. EPA. 1978. Stack Sampling Technical Information. A Collection of Monographs
and Papers. Volumes Nil. EPA-450/2-78-042a,b,c. NTIS PB 80-161672, 80-
161680,80-161698.
U.S. EPA. October 1977. Guidelines for Air Quality Maintenance Planning and
12-136
-------
Analysis. Volume 10 (Revised): Procedures for Evaluating Air Quality Impact of
New Stationary Sources. EPA-450/4-77-001. NTIS PB 274087/661. Office of Air
Quality Planning and Standards. Research Triangle Park, NC 27711.
U.S. EPA. Code of Federal Regulations. 40CFRPart60: Appendix A: Reference
Methods. Office of Federal Register. Washington, D.C.
U.S. EPA. November 1981. Source Sampling and Analysis of Gaseous Pollutants.
EPA-APTI Course Manual 468. Air Pollution Control Institute. Research
Triangle Park, NC 27711.
U.S. EPA. 1979. Source Sampling for Particulate Pollutants. EPA-APTI Course
Manual 450. NTIS PB 80-182439, 80-174360. Air Pollution Control Institute.
Research Triangle Park, NC 27711.
12-137
-------
SECTION 13
SURFACE WATER
13.1 Overview
The objective of an invest.gation of a release to surface water is to
characterize the nature, extent, and rate of migration of the release to this medium.
This section provides the following:
• An example strategy for characterizing releases to the surface water
system (e.g., water column, bottom sediments, and biota), which includes
characterization of the source and the environmental setting of the
release, and conducting a monitoring program that will characterize the
release;
• A discussion of waste and unit source characteristics and operative
release mechanisms;
• A strategy for the design and conduct of monitoring programs
considering specific requirements of different wastes, release
characteristics, and receiving water bodies;
• Formats for data organization and presentation;
• Appropriate field and other methods that may be used in the
investigation; and
• A checklist of information that may be needed for release
characterization.
The exact type and amount of information required for sufficient release
characterization will be facility and site-specific and should be determined through
interactions between the regulatory agency and the facility owner or operator
during the RFI process. This guidance does not define the specific data needed in all
13-1
-------
instances; however, it identifies the information that is likely to be needed to
perform release characterizations and identifies methods for obtaining this
information. The RFI Checklist, presented at the end of this section, provides a tool
for planning and tracking information collection for release characterization. This
list is not a list of requirements for all releases to surface water. Some releases will
involve the collection of only a subset of the items listed, while others will involve
the collection of additional data.
Case Study Numbers 27, 28, 29, 30 and 31 in Volume IV (Case Study Examples)
illustrate various aspects of surface water investigations which are described below.
13.2 Approach for Characterizing Releases to Surface Water
13.2.1 General Approach
A conceptual model of the release should be formulated using all available
information on the waste, unit characteristics, environmental setting, and any
existing monitoring data. This model (not a computer or numerical simulation
model) should provide a working hypothesis of the release mechanism, transport
pathway/mechanism, and exposure route (if any). The model should be
testable/verifiable and flexible enough to be modified as new data become
available. For surface water investigations, this model should account for the
release mechanism (e.g., overtopping of an impoundment), the nature of the source
area (e.g., point or non-point), waste type and degradability, climatic factors (e.g.,
history of floods), hydrologic factors (e.g., stream flow conditions), and fate and
transport factors (e.g., ability for a contaminant to accumulate in stream bottom
sediments). The conceptual model should also address the potential for the transfer
of contaminants in surface water to other environmental media (e.g., soil
contamination as a result of flooding of a contaminated creek on the facility
property).
An example strategy for characterization of releases to surface waters is
summarized in Table 13-1. These steps outline a phased approach, beginning with
evaluation of existing data and proceeding to design and implementation of a
monitoring program, revised over time, as necessary, based on findings of the
previous phase. Each of these steps is discussed briefly below.
13-2
-------
TABLE 13-1
EXAMPLE STRATEGY FOR CHARACTERIZING RELEASES TO SURFACE WATER*
INITIAL PHASE
1. Collect and review existing information on:
Waste
Unit
Environmental setting
Contaminant releases, including inter-media transport
2. Identify any additional information necessary to fully characterize release:
Waste
Unit
Environmental setting
Contaminant releases, including inter-media transport
3. Develop monitoring procedures:
Formulate conceptual model of release
Determine monitoring program objectives
Select monitoring constituents and indicator parameters
Select monitoring locations
Determine monitoring frequency
Incorporate hydrologic monitoring as necessary
Determine role of biomonitoring and sediment monitoring
4. Conduct initial monitoring:
Collect samples under initial monitoring phase procedures and complete
field analyses
Analyze samples for selected parameters and constituents
5. Collect, evaluate, and report results:
Compare analytical and other monitoring procedure results to health
and environmental criteria and identify and respond to emergency
situations and identify priority situations that may warrant interim
corrective measures - Notify regulatory agency
Summarize and present data in appropriate format
Determine if monitoring program objectives were met
Determine if monitoring locations, constituents and frequency were
adequate to characterize release (nature, extent, and rate)
Report results to regulatory agency
13-3
-------
TABLE 13-1 (continued)
EXAMPLE STRATEGY FOR CHARACTERIZING RELEASES TO SURFACE WATER*
SUBSEQUENT PHASES (If necessary)
1. Identify additional information necessary to characterize release:
Identify additional information needs
Determine need to include or expand hydrologic, and sediment and bio-
monitoring
Evaluate potential role of inter-media transport
2. Expand initial monitoring as necessary:
Relocate, decrease, or increase number of monitoring locations
Add or delete constituents and parameters of concern
Increase or decrease monitoring frequency
Delete, expand, or include hydrologic, sediment or bio-monitoring
3. Conduct subsequent monitoring phases:
Collect samples under revised monitoring procedures and complete field
analyses
Analyze samples for selected parameters and constituents
4. Collect, evaluate and report results/identify additional information necessary
to characterize release:
Compare analytical and other monitoring procedure results to health
and environmental criteria and identify and respond to emergency
situations and identify priority situations that may warrant interim
corrective measures - Notify regulatory agency
Determine if monitoring program objectives were met
Determine if monitoring locations, constituents, and frequency were
adequate to characterize release (nature, extent, and rate)
Identify additional information needs
Determine need to include or expand hydrologic, sediment, or bio-
monitoring
Evaluate potential role of inter-media transport
Report results to regulatory agency
Surface water system is subject to inter-media transport. Monitoring program
should incorporate the necessary procedures to characterize the relationship,
if any, with ground water, sediment deposition, fugitive dust and other
potential release migration pathways.
13-4
-------
The first step in the general approach is the collection and review of available
information on the contaminant source and the environmental setting. Some
information on the contaminant source will be available from several reports and
other documents. The RCRA permit, compliance order, or RFA report will provide a
summary of information regarding actual or suspected releases from the various
units. The facility owner or operator should be familiar with this information as a
basis for further characterization of the release(s) in the RFI. In addition, a
thorough understanding of the environmental setting is essential to an adequate
determination of the nature and extent of releases to surface waters. Monitoring
data should also be reviewed focusing on the quality of the data. If the quality
is determined to 36 acceptable, then the data may be used in the design of
the monitoring program. Guidance on obtaining and evaluating the necessary
information on the contaminant source and the environmental setting is given in
Section 13.3.
During the initial investigation particular attention should be given to
sampling run-off from contaminated areas, leachate seeps and other similar sources
of surface water contamination, as these are the primary overland release pathways
for surface water. Releases to surface water via ground-water discharge should be
addressed as part of the ground-water investigation, which should be coordinated
with surface water investigations, for greater efficiency.
Based on the collection and review of existing information, the design of the
monitoring program is the next major step in the general approach. The
monitoring program should include clear objectives, monitoring constituents and
indicator parameters, monitoring locations, frequency of monitoring, and
provisions for hydrologic monitoring. In addition to conventional water quality and
hydrologic monitoring, sediment monitoring and biomonitoring may also have a
role in the surface water evaluation for a given RFI. Guidance on the design of the
monitoring program is given in Section 13.4.
Implementation of the monitoring program is the next major step in the
general strategy for characterizing releases to surface water. The program may be
implemented in a phased manner that allows for modifications to the program in
subsequent phases. For example, initial monitoring results may indicate that
13-5
-------
downstream monitoring locations have been placed either too close to or too far
from the contaminant source to accurately define the complete extent of
downstream contamination. In this case, the program should be modified to
relocate monitoring stations for subsequent monitoring phases. Similarly, initial
monitoring may indicate that biomonitoring of aquatic organisms is needed in the
next phase. Guidance on methods that can be used in the implementation of the
program is given in Section 13.6.
Finally, the results of the characterization of releases to surface waters must be
evaluated and presented in conformance with the requirements of the RFI. Section
13.5 provides guidance on data presentation. Table 13-2 summarizes techniques
and data-presentation methods for the key characterization tasks.
As monitoring data become available, both within and at the conclusion of
discrete investigation phases, they should be reported to the regulatory agency as
directed. The regulatory agency will compare the monitoring data to applicable
health and environmental criteria to determine the need for (1) interim corrective
measures; and/or (2) a Corrective Measures Study. In addition, the regulatory
agency will evaluate the monitoring data with respect to adequacy and
completeness to determine the need for any additional monitoring efforts. The
health and environmental criteria and a general discussion of how the regulatory
agency will apply them are supplied in Section 8. A flow diagram illustrating RFI
decision points is provided in Section 3 (See Figure 3-2).
Notwithstanding the above process, the owner or operator has a continuing
responsibility to identify and respond to emergency situations and to define priority
situations that may warrant interim corrective measures. For these situations, the
owner or operator is directed to follow the RCRA Contingency Plan requirements
under 40 CFR Part 264, Subpart D and Part 265, Sub part D.
13-6
-------
TABLE 13-2
RELEASE CHARACTERIZATION TASKS FOR SURFACE WATER
Investigatory Tasks
Investigatory Techniques
Data Presentation
Formats/Outputs
1. Waste/Unit
Characterization
- Waste Composition and
Analysis
- Unit or Facility
Operations
Release Mechanisms
See Section 13.3.1
Review waste handling and
disposal practices and
schedules
Review environmental
control strategies
See Section 13.3.1, Review
operational information
Data Tables
Schematic diagrams of flow
paths, narrative
Site-specific diagrams,
maps, narrative
2. Environmental Setting
Characterization
- Geographic Description
Classification of Surface
Water and Receptors
Define Hydrologic
Factors
Review topographic, soil
and geologic setting
information
See Section 13.3.3.1
See Section 13.3.3.1
Maps, Tables, Narrative
Maps, Cross Sections,
Narrative
Tables, Graphs, Map
3. Release Characterization
- Delineate Areal Extent
of Contamination
Define Distribution
Between Sediment,
Biota and Water
Column
Determine Rate of
Migration
Describe Seasonal
Effects
Sampling and Analysis
Sampling and Analysis
Flow Monitoring
Repetitive Monitoring
Tables of Results, Contour
Maps, Maps of Sampling
Locations
Graphs and Tables
Graphs and Tables
Graphs and Tables
13-7
-------
13.2.2 Inter-media Transport
Surface waters are subject to inter-media transport, both as a receptor of
contamination and as a migration pathway. For example, surface waters are
generally engaged in a continual dynamic relationship with ground water. Ground
water may discharge to a surface water body that may, in turn, recharge an aquifer.
Hence, contamination may be transported from ground water to surface water and
from surface water to ground water. Release of contaminants from a receiving
water body to soil can also occur through deposition of the contaminants in
floodplain sediments. These sediments may be exposed to wind erosion and
become distributed through fugitive dust. Sediments may be exposed to air during
periods of low flow of water in streams and lakes and when sediments are
deposited by overland flow during rainfall-runoff events. Contaminants may also
enter the air from surface water through volatilization.
13.3 Characterization of the Contaminant Source and Environmental Setting
The initial step in developing an effective monitoring program for a release to
surface waters is to investigate the unit(s) that is the subject of the RFI, the waste
within the unit(s), the constituents within the waste, the operative release
mechanisms and migration pathways to surface water bodies, and the surface water
receptors. From this information, a conceptual model of the release can be
developed for use in designing a monitoring program to characterize the release.
13.3.1 Waste Characterization
Knowledge of the general types of wastes involved is an important
consideration in the development of an effective monitoring program. The
chemical and physical properties of a waste and the waste constituents are major
factors in determining the likelihood that a substance will be released. These waste
properties may also be important initially in selecting monitoring constituents and
indicator parameters. Furthermore, once the wastes are released, these properties
play a major role in controlling the constituent's migration through the
environment and its fate. Table 13-3 lists some of the significant properties in
evaluating environmental fate and transport in a surface water system. Without
data on the wastes, the investigator may have to implement a sampling program
13-8
-------
TABLE13-3
IMPORTANT WASTE AND CONSTITUENT PROPERTIES
AFFECTING FATE AND TRANSPORT IN A SURFACE WATER ENVIRONMENT
Bulk waste properties affecting mobility*
• Physical state (solid, liquid, gas) of waste
• Chemical nature (e.g., aqueous vs non-aqueous) of waste
• Density (liquid)
• Viscosity (liquid)
• Interfacial tension (with water and minerals) (liquid)
Properties to assess mobility of constituents^
Solubility
Vapor pressure
Henry's law constant (or vapor pressure and water solubility)
Bioconcentration factor
Soil adsorption coefficient
Diffusion coefficient (in air and water)
Acid dissociation constant
Octanol-water partition coefficient
Activity coefficient
Mass transfer coefficients (and/or rate constants) for intermedia transfer
Boiling point
Melting point
Properties to assess persistence^
• Rate of biodegradation (aerobic and anaerobic)
• Rate of hydrolysis
• Rate of oxidation or reduction
• Rate of photolysis
a These waste properties will be important when it is known or suspected that
the waste itself has migrated into the environment (e.g., due to a spill).
b These properties are important in assessing the mobility of constituents
present in low concentrations in the environment.
c For these properties, it is generally important to know (1) the effects of key
parameters on the rate constants (e.g., temperature, concentration, pH) and
(2) the identity of the reaction products.
Sources of values for these and other parameters include Mabey, Smith, and Podall,
(1982), and Callahan, et al. (1979). Parameter estimation methods are described by
Lyman, Riehl, and Rosenblatt, (1982), and Neely and Blau (1985).
13-9
-------
involving many constituents to ensure that all potential constituents have been
addressed. General guidance on defining physical and chemical properties and
identifying possible monitoring constituents and indicator parameters is provided
in Sections 3 and 7.
Below are brief synopses of several of the key release, mobility, and fate
parameters summarized in Table 13-3. Figure 13-1 shows the qualitative
relationship between various environmental partitioning parameters. Neely and
Blau (1985) provide a description of environmental partitioning effects of
constituents and application of partition coefficients.
• Physical State:
Solid wastes would appear to be less susceptible to release and migration
than liquids. However, processes such as dissolution (i.e., as a result of
leaching or runoff), and physical transport of waste particulates can act
as significant release mechanisms.
• Water Solubility:
Solubility is an important factor affecting a constituent's release and
subsequent migration and fate in the surface water environment. Highly
soluble contaminants (e.g., methanol at 4.4 x 106 mg/L at 77oF) are easily
and quickly distributed within the hydrologic cycle. These contaminants
tend to have relatively low adsorption coefficients for soils and
sediments and relatively low bioconcentration factors in aquatic life. An
example of a less soluble constituent is tetrachloroethylene at 100 mg/L
at 77oF.
• Henry's Law Constant:
Henry's Law Constant indicates the relative tendency of a constituent to
volatilize from aqueous solution to the atmosphere based on the
competition between its vapor pressure and water solubility.
Contaminants with low Henry's Law Constant values (e.g., methanol,
1.10x10-6 atm-m3/mole at 77oF)will tend to favor the aqueous phase
13-10
-------
w
O
I/I
z
z
o
t/1
,_ o
5 1
u >
« 2
u
CO
U
00
a.
x
t/i
O
5
f=
O
o
c
«
'w
V O
O w
w c
C w O
•2 3 '«
«* u Q.
t 42 o
a s -s
"• P IB
i«V
u
03
^H
^^
Solubilii
5
7
Octanol
i
c
01
Bioconc
uJ
u
c
tat
Soil/sed
3
1
13-11
-------
and volatilize to the atmosphere mere slowly than constituents with high
values (e.g., carbon tetrachloride, 2.3 x 10-2 atm-m3/mole at 77°F). This
parameter is important in determining the potential for inter-media
transport to the air media.
• Octanol/Water Partition Coefficient (Kow):
The octanol/water partition coefficient (Kow) is defined as the ratio of an
organic constituent's concentration in the octanol phase (organic) to its
concentration in the aqueous phase in a two-phase octanol/water
system. Values of K0w carry no units. Kow can be used to predict the
magnitude of an organic constituent's tendency to partition between
the aqueous and organic phases of a two phase system such as surface
water and aquatic organisms. The higher the value of K0w, the greater
the tendency of an organic constituent to adsorb to soil or waste
matrices containing appreciable organic carbon or to accumulate in
biota. Generally, constituents with Kow values greater than or equal to
2.3 are considered potentially bioaccumulative (Veith, et al., 1980).
• Soil-Water Partition Coefficient (K
-------
bioaccumulation, and therefore to determine whether sampling of the
biota may be necessary. Another source of BCFs for constituents is
contained in EPA's Ambient Water Quality Criteria (for priority
pollutants). BCFs can also be predicted by structure-activity relationships.
Constituents exhibiting a BCF greater than 1.0 are potentially
bioaccumulative. Generally, constituents exhibiting a BCF greater than
100 cause the greatest concern.
• The Organic Carbon Adsorption Coefficient (K0c):
The extent to which an organic constituent partitions between the solid
and solution phases of a saturated or unsaturated soil, or between runoff
water and sediment, is determined by the physical and chemical
properties of both the constituent and the soil (or sediment). The
tendency of a constituent to be adsorbed to soil is dependent on its
properties and on the organic carbon content of the soil or sediment. Koc
is the ratio of the amount of constituent adsorbed per unit weight of
organic carbon in the soil or sediment to the concentration of the
constituent in aqueous solution at equilibrium. Koc can be used to
determine the partitioning of a constituent between the water column
and the sediment. When constituents have L high K0c, they have a
tendency to partition to the soil or sediment. In such cases, sediment
sampling would be appropriate.
• Other Equilibrium Constants:
Equilibrium constants are important predictors of a compound's chemical
state in solution. In general, a constituent which is dissociated (ionized)
in solution will be more soluble and therefore more likely to be released
to the environment and more likely to migrate in a surface water body.
Many inorganic constituents, such as heavy metals and mineral acids, can
occur as different ionized species depending on pH. Organic acids, such
as the phenolic compounds, exhibit similar behavior. It should also be
noted that ionic metallic species present in the release may have a
tendency to bind to particulate matter, if present in a surface water
body, and settle out to the sediment over time and distance. Metallic
species also generally exhibit bioaccumulative properties. When metallic
13-13
-------
species are present in a release, both sediment and biota sampling would
be appropriate.
• Biodegradation:
Biodegradation results from the enzyme-catalyzed transformation of
organic constituents, primarily from microorganisms. The ultimate fate
of a constituent introduced into a surface water or other environmental
system (e.g., soil), could be a constituent or compound other than the
species originally released. Biodegradation potential should therefore
be considered in designing monitoring programs. Section 9.3 (Soils)
presents additional information on biodegradation.
• Photolysis:
Photodegradation or photolysis of constituents dissolved in aquatic
systems can also occur. Similar to biodegradation, photolysis may cause
the ultimate fate of a constituent introduced into a surface water or
other environmental system (e.g., soil) to be different from the
constituent originally released. Hence, photodegradation potential
should also be considered in designing sampling and analysis programs.
• Chemical Degradation (Hydrolysisand Oxidation/Reduction):
Similar to photodegradation and biodegradation, chemical degradation,
primarily through hydrolysis and oxidation/reduction (REDOX) reactions,
can also act to change constituent species once they are introduced to
the environment. Hydrolysis of organic compounds usually results in the
introduction of a hydroxyl group (-OH) into a chemical structure.
Hydrated metal ions, particularly those with a valence of 3 or more, tend
to form ions in aqueous solution, thereby enhancing species solubility.
Mabey and Mill (1978) provide a critical review of the hydrolysis of
organic compounds in water under environmental conditions. Stumm
and Morgan (1982) discuss the hydrolysis of metals in aqueous systems.
Oxidation may occur as a result of oxidants being formed during
photochemical processes in natural waters. Similarly, in some surface
water environments (primarily those with low oxygen levels) reduction
of constituents may take place.
13-14
-------
Degradation, whether biological, physical or chemical, is often reported in the
literature as a half-life, which is usually measured in days. It is usually expressed as
the time it takes for one half of a given quantity of a compound to be degraded.
Long half-lives (e.g., greater than a month or a year) are characteristic of persistent
constituents. It should be noted that actual half-life can vary significantly over
reported values based on site-specific conditions. For example, the absence of
certain microorganisms at a site, or the number of microorganisms, can influence
the rate of biodegradation, and therefore, half-life. Other conditions (e.g.,
temperature) may also affect degradation and change the half-life. As such, half-
life values should be used only as general indications of a chemical's persistence.
In addition to the above, reactions between constituents present in a release
may also occur. The owner or operator should be aware of potential
transformation processes, based on the constituents' physical, chemical and
biological properties, and account for such transformations in the design of
monitoring procedures and in the selection of analytical methods.
Table 13-4 provides an application of the concepts discussed above in assessing
the behavior of waste material with respect to release, migration, and fate. The
table gives general qualitative descriptors of the significance of some of the me -e
important properties and environmental processes for the major classes of organic
compounds likely to be encountered.
Table 13-4 can be used to illustrate several important relationships.
*
• Generally, water solubility varies inversely with sorption,
bioconcentration, and to a lesser extent, volatilization.
• Oxidation is a significant fate process for some classes of constituents
which can volatilize from the aqueous phase.
• Variations in properties and environmental processes occur within classes
as indicated by the pesticides, monocyclic aromatics, polycyclic aromatics,
and the nitrosamines and other nitrogen-containing compounds.
13-15
-------
^ z
S2
(/I H
"1
Is
II
°9
£ £
Z O
5 UL
go
a
s,
"o
-5^
c
g
~
•5
X
O
«
2
0.
c
g
2
•^
^
o
c
g
•9
C
2
c
o
o
a
0
a
o
^
-^
.5
^
Q
Chemical Class
Ol
x
^ fl/
2 2
« "2 "2
* o o
355
41
fl
9^
> -C "O
S Ol O
3x5
31 4)
2 2
9) 91
•o -c -o
O Ol O
5x5
01 5 |
x 3 3
91
**
W
f | |
x -i 5
91 41
"» «J
-c -o -S
III
41 41
19 "•
ill
^
41
ides
Organochlormes
Organophosphat
Carbamates
w*
o-
1
|
|
41
^0
J
i
Ol
x
Ol
X
1
ilormated Biphenyls
^m
VJ
f
o
m
Ol
X
1
Ol
X
§
_J
1
0»
-"
I!
.c
0)
X
i
2
91
O Ol
S x
,
41 C
5* p
2 19
amines and other Nit
ining Compounds
Benzedme
Di-n-propylnitros
S 2
I!
J2
w
£
w
111
C
'£
•a
a*
«»
01
(/«
.a
o
f
I/I
o
1)
(0
w
e
a»
-5 «
w C
C O
fl "O
O fl
5 €
2 = 2
x O =
o 'P
-------
Characterizing the environmental processes and properties of inorganic waste
constituents takes a similar approach to that shown on Table 13-4 for organics.
However, characterizing the metals on a class-by-class basis is not advisable because
of the complex nature of each metal and the many species in which the metals
generally occur. The interaction of each metal species with the surface water
environment is generally a function of many parameters including pH, REDOX
potential, and ionic strength. See Stumm and Morgan (1982) for additional
discussions on this subject. Generally, however, when metal species are present in a
release, it is advisable to monitor the sediment and biota, in addition to the water
column. This is due to likely deposition of metals as particulate matter, and to
potential bioaccumulation.
13.3.2 Unit Characterization
The relationship between unit characteristics and migration pathways
provides the framework in this section for a general discussion of release
mechanisms from units of concern to surface waters.
13.3.2.1 Unit Characteristics
Information on design and operating characteristics of a unit can be helpful in
characterizing a release. Unsound unit design and operating practices can allow
waste to migrate from a unit and possibly mix with runoff. Examples include
surface impoundments with insufficient freeboard, allowing for periodic
overtopping; leaking tanks or containers; or land-based units above shallow, low-
permeability materials which, if not properly designed and operated, can fill with
water and spill over. In addition, precipitation falling on exposed wastes can
dissolve and thereby mobilize hazardous constituents. For example, at uncapped
active or inactive waste piles and landfills, precipitation and leachate are likely to
mix at the toe of the active face or the low point of the trench floor. Runoff may
then flow into surface water through drainage pathways.
13-17
-------
13.3.2.2 Frequency of Release
Releases to surface waters may be intermittent, continuous, or a past
occurrence. It is important to consider the anticipated frequency of a release to
establish an effective monitoring program.
Most direct releases to surface waters are intermittent. Intermittent discharges
may be periodic, but may occur more often in a non-periodic manner, for example,
in response to rainfall runoff. Other common factors affecting intermittent releases
include fluctuations in water levels and flow rates, seasonal conditions (e.g., snow
melt), factors affecting mass stability (e.g., waste pile mass migration), basin
configuration, quantity/quality of vegetation, engineering control practices,
integrity of the unit, and process activities.
Erosion of contaminated materials from a unit (e.g., a landfill) is generally
intermittent, and is generally associated with rainfall-runoff events. Similarly,
breaches in a dike are generally short-term occurrences when they are quickly
corrected following discovery. Leaks, while still predominantly intermittent in
nature, may occur over longer spans of time and are dependent on the rate of
release and the quantity of material available.
Direct placement of wastes within surface waters (e.g., due to movement of an
unstable waste pile) has the potential to continuously contribute waste constituents
until the wastes have been removed or the waste constituents exhausted. Direct
placement is usually easily documented by physical presence of wastes within the
surface water body.
The frequency of sample collection should be considered in the design of the
monitoring program. For example, intermittent releases not associated with
precipitation runoff may require more frequent or even continuous sample
collection to obtain representative data on the receiving water body. Continuous
monitoring is generally feasible only for the limited number of constituents and
indicator parameters for which reliable automatic sampling/recording equipment is
available. Intermittent releases that are associated with precipitation runoff may
require event sample collection. With event sampling, water level or flow-activated
automatic sampling/recording equipment can be used. For continuous releases, less
13-18
-------
frequent sample collection is generally adequate to obtain representative data on
the receiving water body.
Previous intermittent releases may be identified through the analysis of
bottom sediments, and whole body or tissue analyses of relatively sessile and long-
lived macroinvertebrates (e.g., clams), or other species, such as fish. These analyses
may identify constituents that may have adsorbed onto particulates and settled to
the sediment, as well as bioaccumulative contaminants. In addition, intermittent
releases may be detected through the use of in situ bioassays. Using these
procedures, the test specie(s) is held within the effluent or stream flow and
periodically checked for survival and condition.
13.3.2.3 Form of Release
Releases to surface waters may be generally categorized as point sources or
non-point sources. Point sources are those that enter the receiving water at a
definable location, such as piped discharges. Non-point source discharges are all
other discharges, and generally cover large areas.
In general, most unit releases to surface waters are likely to be of a point
source nature. Most spills, leaks, seeps, overtopping episodes, and breaches occur
within an area which can be easily defined. Even erosion of contaminated soil and
subsequent deposition to surface water can usually be identified in terms of point
of introduction to the surface water body, through the use of information on
drainage patterns, for example. However, the potential for both point and non-
point sources should be recognized, as monitoring programs designed to
characterize these types of releases can be different. For example, the generally
larger and sometimes unknown areal extent of non-point source discharges may
require an increase in the number of monitoring locations from that routinely
required for point source discharges. The number of monitoring locations must be
carefully chosen to ensure representative monitoring results.
13.3.3 Characterization of the Environmental Setting
The environmental setting includes the surface water bodies and the physical
and biological environment. This section provides a general classification scheme
for surface waters and discusses collection of hydrologic data that may be important
in their characterization. Collection of specific geographical and climatological
13-19
-------
data are also discussed. Characterization of the biotic environment is treated in
Section 13.4.
Note that individual states have developed water quality standards for surface
waters pursuant to the Clean Water Act. These standards identify the designated
uses (e.g., drinking, recreation, etc.) of a surface water and a maximum
contaminant level to support the use. If applicable, the owner or operator should
report such standards.
13.3.3.1 Characterization of Surface Waters
Surface waters can be classified into one of the following categories. These
are obviously not pure classifications; intergrades are common.
• Streams and rivers;
• Lakes and impoundments;
• Wetlands; and
• Marine environments.
13.3.3.1.1 Streams and Rivers
Streams and rivers are conduits of surface water flow having defined beds and
banks. The physical characteristics of streams and rivers greatly influence their
reaction to contaminant releases and natural purification (i.e., assimilative
capacity). An understanding of the nature of these influences is important to
effective planning and execution of a monitoring program. Important
characteristics include depth, velocity, turbulence, slope, changes in direction and in
cross sections, and the nature of the bottom.
The effects of some of these factors are so interrelated that it is difficult to
assign greater or lesser importance to them. For example, slope and roughness of
the channel influence depth and velocity of flow, which together control
turbulence. Turbulence, in turn, affects rates of contaminant dispersion,
13-20
-------
reaeration, sedimentation, and rates of natural purification. The nature of
contaminant dispersion is especially critical in the location of monitoring stations.
All these factors may be of greater or lesser importance for specific sites. It should
also be noted that these factors may differ at the same site depending on when the
release occurred. For example, differences between winter and summer flow may
greatly influence the nature of contaminant dispersion.
Of further relevance to a surface water investigation are the distinctions
between ephemeral, intermittent, and perennial streams, defined as follows:
• Ephemeral streams are those that flow only in response to precipitation
in the immediate watershed or in response to snow melt. The channel
bottom of an ephemeral stream is always above the local water table.
• Intermittent streams are those that usually drain watersheds of at least
one square mile and/or receive some of their flow from baseflow
recharge from ground water during at least part of the year, but do not
flow continually.
• Perennial streams flow throughout the year in response to ground water
discharge and/or surface water runoff.
The distinction between ephemeral, intermittent and perennial streams will
also influence the selection of monitoring frequency, monitoring locations and
possibly other monitoring program design factors. For example, the frequency of
monitoring for ephemeral streams, and to a lesser extent intermittent streams, will
depend on rainfall runoff. For perennial-stream monitoring, the role of rainfall
runoff in monitoring frequency may be of less importance under similar release
situations.
The location of ephemeral and intermittent streams may not be apparent to
the owner or operator during periods of little or no precipitation. Generally,
intermittent and ephemeral streams may be associated with topographic
depressions in which surface water runoff is conveyed to receiving waters. In
addition to topography, a high density of vegetation in such areas may be an
indicator of the presence of ephemeral or intermittent drainage.
13-21
-------
Perennial streams and rivers are continually engaged in a dynamic relationship"
with ground water, either receiving ground-water discharge (gaining stream) or
recharging the ground water (losing stream) over any given stream reach. These
characteristics should be considered in the evaluation of contaminant transport and
fate.
The Ecology of Running Waters (Hynes 1970) and Introduction to Hydrology
(Viessman et al., 1977) may be reviewed for basic discussions of surface water
hydrology.
13.3.3.1.2 Lakes and Impoundments
Lakes are typically considered natural, while impoundments may be man-
made. The source for lakes and impoundments may be either surface water or
ground water, or both. Impoundments may be either incised into the ground
surface or may be created via the placement of a dam or embankment. As with
streams and rivers, the physical characteristics of lakes and impoundments influence
the transport and fate of contaminant releases and therefore the design of the
monitoring program. The physical characteristics that should be evaluated include
dimensions (e.g., length, width, shoreline, and depth), temperature distribution,
and flow pathways.
Especially in the case of larger lakes and impoundments, flow paths are not
clearcut from inlet to outlet. Not only is the horizontal component of flow in
question, but as depth of the water body increases in the open water zone, chemical
and more commonly physical (i.e., temperature) phenomena create a vertical
stratification or zonation. Figure 13-2 provides a typical lake cross section, showing
the various zones of a stratified lake.
Because of stratification, deeper water bodies can be considered to be
comprised of three lakes. The upper lake, or epilimnion, is characterized by good
light penetration, higher levels of dissolved oxygen, greater overall mixing due to
wave action, and elevated biological activity. The lower lake, or hypolimnion, is the
opposite of the epilimnion. Lying between these is what has been termed the
13-22
-------
/
/
*^:*$&*&&!*'r ~':^^-~~'$^?£$5szZ^^^£'*'
^^^^^^^^^^^^^^^
FIGURE 13-2. TYPICAL LAKE CROSS SECTION
(Source Adapted from Col*. 1975).
13-23
-------
middle lake or mesolimnion, characterized by a rapid decrease in temperature with
depth. Were it not for the phenomenon of lake overturn, or mixing, contaminants
with specific gravities greater than water might be confined to the lowermost lake
strata, where they might remain for some time. Due to the potential importance of
lake mixing to contaminant transport, it is discussed below.
Temperatures within the epilimnion are relatively uniform because of the
mixing that occurs there. Water is most dense at 4o Centigrade (C); above and
below 4oC its density decreases. In temperate climates, lake mixing is a seasonal
occurrence. As the surface of the epilimnion cools rapidly in the fall, it becomes
denser than the underlying strata. At some point, the underlying strata can no
longer support the denser water and an "overturn" occurs, resulting in lake mixing.
A similar phenomenon occurs in the spring as the surface waters warm to 4oc and
once again become denser than the underlying waters.
Because of the influence of stratification on the transport of contaminants
within a lake or reservoir, the location of monitoring points will largely depend on
temperature stratification. The monitoring points on water bodies that are not
stratified will be more strongly influenced by horizontal flowpaths, shoreline
configuration and other factors. The presence of temperature stratification can be
determined by establishing temperature-depth profiles of the water body.
More information on lakes and impoundments may be found in the following
references:
A Treatise on Limnology. Volumes I and II (Hutchinson. 1957,1967) or
Textbook of Limnology (Cole. 1975)
13.3.3.1.3 Wetlands
Wetlands are those areas that are inundated or saturated by surface or ground
water at a frequency and duration sufficient to support, and that under normal
circumstances do support, a prevalence of vegetation typically adapted for life in
saturated soil conditions. Wetlands include, but are not limited to, swamps,
marshes, bogs, and similar areas.
13-24
-------
Wetlands are generally recognized as one of the most productive and sensitive
of biological habitats, often associated with critical habitat for State or Federally
listed special-status species of plants or wildlife. Wetlands also may play a
significant role in basin hydrology, moderating peak surface water flows and
providing recharge to the ground water system. The definition of the extent and
sensitivity of wetlands that may be affected by a release is essential to release
characterization.
High organic content, fine-grained sediments, slow surface water movement
and lush vegetative growth and biological activity contribute to a high potential for
wetlands to concentrate contaminants from releases. This is especially true for
bioaccumulative contaminants, such as heavy metals. The pH/Eh conditions
encountered in many wetlands are relatively unique and can have a significant
effect on a contaminant's toxicity, fate, etc. Seasonal die-off of the vegetation and
flooding conditions within the basin may result in the wetlands serving as a
significant secondary source of contaminants to downstream surface water
receptors.
13.3.3.1.4 Marine Environments
For the purpose of this guidance, marine environments are restricted to
estuaries, intermediate between freshwater and saline, and ocean environments.
Industrial development near the mouths of rivers and near bays outletting directly
into the ocean is relatively widespread, and the estuarine environment may be a
common receptor of releases from industrial facilities.
t
Estuaries are influenced by both fresh water and the open ocean. They have
been functionally defined as tidal habitats that are partially enclosed by land but
have some access to the open sea, if only sporadically, and in which ocean water is
partially diluted by fresh water. Estuaries may also experience conditions where
salinities are temporarily driven above the ocean levels due to evaporative losses.
Because of the protection afforded by encircling land areas, estuaries are termed
"low-energy" environments, indicating that wave energy and associated erosive
and mixing processes are reduced.
13-25
-------
The physical characteristics of an estuary that will influence the design of a
monitoring program are similar to those considered for lakes and impoundments
(i.e., length, width, shoreline, depth, and flow pathways). However, the increased
probability for chemical stratification due to varying salinities may be most
pronounced in areas where freshwater streams and rivers discharge into the
estuary. The monitoring program design should also consider tidal influences on
stratification and contaminant dispersion.
In addition, estuaries, or some portions of estuaries, can be areas of
intergrained sediment deposition. These sediments may contain a significant
organic fraction, which enhances the opportunity for metal/organic adsorption, and
subsequent bioaccumulation. Hence, biomonitoring within an estuary may also be
appropriate. The ionic strength of contaminants may also have an important effect
on their toxicity, fate, etc., in the marine environment.
13.3.3.2 Climatic and Geographic Conditions
A release to the surface water system will be influenced by local
climatological/meteorological and geographic conditions. The release may be
associated only with specific seasonal conditions like spring thaws or meteorological
events such as storms. If the release is intermittent, the environmental conditions at
the time of the release may help identify the cause of and evaluate the extent of the
release. If the release is continuous, seasonal variations should also be evaluated.
The local climatic conditions should be reviewed to determine:
• The annual precipitation distribution (monthly averages);
• Monthly temperature variations;
• Diurnal temperature range (daytime/nighttime difference);
• Storm frequency and severity;
• Wind direction and speed; and
13-26
-------
• Snowfall and snow pack ranges (if applicable).
This information will be useful in developing a sampling schedule and in
selecting sampling methods. From these data, it should be possible to anticipate
the range of climatic conditions at the site. These conditions may be far more
complex than simple cold/hot or wet/dry seasons. Some areas have two or more
"wet seasons", one characterized by prolonged showers, another by brief intense
storms, and perhaps a third as a result of snowmelt. Cold/hot seasons may overlap
these wet/dry seasons to create several climatologically identifiable seasons. Each
season may affect the release differently and may require a separate
characterization. The unique climatological seasons that influence the site should
be identified. Typical winter, spring, summer and fall seasonal descriptions may not
be appropriate or representative of the factors influencing the release. Sources of
climatological data are given in Section 12 (Air).
In addition to the climatological/meteorologica! factors, local geographic1
conditions will influence the design of the sampling program. Topographic1
conditions and soil structure may make some areas prone to flash floods and stream
velocities that are potentially damaging to sampling equipment. In other areas
(e.g., the coastal dune areas of the southeastern states), virtually no runoff oca rs.
Soil porosity and vegetation are such that all precipitation either enters the ground
water or is lost to evapotranspiration. (See Section 9 (Soil) for more information).
A description of the geographic setting will aid in developing a sampling
program that is responsive to the particular conditions at the facility. When
combined with a detailed understanding of the climatological/meteorological
conditions in the area, a workable monitoring framework can be created.
13.3.4 Sources of Existing Information
Considerable information may already be available to assist in characterizing a
release. Existing information should be reviewed to avoid duplication of previous
efforts and to aid in focusing the RFI. Any information relating to releases from the
unit, and to hydrogeological, meteorological, and environmental factors that could
influence the persistence, transport or location of contaminants should be
reviewed. This information may aid in:
13-27
-------
• Delineating the boundaries of the sampling area;
• Choosing sampling and analytical techniques; and
• Identifying information needs for later phases of the investigation.
Information may be obtained from readily available sources of geological and
meteorological data, waste characteristics, and facility operations records. (See also
Sections 2,3, 7 and Appendix A).
13.4 Design of a Monitoring Program to Characterize Releases
Following characterization of the contaminant source and environmental
setting, a monitoring program is developed. This section outlines and describes
factors that should be considered in design of an effective surface water monitoring '
program. The characterization of contaminant releases may take place in multiple
phases. While the factors discussed in this section should be carefully considered in
program design, each of these generic approaches may require modification for
specific situations.
The primary considerations in designing a surface water monitoring program
are:
• Establishing the objectives of the monitoring program;
• Determining the constituents of concern;
• Establishing the hydrologic characteristics of the receiving water and
characteristics of the sediment and biota, if appropriate;
• Selecting constituents and/or indicators for monitoring;
• Selecting monitoring locations and monitoring frequency; and
13-28
-------
I
• Determining the need for sediment monitoring and, hydrologic and
biomonitoring.
13.4.1 Objectives of the Monitoring Program
The principal objectives of a monitoring program are to:
• Identify the characteristics of releases (e.g., continuous vs intermittent);
• Identify the fate of constituents;
• Identify the nature, rate, and extent of the release and actual or
potential effects on water quality and biota; and
• Identify the effect of temporal variation on constituent fate and identify
impacts on water quality and biota.
Periodic monitoring of the surface water system is often the only effective
means of identifying the occurrence of releases and their specific effects. Releases
can be continuous or intermittent, point source, or non-point source. The concept
of monitoring is the same, regardless of the frequency or form of the release. A
series of measurements, taken over time, better approximate the actual release to
surface waters than a one-time grab sample.
The functional difference between monitoring the various types of discharges
is the point of measurement. Point source discharges may be monitored at and/or
near the discharge point to surface waters. The fate and potential effects of non-
point source discharges should be inferred through measurement of the presence of
constituents of concern or suitable indicators of water quality within the receiving
water body.
The monitoring program should also establish the background condition
against which to measure variations in a continuous release or the occurrence of an
intermittent release. Such information will enable the facility owner or operator to
compile data that will establish trends in releases from a given unit(s) as well as to
identify releases from other sources.
13-29
-------
Monitoring programs should characterize contaminant releases as a function
of time. Climatologic factors such as frequency of intense rainfall, added effects of
snowmelt, temperature extremes, and mixing in lakes and estuaries should be
evaluated and quantified as causative agents for intermittent contaminant release.
Important concepts to consider in designing the monitoring program for
surface water to help meet the above-stated objectives are described below.
13.4.1.1 Phased Characterization
The initial phase of a surface water release characterization program may be
directed toward verification of the occurrence of a release identified as suspected
by the regulatory agency. It may also serve as the first step for characterizing
surface water systems and releases to those systems in cases where a release has
already been verified.
The initial characterization wilt typically be a short-duration activity, done in
concert with evaluation of other media that may either transport contaminants to
surface waters, or may themselves be affected by discharges from surface waters
(i.e., inter-media transport). It may be particularly difficult to define intermittent
discharges in the initial characterization effort, especially if the contaminants from
these releases are transient in the surface water body.
If the waste characterization is adequate, the initial characterization phase
may rely upon monitoring constituents and suitable indicator parameters to aid in
defining the nature, rate, and extent of a release. Subsequent phases of release
characterization will normally take the form of an expanded environmental
monitoring program and hydrologic evaluation, sensitive to seasonal variations in
contaminant release and loading to the receiving water bodies, as well as to natural
variation in hydrologic characteristics (e.g., flow velocity and volume, stream cross
section).
13-30
-------
13.4.1.2 Development of Conceptual Model
To effectively design a monitoring program, it is important to develop a
conceptual model or understanding of the fate of constituents of the release in the
receiving water body. This conceptual understanding will assist in answering the
following questions.
• What portion of the receiving water body will be affected by the release
and what conditions (e.g., low flow, immediate stormwater runoff)
represent reasonable worst case conditions under which sampling should
occur?
• What should the relative concentrations of contaminants be at specific
receptor points within the water body (e.g., public water supply intakes
downstream of a site)?
i
• How does the release of concern relate to background contamination in'
the receiving water body as a result of other discharges?
• How might the monitoring program be optimized, based on
contaminant dispersion and relative concentrations within the receiving
water body?
The fate of waste constituents entering surface waters is highly dependent on
the hydrologic characteristics of the various classifications of water bodies, (i.e.,
streams and rivers, lakes and impoundments, wetlands, and estuaries, as discussed
earlier). Because of their complexity, methods for characterization of contaminant
fate in wetlands and estuaries is not presented in detail in this guidance. The reader
is referred to Mills (1985) for further detail on characterizing contaminant fate in
wetlands and estuaries.
13.4.1.3 Contaminant Concentration vs Contaminant Loading
Concentration and loading are different means of expressing contaminant
levels in a release or receiving water body. The concept is important in the selection
of constituents for monitoring. Both concentration and loading should be
13-31
-------
evaluated with respect to the release and the receiving waters. Basing an
evaluation solely on concentration may obscure the actual events. In addition, it is
essential to quantify individual sources of contaminants and the relationships
between media, as well as the loading found in the receiving water body, to
effectively define the nature and extent of the contaminant release.
Contaminant concentrations in receiving waters have specific value in
interpreting the level of health or environmental effects anticipated from the
release. Contaminant loading provides a common denominator for comparison of
contaminant inputs between monitoring points. In addition, especially in the case
of contaminants that are persistent in sediments (e.g., heavy metals), loadings are a
convenient means of expressing ongoing contributions from a specific discharge.
The distinction between concentration and loading is best drawn through the
following example.
A sample collected from a stream just upgradient of a site boundary (Station
A) has a concentration of 50 micrograms per liter (ug/l) of chromium. A second
sample collected just downstream of the site (Station B) has a chromium
concentration of 45 ug/l- From these data it appears that the site is not releasing
additional chromium to the stream. If, however, the stream flow is increasing
between these two sampling locations, a different interpretation is apparent. If the
stream flow at the upstream location is 1,000 gallons per minute (gpm) and the
downstream location is 1,300 gpm, the actual loading of chromium to the stream at
the two locations is as follows:
Station A
Chromium = (50.0 ug/l)(1,000 gal/min)(10-9 kg/ug)(60 min/hr)(3.785 I/gal) - 0.0114
kg/hr
Station B
Chromium = (45.0 ug/l)(1,300gal/min)(10-9kg/ng)(60 min/hr)(3.785 I/gal) = 0.0133
kg/hr
It is now apparent that somewhere between the two sampling stations is a
source(s) contributing 0.0019 kg/hr of chromium. If all of the flow difference (i.e.,
13-32
-------
300 gpm) is from a single source, then this source would have a chromium
concentration of 27.9 ug/l:
Chromium = [(0.0019 kg/hr)(1Q9 ug/kg)(1hr/60min)(1 min/300 gal)(1 gal/3.785 I)] =
27.9 ug/l
If, however, 90 percent of this flow difference (i.e., 270 gpm) was due to
ground-water discharge with a chromium concentration below detectable limits
and the remaining 10 percent (i.e., 30 gpm) was the result of a direct discharge from
the facility, this discharge could have a chromium concentration of 279 ug/l.
13.4.1.4 Contaminant Dispersion Concepts
Contaminant dispersion concepts and models of constituent fate can be used
to define constituents to be monitored and the location and frequency of
monitoring. Dispersion may occur in streams, stratified lakes or reservoirs, and in1
estuaries. Dispersion may be continuous, seasonal, daily, or a combination of these.
The discussion below is based on information contained in the Draft
Superfuno' Exposure Assessment Manual (EPA, 1987) relative to simplified models
useful in surface water fate analyses. The reader is directed to that document for a
more in-depth discussion of models. The equations presented below are based on
the mixing zone concept originally developed for EPA's National Pollutant
Discharge Elimination System (NPDES) under the Clean Water Act. To avoid
confusion over regulatory application of these concepts in the NPDES program, and
the approach presented below (basically to aid in the development of a monitoring
program), the following discussion refers to use of the "Dispersion Zone".
The following equation provides an approximate estimate of the
concentration of a substance downstream from a point source release, after
dilution in the water body:
CuQu + CWQW
Cr =
Qu + Qw
13-33
-------
where:
Cr = downstream concentration of substance following complete
dispersion (mass/volume)
Cu = upstream concentration of substance before effluent release point
(mass/volume)
Cw = concentration of substance in effluent (mass/volume)
Qw = effluent flow rate (volume/time)
Ou - upstream flow rate before effluent release point (volume/time)
The following equation may be used to estimate instream concentrations after
dilution in situations where waste constituents are" introduced via inter-media
transfer or from a non-point source, or where the release rate is known in terms of
mass per unit time, rather than per unit effluent volume:
Tr + Mu
where:
Tr = inter-media transfer rate (mass/time)
MU = upstream mass discharge rate (mass/time)
Qt - stream flow rate after inter-media transfer or non-point source
release (volume/time)
The above two equations assume the following :
• Dispersion is instantaneous and complete;
• The waste constituent is conserved (i.e., all decay or removal processes
are disregarded); and
• Stream flow and rate of contaminant release to the stream are constant
(i.e., steady-state conditions).
13-34
-------
For a certain area downstream of the point of release, the assumption of
complete dispersion may not be valid. Under certain situations, the dispersion zone
can extend downstream for a considerable distance, and concentrations can be
considerably higher within the dispersion zone than those estimated by the
equation. The length of this zone can be approximated by the following equation:
0.4
D2 =
O.Sdygds
where:
DZ = dispersion zone length (length units)
w = width of the water body (length units)
u = stream velocity (length/time)
d = stream depth (length units)
s = slope (gradient) of the stream channel (length/length) '
g = acceleration due to gravity (32 ft/sec2)
Within the dispersion zone, contaminant concentrations will show spatial
variation. Near the release point the contaminant will be restricted (for a discharge
along one shoreline) to the nearshore area and (depending on the way the
discharge is introduced and its density) can be vertically confined. As the water
moves downstream, the contaminant will disperse within surrounding ambient
water and the plume will widen and deepen. Concentrations will generally
decrease along the plume centerline and the concentration gradients away from
the centerline will decrease. Eventually, as described above, the contaminant will
become fully dispersed within the stream; downstream from this point
concentration will be constant throughout the stream cross-section, assuming that
the stream flow rate remains constant.
It is important to understand this concentration variability within the
dispersion zone if measurements are to be made near the release. Relatively
straightforward analytical expressions (See Neely, 1982) are available to calculate
the spatial variation of concentration as a function of such parameters as stream
width, depth, velocity, and dispersion coefficients. Dispersion coefficients
characterize the dispersion between the stream water and contaminated influx;
13-35
-------
they can, in turn, be estimated from stream characteristics such as depth, gradient,
and path (i.e., straight or bends).
The above considerations are for instream concentrations resulting from the
releases of concern. If total instream concentrations are required, the
concentrations determined from background water samples should also be
considered. In addition, if introduction of the contaminant occurs over a fixed
stream reach, as mig : be the case with a non-point discharge, it should be assumed
that the dispersion zone begins at the furthest downstream point within this reach.
13.4.1.5 Conservative vs Non-Conservative Species
The expressions presented thus far have assumed that the contaminant(s) of
concern is conservative (i.e., that the mass loading of the contaminant is affected
only by the mechanical process of dilution). For contaminants that are non-
conservative, the above equations would provide a conservative estimate of '
contaminant loading at the point of interest within the receiving water body.
In cases where the concentration after dilution of a non-conservative
substance is still expected to be above a level of concern, it may be useful to
estimate the distance downstream where the concentration will remain above this
level and at selected points in between. The reader is referred to the Draft
Superfund Exposure Assessment Manual (EPA, 1987), for details regarding this
estimation procedure and to specific State Water Quality Standards for
determination of acceptable instream concentrations.
13.4.2 Monitoring Constituents and Indicator Parameters
13.4.2.1 Hazardous Constituents
The facility owner or operator should propose a list .of constituents and
indicator parameters, if appropriate, to be included in the Surface Water
investigation. This list should be based on a site-specific understanding of the
composition of the release source(s) and the operative release mechanisms, as well
as the physical and chemical characteristics of the various classes of contaminants.
These factors, as well as potential release mechanisms and migration pathways,
13-36
-------
have been discussed in Sections 13.3 and 13.4.1. Also refer to Sections 3 and 7 of this
guidance, and to the lists of constituents provided in Appendix B.
13.4.2.2 Indicator Parameters
Indicator parameters (e.g., chemical and biochemical oxygen demand, pH,
total suspended solids, etc.) may also play a useful role in release characterization.
Though indicators can provide useful data for release verification and
characterization, specific hazardous constituent concentrations should always be
monitored. Furthermore, many highly toxic constituents may not be detected by
indicators because they do not represent a significant amount of the measurement.
Following are brief synopses of some common indicator parameters and field
tests that can be used in investigations of surface water contamination. The use of
biomonitoring as an indicator of contamination is discussed in Section 13.4.5.
i
Biochemical Oxygen Demand (BOD) and Chemical Oxygen Demand (CQD)--BOD is
an estimate of the amount of oxygen required for the biochemical degradation of
organic material (carbonaceous demand) and the oxygen used to oxidize inorganic
material such as sulfides and ferrous iron. It may also measure the oxygen used to
oxidize reduced forms of nitrogen (nitrogenous demand! unless their oxidation is
prevented by an inhibitor. Because the complete stabilization of a BOD sample may
require an extended period, 5 days has been accepted as the standard incubation
period. While BOD measures only biodegradable organics, non-biodegradable
materials can exert a demand on the available oxygen in an aquatic environment.
COD measures the total oxygen demand produced by biological and chemical
oxidation of waste constituents. Availability of results for the COD in approximately
4 hours, versus 5 days for the BOD, may be an important advantage of its use in
characterizing releases of a transient nature.
COD values are essentially equivalent to BOD when the oxidizable materials
present consist exclusively of organic matter. COD values exceed BOD values when
non-biodegradable materials that are susceptible to oxidation are present. The
reverse is not often the case; however, refinery wastes provide a notable exception.
There are some organic compounds, such as pulp and paper mill cellulose, that are
non-biodegradable, yet oxidizable. Nitrogenous compounds, which may place a
13-37
-------
significant drain on available oxygen in aquatic environments, are not measured in
the COD test. In addition, chlorides interfere with the COD test, leading to
overestimates of the actual COD. BOD/COD ratios, as an indicator of
biodegradability, are discussed in Section 9 (Soil). BOD and COD may be useful
indicator parameters if the release is due primarily to degradable organic wastes.
Total Organic Carbon (TOC)--Total organic carbon is valuable as a rapid estimator of
organic contamination in a receiving water. TOC, however, is not specific to a given
contaminant or even to specific classes of organics. In addition, TOC measurements
have little use if the release is primarily due to inorganic wastes.
Dissolved Oxygen (DO)--Measurements of DO may be readily made in the field with
an electronic DO meter, which has virtually replaced laboratory titrations.
Especially in lake environments, it is valuable to know the DO profile with depth.
The bottoms of lakes are often associated with anoxic conditions (absence of
oxygen) because of the lack of mixing with the surface and reduced or non-existent'
photosynthesis. Influx of a contaminant load with a high oxygen demand can
further exacerbate oxygen deficiencies under such conditions. In addition, low DO
levels favor reduction, rather than oxidation reactions, thus altering products of
chemical degradation of contaminants. DO levels less than 3 mg/liter (ppm) are
considered stressful to most aquatic vertebrates (e.g., fish and amphibians).
pJH-pH is probably one of the most common field measurements made of surface
waters. It is defined as the inverse log of the hydrogen ion concentration of an
aqueous medium. pH is generally measured in the field with analog or digital
electronic pH meters.
As an indicator of water pollution, pH is important for two reasons:
• The range within which most aquatic life forms are tolerant is usually
quite narrow. Thus, this factor has significant implications in terms of
impact to aquatic communities; and
• The pH of a solution may be a determining factor in moderating other
constituent reactions.
13-38
-------
Temperature-Alonq with pH, temperature is a fundamental parameter that should
always be recorded in the field when a water sample is collected. Temperature is
most often measured by electronic meters that can simultaneously record pH and/or
specific conductance. Temperature is a significant parameter because:
• Most aquatic species are sensitive to elevated temperatures;
• Elevated temperatures can be an indication of a contaminant plume;
• Most chemical reactions are temperature-dependent; and
• Temperature defines strata in thermally-stratified lakes.
Alkalinity-Alkalinity is the capacity of water to resist a depression in pH. It is,
therefore, a measure of the ability of the water to accept hydrogen ions without
resulting in creation of an acid medium. Most natural waters have substantial
buffering capacity (a resistance to any alteration in pH, toward either the alkaline
or acid side) through dissolution of carbonate-bearing minerals, creating a
carbonate/bicarbonate buffer system.
Alkalinity is usually expressed in calcium carbonate (CaCOa) equivalents and is
the sum of alkalinities provided by the carbonate, bicarbonate, and hydroxide ions
present in solution. Alkalinities in the natural environment usually range from 45 to
200 milligrams per liter (mg/l). Some limestone streams have extremely high
buffering capacities, while other natural streams are very lightly buffered and are
extremely sensitive to acid (or alkaline) loadings.
Hardness-The sum of carbonate and bicarbonate alkalinities is also termed
carbonate hardness. Hardness is generally considered a measure of the total
concentration of calcium and magnesium ions present in solution, expressed as
CaCO? equivalents.
Calcium and magnesium ions play a role in plant and animal uptake of
contaminants; knowledge of the hardness of a surface water is necessary for
evaluation of the site-specific bioaccumulative potential of certain contaminants
(e.g., heavy metals).
13-39
-------
Total Solids-Analytically. the total solids (TS) content of a water is that remaining
after evaporation at 103-115° C or 180oC, depending on the method. The residue
remaining represents a sum of the suspended, colloidal, and dissolved solids.
Hazardous constituents with high vapor pressures (i.e., volatiles, semi-volatiles) will
not remain after evaporation, and will not contribute to the TS determination.
Suspended Solids-Suspended solids are those materials that will not pass a glass-
fiber filter. Suspended solids contain both organic and inorganic compounds. For
the purpose of comparison to water samples, the average domestic wastewater
contains about 200 ppm (mg/l) of suspended solids.
Volatile Suspended Solids-Volatile suspended solids are the volatile organic portion
of the suspended solids. Volatile suspended solids are the components of
suspended solids that volatilize at a temperature of 600° C. The residue or ash is
termed fixed suspended solids and is a measure of the inorganic fraction (i.e.,1
mineral content). The only inorganic salt that will degrade below 600<> C is'
magnesium carbonate.
Total Dissolved Solids-Total dissolved solids context is obtained by subtracting
suspended solids from total solids. Its significance lies in the fact that it cannot be
removed from a surface water or effluent stream through physical means or simple
chemical processes, such as coagulation.
Salinity-The major salts contributing to salinity are sodium chloride (NaCI) and
sulfates of magnesium and calcium (MgSOd, CaSO4). The following represents an
example of classification of saline waters on the basis of salt content.
Type of Water Total Dissolved Solids (As Salts)
brackish 1,000 to 35,000 mg/l
seawater 35,000 mg/l
brine > 35,000 mg/l
Specific Conductances-Conductivity measures the capacity to conduct current. Its
counterpart is, of course, resistance, measured in ohms. The unit of conductivity has
been defined as the mho. Specific conductance is conductivity/unit length. The most
common units for specific conductance are mho/cm. Specific conductance can be
13-40
-------
measured instantaneously with electronic conductivity meters to comparatively
high levels of accuracy and precision in the field and is an excellent real-time
indicator parameter.
Conductivity generally rises with increased concentration of dissolved (ionic)
species. Therefore, waters with high salinities, or high total dissolved solids, can be
expected to exhibit high conductivities. Variations in specific conductance within a
stream reach or a portion of an impoundment may indicate the presence of
contaminant release points.
Major Ion Chemistrv-The nature and prevalence of ionic species may serve as
indicators of pollution from waste sources containing inorganics. Ions result from
the dissociation of metal salts. The cation (e.g., Na + , Ca + , Mg+ +) is typically a
metallic species and the anion (e.g., CI-, $04--) a non-metallic species.
i
A common approach to use of ion chemistry as an indicator of waste
contamination in surface waters is to analyze for anions. Standard Methods
(American Public Health Association, 1985), protocol no. 429 includes the following
common anions as analytes:
Chloride (CI-)
Fluoride (F-)
Bromide (Br-)
Nitrate (NOs-)
Nitrite (NO2-)
Phosphate (PO4—)
Sulfate (S04-)
While elevated concentrations of these anions may indicate the presence of
inorganic constituents or other contaminants, no information will be provided
regarding the identity of specific constituents or contaminants. In addition,
elevated levels of anions may be associated with effluent from domestic refuse
and/or runoff from fertilized agricultural fields.
13-41
-------
The nature and concentrations of naturally-occurring ions in surface waters
are a function of the geologic setting of the area, and may be temporarily affected
by stormwater runoff, which may cause resuspension of streambed sediments.
In reference to their inertness with respect to constituent and biological
degradation, ionic species are termed "conservative." The fact that their mass is not
altered (i.e., is conserved) in surface waters permits them to be used in simple
dilution modeling.
13.4.3 Selection of Monitoring Locations
The selection of monitoring locations should be addressed prior to sample
acquisition because it may affect the selection of monitoring equipment and
because monitoring locations will affect the representativeness of samples taken
during the monitoring program. Samples must be taken at locations representative
of the water body or positions in the water body with specific physical or chemical
characteristics. As discussed in Section 13.4.1.2 (Development of Conceptual
Model), one of the most important preliminary steps in defining monitoring
locations in a surface water monitoring program is developing a conceptual model
of the manner in which the release is distributed within the receiving water body.
This is dependent on the physical and chemical characteristics of the receiving
water, the point source or non-point source nature of the discharge, and the
characteristics of the constituents themselves.
As a practical example, if a release contains contaminants whose specific
gravities exceed that of water, it may behave almost as a separate phase within the
receiving water body, traveling along the bottom of the water body. As another
example, certain contaminants may be found in comparatively low concentrations
in sediments or within the water column, yet may accumulate in aquatic biota via
bioaccumulation. In this case monitoring of the biota would be advised. If the
facility owner or operator is unaware of these phenomena, it would be possible for
the monitoring program to show no evidence of contamination.
In general, it will be desirable to locate monitoring stations in three areas
relative to the discharge in question:
13-42
-------
Background monitoring stations:
Background monitoring should be performed in an area known not
to be influenced by the release of concern (e.g., upstream of a
release).
Monitoring stations at the release point(s) or area:
If the release is a point source or area source, periodic monitoring
should be performed at monitoring stations near the discharge
origin to determine the range of contaminant concentrations. The
contaminant stream (e.g., leachate seep, runoff) should also be
subjected to monitoring.
Monitoring of the receiving water body within the area of
influence: '
i
One means of evaluating the water quality effects of a discharge is
to monitor the discharge point and model its dispersion (e.g., using
dispersion zone concepts discussed previously) within the receiving
water body. The results of this modeling may be used to determine
appropriate sampling locations. Actual sampling of the area
thought to be influenced by the release is required. The "area of
influence" may be defined as that portion of the receiving water
within which the discharge would show a measurable effect. As
described previously, the area to be sampled is generally defined in
a phased fashion, based on a growing base of monitoring data. It is
usually prudent to start with a conservatively large area and
continually refine its boundaries. This is particularly true where
sensitive receptors (e.g., public water supply intakes, sensitive
wetlands, recreation areas) lie downstream of the release. In
addition, in order to determine the full extent of the release (and
its effects), samples should be taken at locations beyond the
perceived area of influence.
13-43
-------
The majority of the effort of the monitoring program will take place within
the area of influence, as defined above. Many factors are involved in selecting
monitoring stations within this area, the most critical being:
• The homogeneity of the water body in terms of temperature, flow,
salinity, and other physical and chemical characteristics;
• The representativeness of the monitoring point, in terms of both
contaminant characteristics and use factors;
• The presence of areas of pronounced water quality degradation; and
• Defensible monitoring design, including the choice of the monitoring
scheme (random, stratified random, systematic, etc.), the experimental
design, and adequate sample size determination.
i
Estuarine areas are particularly difficult in terms of selecting monitoring
locations that will allow an adequate evaluation of constituent distribution,
because detailed knowledge of the hydrologic characteristics of the estuary is
required to accurately locate representative monitoring points. Freshwater- salt
water stratification is a particularly important consideration. If stratification is
known to occur or is suspected, sampling should be conducted at a range of depths
within the estuary as well as at surface locations.
The selection of sampling locations is described in much greater detail in EPA
(1973,1982).
13.4.4 Monitoring Schedule
The monitoring schedule or frequency should be a function of the type of
release (i.e., intermittent vs continuous), variability in water quality of the receiving
water body (possibly as a result of other sources), stream flow conditions, and other
factors causing the release (e.g., meteorological or process design factors).
Therefore, frequency of monitoring should be determined by the facility owner or
operator on a site-specific basis. Sampling points with common monitoring
13-44
-------
objectives should be sampled as close to simultaneously as possible, regardless of
the monitoring frequency established.
Factors important in determining the required frequency of monitoring
include:
• The homogeneity of the receiving water in terms of factors that
may affect the fate of constituents. The most important of these
are flow and seasonal or diurnal stratification.
• The characteristics of the releases. Releases may be continuous or
event-associated.
As an example, continuous, point source releases of low variability subject to
few, if any, additional releases may require relatively infrequent monitoring. On
the other hand, releases known to be related to recurrent causes, such as rainfall'
and runoff, may require monitoring associated with the event. Such monitoring is
termed "event" sampling. To evaluate the threshold event required to trigger
samplinn, as well as the required duration of the monitoring following the event, it
is necessary that the role of the event in creating a release from the unit be well
understood. In what is probably a very common example, if stormwater runoff is
the event of concern, a hydrograph for various storm return intervals and durations
should be estimated for the point or area of interest and the magnitude and
duration of its effects evaluated.
Continuous monitoring can be accomplished through in situ probes that
provide frequent input to field data storage units. However, continuous
monitoring is feasible only for the limited number of constituents and indicator
parameters for which reliable automatic sampling/recording equipment is available.
In estuaries, samples are generally required through a tidal cycle. Two sets of
samples are taken from an area on a given day, one at ebb or flood slack water and
another at three hours earlier or later at half tide interval. Sampling is scheduled
such that the mid-sampling time of each run coincides with the calculated
occurrence of the tidal condition.
13-45
-------
Where investigating discharges of contaminated ground water to streams or
rivers, it is important to sample during low flow conditions (e.g., using State critical
low flow designations) to better assess the possible effects of the release(s) of
concern.
13.4.5 Hydrologic Monitoring
The monitoring program should also include provisions for hydrologic
monitoring. Specifically, the program should provide for collection of data on the
hydrologic condition of the surface water body at the time of sampling.
For example, some indication of the stage and discharge of a stream being
monitored needs to be recorded at the time and location each water sample is
collected. Similarly, for sampling that occurs during storms, a record of rainfall
intensity over the duration of the storm needs to be obtained. Without this
complementary hydroiogic data, misinterpretation of the water quality data in1
terms of contaminant sources and the extent of contamination is possible.
The techniques for hydrologic monitoring that could be included in a
monitoring program range in complexity from use of simple qualitative descriptions
of streamflow to permanent installation of continuously-recording stream gages.
The techniques appropriate in a given case will depend on the characteristics of the
unit and of the surface waters being investigated. Guidance on hydrologic
monitoring techniques can be found in the references cited in Section 13.6.1.
13.4.6 The Role of Biomonitoring
The effects of contaminants may be reflected in the population density,
species composition and diversity, physiological condition, and metabolic rates of
aquatic organisms and communities. Biomonitoring techniques can provide an
effective complement to detailed chemical analyses for identifying chemical
contamination of water bodies. They may be especially useful in those cases where
releases involve constituents with a high propensity to bioaccumulate. This includes
most metal species and organicswith a high bioconcentration factor (e.g., > 10) or a
high octanol/water partition coefficient (e.g., .>2.3). These properties were
discussed in Section 13.3.
13-46
-------
Biomonitoring techniques may include:
• Community ecology studies;
• Evaluation of food chain/sensitive species impacts; and
• Bioassays.
These techniques are discussed below.
13.4.6.1 Community Ecology Studies
Indicator species are useful for evaluating the well-being of an aquatic
community that may be stressed by the release of contaminants. For example, the
condition of the benthic macroinvertebrate community is commonly used as an
indicator of the presence of contaminants. The objective of studying the naturally-
occurring biological community is to determine community structure that would be'
expected, in an undisturbed habitat. If significant changes occur, perturbations in1
the community ecology may be linked to the disturbance associated with release of
contaminants to the water bodv.
EPA is engaged in research to develop rapid bioassessment techniques using
benthic macroinvertebrates. Although protocols are being considered, in general
these techniques suffer from lack of data on undisturbed aquatic communities and
associated water quality information. For some areas (e.g., fisheries), however,
indices to community health based on benthic invertebrate communities are
available (Hilsenhoff 1982, Cummins and Wilgbach, 1985).
Because species diversity is a commonly-used indicator of the overall health of
a community, depressed community diversity may be considered an indicator of
contamination. For example, if a release to surface waters has a high chemical
oxygen demand (COD) and, therefore, depresses oxygen levels in the receiving
water body, the number of different species of organisms that can colonize the
water body may be reduced. In this case the oxygen-sensitive species (e.g., the
mayfly), is lost from the community and is replaced by more tolerant species. The
number of tolerant species is small, but the number of individuals within these
species that can colonize the oxygen-deficient waters may be quite large. Therefore,
13-47
-------
the overall species diversity could be low, even though the numbers of organisms
may be high.
Evaluations of community ecology should however, be sensitive to the role
that habitat variability may play in altering community structure. Diversity of
habitat may be altered by natural physical conditions (e.g., a rapid increase in
stream gradient), substrate characteristics (e.g.,silty versus rocky substrate), and so
forth. It may also be difficult to directly link contaminant levels with the presence or
absence of aquatic organisms, unless there is a secondary impact that is more self-
evident, such as high oxygen demand, turbidity, or salinity.
13.4.6.2 Evaluation of Food Chain Sensitive/Species Impacts
At this level of biomonitoring, the emphasis is actually on the threat to specific
fish or wildlife species, or man, as a result of bioaccumulation of constituents from
the release being carried through the food web. Bioaccumulative contaminants are'
not rapidly eliminated by biological processes and accumulate in certain organs or
body tissues. Their effect may not be felt by individual organisms that initially
consume the contaminated substrate or take up the contaminants from the water.
However, organisms at higher trophic levels consume the organisms of the lower
trophic levels. Consequently, contaminants may become bioaccumulated in
organisms and biomagnified through the food web.
Examination of the potential for bioaccumulation and biomagnification of
contaminants requires at least a cursory characterization of the community to
define its trophic structure, that is, which organisms occupy which relative positions
within the community. Based on this definition, organisms representative of the
various trophic levels may be collected, sacrificed, and analyzed to determine the
levels of the contaminants of interest present.
If a specific trophic level is of concern, it may be possible to short-cut the
process by selectively collecting and analyzing organisms from that level for the
contaminants of concern. This may be the case, for instance, if certain organisms
are taken by man either commercially or through recreational fishing, for
consumption. It may also be necessary to focus on the prey of special-status fish or
wildlife (e.g., eagles and other birds of prey) to establish their potential for
13-48
-------
exposure. This type of biomonitoring may be especially useful if constituents
released have a relatively high potential to bioaccumulate. A discussion of
indicators that are generally predictive of constituents which have a significant
potential for bioaccumulation was presented in Section 13.3.
In addition, in the selection of organisms it is important to consider the ability
of a given organism to accumulate a class of contaminants and the residential vs
migratory nature of the organisms. For example, bullfrogs are superior for
accumulating metals but poor for organics; spawning (thus migratory) salmon
would be much less useful for characterizing a release from a local facility than
would resident fish.
13.4.6.3 Bioassay
Bioassay may be defined as the study of specially selected representative
species to determine their response to the release of concern, or to specific*
constituents of the release. The organisms are "monitored" for a period of timef
established by the bioassay method. The objective of bioassay testing is to establish
a concentration-response relationship between the contaminants of concern and
representative biota that can be used to evaluate the effects of the release.
Bioassay testing may involve the use of indigenous organisms (U.S. EPA, 1973) or
organisms available commercially for this purpose. Bioassays have an advantage
over strict constituent analyses of surface waters and effluents in that they measure
the total effect of all constituents within the release on aquatic organisms (within
the limits of the test). Such results, therefore, are not as tightly constrained by
assumptions of contaminant interactions. Discussions of bioassay procedures are
provided by Peltier and Weber (1985) and Horning and Weber (1985).
The criterion commonly used to establish the endpoint for a bioassay is
mortality of the test organisms, although other factors such as depressed growth
rate, reproductive success, behavior alteration, and flesh tainting (in fish and
shellfish) can be used. Results are commonly reported as the LC50 (i.e., the lethal
concentration that resulted in 50 percent mortality of the test organisms within the
time frame of the test) or the EC50 (i.e., the effective concentration that resulted in
50 percent of the test organisms having an effect other than death within the time
frame of the test).
13-49
-------
One potential use of bioassays during the RFI is to predict the effect of a
release on sensitive species residing in the affected surface water(s). Bioassay may
be especially useful if the release is intermittent. In this case, samples of the waste
may be taken from the unit of concern and used to conduct bioassay tests. The
bioassay may be conducted using the waste at 100 percent strength, and in diluted
form, to obtain a concentration response relationship. The results of this testing
may then be used to predict the effects of a release on the surface water biota.
Bioassays can serve as important complements to the overall monitoring
program. In considering the role and design of bioassays in a monitoring program,
the facility owner or operator should be aware of the advantages and limitations of
toxicity testing. The study design must account for factors such as species sensitivity
and frequency of monitoring which may be different from the considerations that
feed into chemical monitoring programs. Toxicity testing techniques are an integral
part of the Clean Water Act program to control the discharge of toxic substances.
Many issues associated with toxicity testing have been addressed in this context in
the Technical Support Document for Water Quality-Based Toxics Control (Brandes et
al, 1985).
13.5 Data Management and Presentation
The owner or operator will be required to report on the progress of the RFI at
appropriate intervals during the investigation. The data should be reported in a
clear and concise manner, and interpretations should be supported by the data. The
following data presentation methods are suggested for the various phases of the
surface water investigation. Further information on the various procedures is given
in Section 5. Section 5 also provides guidance on various reports that may be
required.
13.5.1 Waste and Unit Characterization
Waste and unit characteristics should be presented as:
• Tables of waste constituents, concentrations, effluent flow and
mass loadings;
13-50
-------
• Tables of relevant physical and chemical properties of potential
contaminants (e.g., solubility);
• Narrative description of unit operations;
• Surface map and plan drawings of facility, unit(s), and surface
waters; and
• Identification of "reasonable worst case" contaminant release to
surface waters.
13.5.2 Environmental Setting Characterization
The environment of the waste unit(s) and surface waters should be described
in terms of physical and biological environments in the vicinity. This description
should include:
• A map of the area portraying the location of tr.e waste unit in
relation to potential receiving waters;
• A map or narrative classification of surface waters (e.g., type of
surface water, uses of the surface water, and State classification, if
any);
*
• A description of the climatological setting as it may affect the
surface hydrology or release of contaminants; and
• A narrative description of the hydrologic conditions during
sampling periods.
13.5.3 Characterization of the Release
The complex nature of the data involving multiple monitoring events,
monitoring locations, matrices (water, sediment, biota), and analytes lends itself to
graphic presentation. The most basic presentation is a site map or series of maps
13-51
-------
that locate the monitoring stations for each monitoring event. These maps may
also be adapted to include isopleths for specific analytes; however, since the
isopleths imply a continuity within their borders, they may not be appropriate
unless they are based on an adequate number of monitoring points and
representative data. The contours should be based on unit intervals whose accuracy
ranges do not overlap. In most situations, two separate reporting formats are
appropriate. First, the data should be included as tables. These tables should
generally be used to present the analytical results for a given sample. Each table
could include samples from several locations for a given matrix, or could include
samples from each location for all sample matrices. Data from these tables can then
be summarized for comparison purposes using graphs.
Graphs are most useful for displaying spatial and temporal variations. Spatial
variability for a given analyte can be displayed using bar graphs where the vertical
axis represents concentration and the horizontal axis represents downstream
distance from the discharge. The results from each monitoring station can then be1
presented as a concentration bar. Stacked bar graphs can be used to display these
data from each matrix at a given location or for more than one analyte from each
sample.
Similarly, these types of graphs can be used to demonstrate temporal
variability if the horizontal axis represents time rather than distance. In this
configuration, each graph will present the results of one analyte from a single
monitoring location. Stacked bars can then display multiple analytes or locations.
Line graphs, like isopleths, should be used cautiously because the line implies a
continuity, either spatial or temporal, that may not be accurately supported by the
data.
Scatter plots are useful for displaying correlations between variables. They can
be used to support the validity of indicator parameters by plotting the indicator
results against the results for a specific constituent.
Graphs are used to display trends and correlations. They should not be used to
replace data tables, but rather to enhance the meaning of the data.
13-52
-------
13.6 Field and Other Methods
The purpose of this section is to provide an overview of methods that can be
used to characterize the nature, rate, and extent of contaminant releases to surface
water. Detailed descriptions of specific methods can be found in the indicated
references.
The methods presented in this section relate to four specific areas, as follows:
• Surface Water Hydrology;
• Sampling and Constituent Analysis of Surface Water, Sediments,
and Biota;
• Characterization of the Condition of the Aquatic Community; and
I
• Bioassay Methods.
13.6.1 Surface Water Hydrology
The physical attributes of the potentially affected water body should be
characterized to effectively develop a monitoring program and to interpret results.
Depending on the characteristics of the release and the environmental setting, any
or all of the following hydrologic measurements may need to be undertaken.
• Overland flow:
Hydraulic measurement;
Rainfall/runoff measurement;
Infiltration measurement; and
Drainage basin characterization (including topographic
characteristics, soils and geology, and land use).
• Open channel flow:
Measurement of stage (gaging activities);
Measurement of width, depth, and cross-sectional area;
Measurement of velocity;
13-53
-------
Measurement of channel discharge;
Measurement of channel discharge at controls (e.g., dams and
weirs); and
Definition of flow pathways - solute dispersion studies.
• Closed conduit flow:
Measurement of discharge.
• Lakes and impoundments:
Morphometric mapping;
Bathymetric mapping;
Temperature distributions; and
Flow pathways.
The following references provide descriptions of the measurements described
above.
National Oceanic and Atmospheric Administration. Rainfall Atlas of the U.S.
Viessman.etal., 1977. Introduction to Hydrology.
USGS. 1977. National Handbook of Recommended Methods for Water-Data
Acquisition Chapter 1 (Surface Water) and Chapter 7 (Physical Basin
Characteristics for Hydrologic Analyses).
U.S Department of Interior. 1981. Water Measurement Manual. Bureau of
Reclamation. GPO No. 024-003-00158-9. Washington, D.C.
Chow. 1964. Open Channel Hydraulics. McGraw-Hill. New York, N.Y.
In addition, the following monographs in the Techniques of Water Resources
Investigations series of the USGS (USGS-WSP-1822, 1982) give the reader more
detailed information on techniques for measuring discharge and other
characteristics of various water bodies and hydrologic conditions:
13-54
-------
Benson and Dalrymple. 1967. General Field and Office Procedures for Indirect
Discharge Measurement.
Bodhaine, 1968. Measurement of Peak Discharge at Culverts by Indirect
Methods. USGS-TWI-03-AS.
Buchanan and Somers. 1968. Stage Measurements at Gaging Stations.
Carter and Davidian. 1968. General Procedure for Gaging Streams. USGS-TWI-
03-AL
13.6.2 Sampling of Surface Water, Runoff, Sediment, and Biota
13.6.2.1 Surface Water
The means of collecting water samples is a function of the classification of the
water body, as discussed in Section 13.3.3.1. The following discussion treats lakes
and impoundments separately from streams and rivers although, as indicated
below, the actual sampling methods are similar in some cases. Wetlands are
considered an tntergrade between these waters. Stormwater and snowmelt runoff
is also treated as a separate category (Section 13.6.2.2). Although estuaries also
represent somewhat of an intergrade, estuary sampling methods are similar to
those for large rivers and lakes.
13.6.2.1.1 Streams and Rivers
These waters represent a continuum from ephemeral to intermittent to
perennial. Streams and rivers may exhibit some of the same characteristics as lakes
and impoundments. The degree to which they are similar is normally a function of
channel configuration (e.g., depth, cross sectional area and discharge rate). Larger
rivers are probably more similar to most lakes and impoundments, with respect to
sampling methods, than to free-flowing headwater streams. In general, however,
streams and rivers exhibit a greater degree of mixing due to their free-flowing
characteristics than can be achieved in lakes and impoundments. Mixing and
dilution of inflow can be slow to fast, depending on the point of discharge to the
stream or river and the flow conditions.
13-55
-------
Stream and river sampling methods do not differ appreciably from those
outlined in the following section (Lakes and Impoundments). However, the
selection of monitoring stations must consider additional factors created by
differential flow velocities within the stream cross section. Strong currents and
turbulence as a result of channel configuration may affect the amount of mixing
and the distribution of contaminants in the stream. The reader may wish to refer to
the references provided in Section 13.3.1 for a discussion of the manner in which
differential velocities are handled in stream gaging studies to obtain representative
discharge measurements.
13.6.2.1.2 Lakes and Impoundments
These waters are, by definition, areas where flow velocity is reduced, limiting
the circulation of waters from sources such as discharging streams or ground water.
They often include a shoreline wetland where water circulation is slow, dilution of1
inflowing contaminants is minimal, and sediments and plant life.become significant
factors in sampling strategies. The deeper zones of open water may be vertically
stratified and subject to periodic turnover, especially in temperate climates.
Sampling programs should be designed to obtain depth-specific information as well
as to characterize seasonal variations.
Access to necessary monitoring stations may be impeded by both water depth
and lush emergent or floating aquatic vegetation, requiring the use of a floating
sampling platform or other means to appropriately place the sampling apparatus. It
is common to employ rigid extensions of monitoring equipment to collect surface
samples at distances of up to 30 or 40 feet from the shoreline. However, a boat is
usually the preferred alternative for distances over about six feet. A peristaltic
pump may also be used to withdraw water samples, and has the added advantage
of being able to extract samples to a depth of 20 to 30 feet below the surface.
Many sampling devices are available in several materials. Samples for trace
metals should not be collected in metal bottles, and samples for organics should not
be collected in plastic bottles. Teflon or Teflon-coated sampling equipment,
including bottles, is generally acceptable for both types of constituents. EPA (1982)
and EPA (1986) provide an analysis of the advantages and disadvantages of many
13-56
-------
sampling bottles for specific sampling situations. Detailed descriptions of the use of
dippers/transfer devices, pond samplers, peristaltic pumps, and Kemmerer bottles
are provided by EPA (1984).
Depth-specific samples in lake environments are usually collected with
equipment such as Kemmerer bottles (commonly constructed of brass), Van Dorn
samplers (typically of polyvinyl chloride or PVC construction), or Nansen tubes. The
depth-specific sample closure mechanism on these devices is tripped by dropping a
weight (messenger) down the line. Kemmerer bottles and Nansen tubes may also
be outfitted with a thermometer that records the temperature of the water at the
time of collection.
13.6.2.1.3 Additional Information
Additional information regarding specific surface water sampling methods
may be found in the following general references:
U.S. EPA. 1986. Methods for Evaluating Solid Wastes. EPA/SW-846. GPO No.
955-001-00000-1. Off ice cf Solid Waste. Washington, D.C 20460.
U.S. EPA. 1984. Characterization of Hazardous Waste Sites -- A Methods
Manual: Volume II. Available Sampling Methods. EPA-600/4-84-076. NTIS PB-
168771. Washington, D.C. 20460.
U.S. EPA. 1986. Handbook of Stream Sampling for Wasteload Allocation
Applications. EPA/625/6-83/013.
U.S. EPA. 1982. Handbook for Sampling and Sample Preservation of Water and
Wastewater. NTIS PB 83-124503.
USGS. 1977. National Handbook of Recommended Methods for Water-Data
Acquisition.
I
13-57
-------
13.6.2.2 Runoff Sampling
Runoff resulting from precipitation or snowmelt creates an intermittent
release situation that requires special treatment for effective sampling. The
contaminant release mechanism in runoff situations may be overflow of ponds
containing contaminants or erosion of contaminated soils. Based on an evaluation
of the waste characteristics and the environmental setting, the facility owner or
operator can determine whether waste constituents will be susceptible to this
release mechanism and migration pathway.
Once it has been determined that erosion of contaminated soils is of concern,
the quantity of soil transported to any point of interest, such as the receiving water
body, can be determined through application of an appropriate modification of the
Universal Soil Loss Equation (USLE). The USLE was initially developed by the U.S.
Department of Agriculture, Agricultural Stabilization and Conservation Service
(ASCS) to assist in the prediction of soil loss from agricultural areas. The initial'
formula is reproduced below:
A = RKLSCP
where:
A = Estimated annual average soil loss (tons/acre)
R = Rainfall intensity factor
K = Soil credibility factor
L = Slope-length factor
S = Slope-gradient factor
C = Cropping management factor*
P = Erosion control practice factor*
*C and P factors can be assumed to equal unity in the equation if no specific
crop or erosion management practices are currently being employed. Otherwise,
these factors can be significantly less than unity, depending on crop or erosion
control practices.
13-58
-------
Section 2.6 (Soil Contamination) of the Draft Super-fund Exposure Assessment
Manual (EPA, 1987) provides a discussion of the application of a modified USLE to
characterization of releases through soil erosion. This discussion is summarized in
Appendix H (Soil Loss Calculation).
If the potential for a significant contaminant release exists, based on analysis
of the hydrologic situation and waste site characteristics, event samples should be
taken during high runoff periods. In situations where high runoff is predictable,
such as spring runoff or the summer thundershower season, automatic samplers
may be set to sample during these periods. Perhaps the most effective way to
ensure sampling during significant events is to have personnel available to collect
samples at intervals throughout and following the storm. Flow data should be
collected coincident with sample collection to permit calculation of contaminant
loading in the runoff at various flows during the period. Automated sampling
equipment is available that will collect individual samples and composite them
either over time or with flow amount, with the latter being preferred. Flow-1
proportional samplers are usually installed with a flow-measuring device, such as a
weir with a continuous head recorder. Such devices are readily available from
commercial manufacturers and can be rented or leased. Many facilities with an
NPDES discharge permit routinely use this equipment in compliance monitoring.
Automated samplers are discussed in Section 8 of Handbook for Sampling and
Sample Preservation of Water and Wastewater (EPA. 1982) (NTIS PB 83-124503); this
publication also includes other references to automated samplers and a table of
devices available from various manufacturers.
13.6.2.3 Sediment
Sediment is traditionally defined as the deposited material underlying a body
of water. Sediment is formed as waterborne solids (particulates) settle out of the
water column and build up as bottom deposits.
Sedimentation is greatest in areas where the stream velocity decreases, such as
behind dams and flow control structures, and at the inner edge of bends in stream
channels. Sediments also build up where smaller, fast-flowing streams and runoff
discharge into larger streams and lakes. These areas can be important investigative
13-59
-------
areas. Some sections of a streambed may be virtually without sediments. In some
streams or some areas of streams, water velocity may be too fast for sediments to
deposit and actually may scour the bottom, transporting material and depositing it
further downstream. The stream bed in such an area will be primarily rocks and
debris.
In some situations, such as low-flow conditions, the overlying water
temporarily recedes, exposing sediments to the air. Runoff channels, small lakes,
and small streams and rivers may on occasion dry completely. In these cases,
samples can be collected using the same procedures described in the Soils section
(Section 9) of this document.
For this discussion, the definition of sediment will be expanded to include any
material that may be overlain by water at any time during the year. This definition
then includes what may otherwise be considered submerged soils and sludges.
Submerged soils are found in wetlands and marshes. They may be located on the'
margins of lakes, ponds, and streams, or may be isolated features resulting from
collected runoff, or may appear in areas where the ground-water table exists at or
very near the land surface. In any instance they are important investigative areas.
Sludges are included for discussion here because many RCRA facilities use
impoundments for treatment or storage and these impoundments generally have a
sludge layer on the bottom. Sampling these sludges involves much the same
equipment and techniques as would be used for sediments.
There are essentially two ways to collect sediment samples, either by coring or
with grab/dredges. Corers are metal tubes with sharpened lower edges. The corer
is forced vertically into the sediment. Sediments are held in the core tube by friction
as the corer is carefully withdrawn; they can then be transferred to a sample
container. There are many types and modifications of corers available. Some units
are designed to be forced into the sediments by hand or hydraulic pressure; others
are outfitted with weights and fins and are designed to free fall through the water
column and are driven into the sediment by their fall-force.
Corers sample a greater thickness of sediments than do grab/dredges and can
provide a profile of the sediment layers. However, they sample a relatively small
13-60
-------
surface area. Most corers are less than four inches in diameter and are more
commonly two inches in diameter.
Grab/dredges are basically clamshell-type scoops that sample a larger surface
area but offer less depth of penetration. Typical grab/dredge designs are the Ponar,
Eckman, and Peterson versions; each has a somewhat different operating
mechanism and slightly different advantages. Some use spring force to close the
jaws while others are counter-levered like ice tongs.
In sediment sampling, vertical profiling is not normally required because
deposition of hazardous material is often a recent activity in terms of sedimentary
processes. Grab/dredges that sample a greater surface area may be more
appropriate than corers. Similarly, shallow sludge layers contained in surface
impoundments should be sampled with grab/dredges because corer penetration
could damage the impoundment liner, if present. Thicker sludge layers which may
be present in surface impoundments, may be sampled using coring equipment if it is
important to obtain vertical profile information.
Submerged soils are generally easier to sample with a corer, than with a
grab/dredge because vegetation and roots can prevent the grab/dredges from
sealing completely. Under these conditions, most of the sample may wash out of
the device as it is recovered. Corers can often be forced through the vegetation and
roots to provide a sample. In shallow water, which may overlie submerged soils,
sampling personnel can wade through the water (using proper equipment and
precautions) and choose sample locations in the small, clear areas between
vegetative stems and roots.
A wide variety of sampling devices are available for collection of sediment
samples. Each has advantages and disadvantages in a given situation, and a variety
of manufacturers produce different versions of the same device. As with water
sampling, it is important to remember that metal samplers should not be used when
collecting samples for trace metal analysis, and sampling devices with plastic
components should not be used when collecting samples for analysis of organics.
The following references describe the availability and field use of sediment
samplers:
13-61
-------
U.S. EPA. 1982. Handbook for Sampling and Sample Preservation of Water
and Wastewater. Environmental Monitoring and Support Laboratory,
EPA-600/4-82-029. NTIS PB 83-124503.
U.S. EPA. 1985. Methods Manual for Bottom Sediment Sample Collection. NTIS
PB 86-107414.
USGS. 1977, update June 1983. National Handbook of Recommended Methods
for Water-Data Acquisition.
U.S. EPA. 1984. Characterization of Hazardous Waste Sites - A Methods
Manual: Volume II. Available Sampling Methods. EPA-600/4-84-076. NTIS PB
85-168771.
13.6.2.4 Biota '
Collection of biota for constituent analysis (whole body or tissue) may be
necessary to evaluate exposure of aquatic organisms or man to bioaccumulative
contaminants. For the most part, collection should be restricted to representative
fish species and sessile macroinvertebrates, such as mollusks. Mollusks are filter-
feeders; bioaccumulative contaminants in the water column will be extracted and
concentrated in their tissues. Fish species may be selected on the basis of their
commercial or recreational value, and their resultant probability of being consumed
by man or by special status-species of fish or wildlife.
The literature on sampling aquatic organisms is extensive. Most sampling
methods include capture techniques that be collected using sampling bottles (as for
water samples) or nets of appropriate mesh sizes. Periphyton may be most easily
collected by scraping off the substrate to which the organisms are attached. Other
techniques using artificial substrates are available if a quantitative approach is
required. Aquatic macroinvertebrates may be collected using a wide variety of
methods, depending on the area being sampled; collection by hand or using forceps
may be efficient. Grab sampling, sieving devices, artificial substrates and drift nets
may also be used effectively. EPA (1973) provides a discussion of these techniques,
as well as a method comparison and description of data analysis techniques.
13-62
-------
Fish collection techniques may be characterized generally as follows (USGS,
1977):
• Entangling gear:
Gill nets and trammel nets.
• Entrapping gear:
Hoop nets, basket traps, trap nets, and fyke and wing
nets.
• Encircling gear:
Haul seine, purse seine, bay seine, and Danish seine.
• Electroshocking gear:
Boat shockers, backpack shockers, and electric seines.
Selection of sampling equipment is dependent on the characteristics of the water
body, such as size and conditions, the size of the fish to be collected, and the overall '
objectives of the study. Fisheries Techniques (Nielsen and Johnson, 1983) and
Guidelines for Sampling Fish in Inland Waters (Backiel and Welcomme, 1980)
provide basic descriptions of sampling methods and data interpretation from
fisheries studies.
13.6.3 Characterization of the Condition of the Aquatic Community
Evaluation of the condition of aquatic communities may proceed from two
directions. The first consists of examining the structure of the lower trophic levels as
an indication of the overall health of the aquatic ecosystem. With respect to RFI
studies, a healthy water body would be one whose trophic structure indicates that it
is not impacted by contaminants. The second approach focuses on a particular
group or species, possibly because of its commercial or recreational importance or
because a substantial historic data base already exists.
The first approach emphasizes the base of the aquatic food chain, and may
involve studies of plankton (microscopic flora and fauna), periphyton (including
bacteria, yeast, molds, algae, and protozoa), macrophyton (aquatic plants), and
benthic macroinvertebrates (e.g., insects, annelid worms, mollusks, flatworms,
roundworms, and crustaceans). These lower levels of the aquatic community are
13-63
-------
studied to determine whether they exhibit any evidence of stress. If the community
appears to have been disturbed, the objective is to characterize the source(s) of the
stress and, specifically, to focus on the degree to which the release of waste
constituents has caused the disturbance or possibly exacerbated an existing
problem. An example of the latter would be the further depletion of already low
dissolved oxygen levels in the hypolimnion of a lake or impoundment through the
introduction of waste with a high COD and specific gravity.
The sampling methods referenced in Section 13.6.2.4 may be adapted (by
using them in a quantitative sampling scheme) to collect the data necessary to
characterize aquatic communities. Hynes (1970) and Hutchinson (1967) provide an
overview of the ecological structure of aquatic communities.
Benthic macroinvertebrates are commonly used in studies of aquatic
communities. These organisms usually occupy a position near the base of the food
chain. Just as importantly, however, their range within the aquatic environment is
restricted, so that their community structure may be referenced to a particular
stream reach or portion of lake substrate. By comparison, fish are generally mobile
within the aquatic environment, and evidence of stress or contaminant load may
not be amenable to interpretation with reference to specific releases.
The presence or absence of particular benthic macroinvertebrate species,
sometimes referred to as "indicator species," may provide evidence of a response to
environmental stress. Several references are available in this regard. For more
information, the reader may consult Selected Bibliography on the Toxicology of the
Benthic Invertebrates and Periphyton (EPA. 1984).
A "species diversity index" provides a quantitative measure of the degree of
stress within the aquatic community, and is an example of a common basis for
interpretation of the results of studies of aquatic biological communities. The
following equation (the Sannon-Wiener Index) demonstrates the concept of the
diversity index:
s
H = I (PiMlogjPi)
13-64
-------
where:
H ss species diversity index
s = number of species
Pi = proportion of total sample belonging to the [th species
Measures of species diversity are most useful for comparison of streams with similar
hydrologic characteristics or for the analysis of trends over time within a single
stream. Additional detail regarding the application of other measures of
community structure may be found in the following references:
U.S. EPA. 1973. Biological Field and Laboratory Methods for Measuring the
Quality of Surface Water and Effluents.
USGS. 1977, Update May, 1983. National Handbook of Recommended
Methods of Water-Data Acquisition.
Curns, J. Jr., and K.L Dickson, eds. 1973. ASTMSTP528: Biological Methods
for the Assessment of Water Quality. American Society for Testing and
Materials. STP528. Philadelphia, PA.
The second approach to evaluating the condition of an aquatic community is
through selective sampling of specific organisms, most commonly fish, and
evaluation of standard "condition factors" (e.g., length, weight, girth). In many
cases, receiving water bodies are recreational fisheries, monitored by state or
federal agencies. In such cases, it is common to find some historical record of the
condition of the fish population, and it may be possible to correlate operational
records at the waste management facility with alterations in the status of the fish
population.
Sampling of fish populations to evaluate condition factors employs the same
methodologies referenced in Section 13.6.2.4. Because of the intensity of the effort
usually associated with obtaining a representative sample of fish, it is common to
coordinate tissue sampling for constituent analysis with fishery surveys.
13-65
-------
13.6.4 Bioassay Methods
The purpose of a bioassay, as discussed is more detail in Section 13.4.6.3, is to
predict the response of aquatic organisms to specific changes within the
environment. In the RFI context, a bioassay may be used to predict the potential
adverse environmental effects of releases to surface water. Thus, bioassay is not
generally considered to be an environmental characterization or monitoring
technique. As indicated below, bioassay may be required for Federal water quality
programs or state programs, especially where stream classification (e.g., warm-
water fishery, cold-water fishery) is involved.
Bioassays may be conducted on any aquatic organism including algae,
periphyton, macroinvertebrates, or fish. Bioassay includes two main techniques,
acute toxicity tests and chronic toxicity tests. Each of these may be done in a
laboratory setting or using a mobile field laboratory. Following is a brief discussion
of acute and chronic bioassay tests. .
Acute Toxicitv Tests-Acute toxicity tests are used in the NPDES permit program to
identify effluents containing toxic wastes discharged in toxic amounts. The data are
used to predict potential acute and chronic toxicity in the receiving water, based on
the LC50 and appropriate dilution, and application of persistence factors. Two types
of tests are used; static and flow-through. The selection of the test type will depend
on the objectives of the test, the available resources, the requirements of the test
organisms, and effluent characteristics. Special environmental requirements of
some organisms may preclude static testing.
It should be noted that a negative result from an acute toxicity test with a
given effluent sample does not preclude the presence of chronic toxicity, nor does it
negate the possibility that the effluent may be acutely toxic under different
conditions, such as variations in temperature or contaminant loadings.
There are many sources of information relative to the performance of acute
bioassays. Methods for Measuring the Acute Toxicitv of Effluents to Freshwater and
Marine Organisms (Peltier and Weber, 1985) provides a comprehensive treatment
of the subject.
13-66
-------
Chronic Toxicitv Tests-Chronic toxicity tests may include measurement of effluent
effects on growth and reproductive success. These tests usually require long periods
of time, depending on the life cycles of the test organisms. Chronic bioassays are
generally relatively sophisticated procedures and are more intensive in terms of
manpower, time and expense than are acute toxicity tests. The inherent complexity
of these tests dictate careful planning with the regulatory agency prior to initiation
of the work. Methods for Measuring the Chronic Toxicitv of Effluents to Aquatic
Organisms (Horning and Weber, 1985) is a companion volume to the methods
document noted above, and contains method references for chronic toxicity tests. A
discussion of bioassay procedures is also provided in Protocol for Bioassessment of
Hazardous Waste Sites, NTIS PB 83-241737. (Tetra Tech, 1983).
Chronic toxicity tests are also used in the NPDES permit program to identify
and control effluents containing toxic wastes in toxic amounts.
i
13.7 Site Remediation I
Although the RFI Guidance is not intended to provide detailed guidance on
site remediation, it should be recognized that certain data collection activities that
may be necessary for a Corrective Measures Study may be collected during the RFI.
EPA has developed a practical guide for assessing and remediating contaminated
sites that directs users toward technical support, potential data requirements and
technologies that may be applicable to EPA programs such as RCRA and CERCLA.
The reference for this guide is provided below.
U.S. EPA. 1988. Practical Guide for Assessing and Remediating Contaminated
Sites. Office of Solid Waste and Emergency Response. Washington, D.C.
20460.
The guide is designed to address releases to ground water as well as soil,
surface water and air. A short description of the guide is provided in Section 1.2
(Overall RCRA Corrective Action Process), under the discussion of Corrective
Measures Study.
13-67
-------
13.8 Checklist
RFI CHECKLIST - SURFACE WATER
Site Name/Location
Type of Unit
1. Does waste characterization include the following information? (Y/N)
• Constituents of concern
• Concentrations of constituents
• Mass of the constituent
• Physical state of waste (e.g. .solid, liquid, gas)
• Water solubility
• Henry's Law Constant
• Octanol/Water Partition Coefficient (Kow)
• Bioconcentration Factor (BCF)
• Adsorption Coefficient (Koc)
• Physical, biological, and chemical degradation
2. Does unit characterization include the following information? (Y/N)
• Age of unit
• Type of unit
• Operating practices
• Quantities of waste managed
• Presence of cover
• Dimensions of unit
• Presence of natural or engineered barriers
• Release frequency
• Release volume and rate
• Non-point or point source release
• Intermittent or continuous release
13-68
-------
RFI CHECKLIST - SURFACE WATER (Continued)
3. Does environmental setting information include the following? (Y/N)
• Areal extent of drainage basin
• Location and interconnection of all streams, lakes
and other surface water features
• Flow identification as ephemeral, intermittent or perennial
• Channel alignment, gradient and discharge rate
• Flood and channel.control structures
• Source of lake and impoundment water
• Lake and impoundment depths and surface area
• Vertical temperature stratification of lakes and impoundments
• Wetland presence and role in basin hydrology
• NPDES and other discharges
• USGS gaging stations or other existing flow monitoring systems
• Surface water quality characteristics
• Average monthly and annual precipitation values
• Average monthly temperature
• Average monthly evaporation potential estimates
• Storm frequency and severity
• Snowfall and snow pack ranges
4. Have the following data on the initial phase of the release
characterization been collected? (Y/N)
• Monitoring locations
• Monitoring constituents and indicator parameters
• Monitoring frequency
• Monitoring equipment and procedures ._
• Concentrations of constituents and locations
at which they were detected
• Background monitoring results
13-69
-------
RFI CHECKLIST-SURFACE WATER (Continued)
(Y/N)
• Hydrologic and biomonitoring results
• Inter-media transfer data
• Analyses of rate and extent of contamination .
5. Have the following data on the subsequent phase(s) of the release
characterization been collected? (Y/N)
• New or relocated monitoring locations
• Constituents and indicators added or deleted for monitoring
• Modifications to monitoring frequency, equipment
or procedures
• Concentrations of constituents and locations at which
they were detected
• Background monitoring results
• Hydrologic and biomonitoring results
• Inter-media transfer data
• Analyses of rate and extent of contamination
13-70
-------
13.9 References
American Public Health Association, (APHA). 1985. Standard Methods for the
Examination of Water and Wastewater. 16th Edition. American Public Health
Association, Washington, D.C.
Backiel,T.,and R. Welcomme. 1980. Guidelines for Sampling Fish in Inland Waters.
EIFAC Technical Paper No. 33. Food and Agriculture Organization of the
United Nations, Rome, Italy.
Benson, M.A., and T. Dalrymple. 1967. General Field and Office Procedures for
Indirect Discharge Measurement. Techniques of Water Resources
Investigations series. U.S. Geological Survey, Reston, VA.
Bodhaine, G. L 1968. Measurement of Peak Discharge at Culverts by Indirect.
Methods. Techniques of Water Resources Investigations Series. U.S. Geological i
Survey, Reston, VA.
Brandes, R., B. Newton, M. Owens, and E. Southerland. 1985. The Technical Support
Document for Water Quality-Based Toxics Control. EPA-440/4-85-032. Office
of Water Enforcement and Permits. Washington, D.C. 20460.
Buchanan, T. J., and W. P. Somers. 1968. Stage Measurement at Gaging Stations.
Techniques of Water Resources Investigations Series. U.S. Geological Survey,
Reston, VA.
Cairns, J. Jr., and K. L. Dickson, eds. 1973. Biological Methods for the Assessment of
Water Quality (STP 528). American Society for Testing and Materials,
Philadelphia, PA.
Callahan, M., M. Slimak, N. Gabel, I. May, etal. 1979. Water-Related Environmental
Fate of 129 Priority Pollutants. Volumes I & II. EPA 440/4-79-029a/b.
Monitoring and Data Support Division. NTIS 029A/80-204373 and 029B/80-
204381.Washington, D.C. 20460.
13-71
-------
Carter, R. W., and J. Davidian. 1968. General Procedure for Gaging Streams.
Techniques of Water Resources Investigations Series. U.S. Geological Survey,
Reston, VA.
Chow, V.T. 1964. Open-Channel Hydraulics. McGraw-Hill. New York, NY.
Cole, G. A. 1975. Textbook of Limnology. The C. V. Mosbv Company. St. Louis. MO.
Cowardin, L M., V. Carter, F. C. Golet, and E. T. Laftoe. 1979. Classification of
Wetlands and Deepwater Habitats of the United States. U.S. Fish & Wildlife
Service. NTIS PB 80-168784. Washington, D.C.
Cummins, K. W. and N. A. Wilgbach. 1985. Field Procedures for Analysis of
Functional Feeding GroUPS of Stream Macroinvertebrates. Contribution 1611.
Appalachian Environmental Laboratory, University of Maryland.
i
Hilsenhoff, W. L 1982. Using a Biotic Index to Evaluate Water Quality in Streams.1
Technical Bulletin No. 132. Department of Natural Resources. Madison, Wl.
Horning, W., and C. I. Weber. 1985. Methods for Measuring the Chronic Toxicitv of
Effluents to Aquatic Organisms. U.S. EPA, Office of Research and
Development. Cincinnati, OH.
Hutchinson, G. E. 1957. A Treatise on Limnology: Volume I. Geography. Physics,
and Chemistry. John Wiley & Sons, Inc. New York, NY.
Hutchinson, G. E. 1967. A Treatise on Limnology: Volume II. Introduction to Lake
Biology and Limnoplankton. John Wiley & Sons, Inc. New York, NY.
Hynes, H. B. N. 1970. The Ecology of Running Waters. University of Toronto Press.
Toronto, Ontario.
Lyman, W. J., W. F. Riehl, and D. H. Rosenbaltt. 1982. Handbook of Chemical
Property Estimation Methods. McGraw-Hill. New York, NY.
13-72
-------
Mabey, W.( and T. Mill. 1978. "Critical Review of Hydrolysis of Organic Compounds
in Water Under Environmental Conditions." Journal of Environmental
Chemistry. Vol. 7, No. 2.
Mabey, W. R., J. H. Smith, R. T. Podall, et al. 1982. Aquatic Fate Process Data for
Organic Priority Pollutants. EPA 440/4-81-014. Washington, D.C. 20460.
Mills, W. B., 1985. Water Quality Assessment: A Screening Procedure for Toxic and
Conventional Pollutants in Surface and Ground Water: Parts land 2. EPA
600/6-85-002, a, b. NTIS PB 83-153122 and NTIS PB 83-153130. U.S. EPA, Office
of Research and Development. Athens, GA.
National Oceanic and Atmospheric Administration. Rainfall Atlas of the U.S.
Neely, W. B. 1982. "The Definition and Use of Mixing Zones". Environmental
Science and Technology 16(9): 520A-521 A. i
t
Neely, W. G., and G. E. Blau, eds. 1985. Environmental Exposure from Chemicals.
Volume I. CRC Press. B jca Raton, FL
Nielsen, L A., and D. L Johnson, eds. 1983. Fisheries Techniques. The American
Fisheries Society. Blacksburg, VA, 468 pp.
Peltier, W. H., and C. I. Weber. 1985. Methods for Measuring the Acute Toxicitv of
Effluents to Freshwater and Marine Organisms. EPA 600/4-85/013. NTIS PB 85-
205383. U.S.EPA, Environmental Monitoring and Support Laboratory, Office
of Research and Development. Cincinnati, OH.
Stumm, W. and J. J. Morgan. 1982. Aquatic Chemistry. 2nd Edition. Wiley
Interscience. New York, NY.
TetraTech. 1983. Protocol for Bioassessment of Hazardous Waste Sites. U.S. EPA.
NTIS PB 83-241737. Washington, D.C. 20460.
U.S. Department Of Interior. 1981. Water Measurement Manual. Bureau of
Reclamation. GPO No. 024-003-00158-9. Washington, D.C.
13-73
-------
U.S. EPA. 1973. Biological Field and Laboratory Methods for Measuring the Quality
of Surface Water and Effluents. EPA-67014-73-001. Office of Research and
Development. Washington, D.C. 20460.
U.S. EPA. 1982. Handbook for Sampling and Sample Preservation of Water and
Wastewater. Environmental Monitoring and Support Laboratory. EPA-600/4-
. 82-029. NTIS PB 83-124503. Washington, D.C.
U.S. EPA. 1984. Characterization of Hazardous Waste Sites - A Wetlands Manual-
Volume II - Available Sampling Methods. EPA-600/4-84-076. NTIS PB 85-
168771. Washington, D.C. 20460.
U.S. EPA. 1984. Selected Bibliography on the Toxicology of the Benthic
Invertebrates and Periphyton. Environmental Monitoring and Support
Laboratory. NTIS PB 84-130459. '
I
U.S. EPA. 1985. Methods Manual for Bottom Sediment Sample Collection. NTIS
PB86-107414. Washington, D.C. 20460.
U.S. EPA. 1987. Draft Superfund Exposure Assessment Manual. Office of
Emergency and Remedial Response. Washington, D.C. 20460.
U.S. EPA. 1986. Test Methods for Evaluating Solid Waste. EPA/SW-846. GPO
No.9 55-001-00000-1. Officeof Solid Waste. Washington, D.C. 20460.
U.S. EPA. 1986. Handbook of Stream Sampling for Wasteload Allocation
Applications. EPA/625/6-83/013.
USGS. 1977. National Handbook of Recommended Methods for Water-Data
Acquisition. U.S. Geological Survey. Office of Water Data Coordination. U.S.
Government Printing Office. Washington, D.C.
Veith, G., Macey, Petrocelli and Carroll. 1980. An Evaluation of Using
Partition.Coefficients and Water Solubility to Estimate Biological
13-74
-------
Concentration Factors for Organic Chemicals in Fish. Proceedings, ASTM 3rd
Symposium on Aquatic Toxicity. ASTM STP 707.
Viessman, W., Jr., W. Knapp. G. L Lewis, and T. E. Harbaugh. 1977. Introduction to
Hydrology. 2nd Edition. Harper and Row, Publishers, New York, NY.
13-75
-------
APPENDIX G
AIR RELEASE SCREENING ASSESSMENT METHODOLOGY
-------
DRAFT FINAL
(Revised)
AIR RELEASE
SCREENING ASSESSMENT
METHODOLOGY
MAY 1989
-------
TABLE OF CONTENTS
Section Title
1.0 Introduction 1-1
2.0 Screening Methodology 2-1
2.1 Overview 2-2
2.2 Step 1 -SourceCharacterization Information 2-5
2.3 Step 2 - Release Constituent Surrogates 2-7
2.4 Step 3 - Emission Estimates 2-9
2.5 Step4-Concentration Estimates 2-14
2.6 Step 5 - Health Criteria Comparisons 2-17
3.0 Example Applications 3-1
3.1 Case Study A 3-1
3.2 Case Study B 3-6
4.0 References 4-1
l
Appendix A Background Information
Appendix B Release Constituent Surrogate Data
Appendix C Emission Rate Estimates - Disposal Impoundments
Appendix D Emission Rate Estimates - Storage Impoundments
Appendix E Emission Rate Estimates - Oil Films on Storage
Impoundments
Appendix F Emission Rate Estimates - Mechanically Aerated
Impoundments
Appendix G Emission Rate Estimates - Diffused Air Systems
Appendix H Emission Rate Estimates - Land Treatment (after tilling)
Appendix I Emission Rate Estimates - Oil Film Surfaces on Land
Treatment Units
Appendix J Emission Rate Estimates - Closed Landfills
Appendix K Emission Rate Estimates - Open Landfills
Appendix L Emission Rate Estimates - Wastepiles
Appendix M Emission Rate Estimates - Fixed Roof Tanks
Appendix N Emission Rate Estimates- Floating Roof Tanks
-------
TABLE OF CONTENTS (Continued)
Section Title
Appendix O Emission Rate Estimates - Variable Vapor Space Tanks
Appendix P Emission Rate Estimates - Particles from Storage Piles
Appendix Q Emission Rate Estimates - Particles from Exposed, Flat,
Contaminated Areas
Appendix R Dispersion Estimates
Appendix S Emission Rate Estimation Worksheets
HI
-------
LIST OF FIGURES
Number Page
2-1 Screening Methodology Overview 2-3
2-2 Step 1 - Obtain Source Characterization Information 2-6
2-3 Step 2-Select Release Constituents and Surrogates 2-8
2-4 Step 3 - Calculate Emission Estimates 2-10
2-5 Step 3 - Calculate Emission Estimates (Alternative 2-11
Approach)
2-6 Step4-CalculateConcentration Estimates 2-15
2-7 Step 4 - Calculate Concentration Estimates (Alternative 2-16
Approach)
2-8 Step 5-Compare Results to Health-Based Criteria 2-18
LIST OF EXHIBITS
Number
2-1 Ratio of Scaling Estimates to CHEMDAT6 Emission Rate
Modeling Results
3-1 Table S-2 Emission Rate Estimation Worksheet - Storage
Impoundment
3-2 Table R-1 Concentration Estimation Worksheet - Unit
Category: Storage Impoundment
3-3 Table S-8 Emission Rate Estimation Worksheet - Closed
Landfills
3-4 Table S-8 Emission Rate Estimation Worksheet - Closed
Landfill
3-5 Table R-1 Concentration Estimation Worksheet - Unit
Category: Closed Landfill
3-6 Table R-1 Concentration Estimation Worksheet - Unit
Category: Closed Landfill
2-20
3-2
3-5
3-8
3-9
3-10
3-11
IV
-------
1.0 INTRODUCTION
A screening method has been developed for evaluating which waste management
units have air releases warranting further investigation under a RCRA Facility
Investigation (RFI). This method can be used as an intermediate step between the
general qualitative determination of the RCRA Facility Assessment (RFA) regarding
identification of air emissions that warrant an RFI, and the actual performance of a
complicated and costly RFI. Specifically, this screening methodology provides a basis
for identifying air releases with the potential to have resulted in off-site exposures
that meet or exceed health-based criteria in the RFI Guidance.
This screening methodology has been developed as a technical aid for routine use
by EPA Regional and State staff who may not be familiar with air release
assessments. However, it should also be considered a resource available to prioritize
waste management units which may warrant the conduct of an RFI for the air
media. Alternative resources (e.g., available air monitoring data, more
sophisticated modeling analyses, judgmental factors) may also provide important
input to the RFI decision-making process.
The screening methodology itself is explained in Section 2 and example applications
of it are presented in Section 3. A discussion of background information that
addresses the technical basis for the air release screening methodology is presented
in Appendix A.
1-1
-------
2.0 SCREENING METHODOLOGY
This section presents the air release screening assessment methodology. This
methodology can be used as a transition between the general qualitative
determination made in the RFA regarding air emissions that warrant an RFI, and the
actual performance of an RFI.
The primary (recommended) screening approach involves the application of
available emission rate models and dispersion models. An alternative approach
involves the use of technical aids based on scaling modeling results for a limited set
of source scenarios.
The screening methodology for releases of organics is based on using the
CHEMDAT6 air emission models, available from EPA's Office of Air Quality Planning
and Standards (OAQPS), (U.S. EPA, December 1987). Specifically, the following unit
categories are directly addressed in this section:
• Disposal impoundments
• Storage impoundments
• Oil Films on Storage Impoundments
• Mechanically Aerated Impoundments
• Diffused Air Systems
• Land treatment (emissions after tilling)
• Oil Film Surfaces on Land Treatment Units
• Closed landfills
• Open landfills
• Wastepiles
The alternative approach presented in this section involves scaling the emission rate
results from numerous source scenarios that have been modeled using CHEMDAT6.
These scaling computations can become tedious if numerous source scenarios are
evaluated. In addition, the direct use of CHEMDAT6 models will provide more
representative unit-specific emission estimates. Therefore, it is strongly
recommended that EPA Regional and State agency staff develop a capability to use
CHEMDAT6 directly to model unit-specific and facility-specific scenarios.
2-1
-------
CHEMDAT6 has been developed for use on a microcomputer using LOTUS
spreadsheet software; therefore, these models can easily be used by staff familiar
with LOTUS applications. However, the basic strategy described in this section to
estimate ambient concentrations can still be successfully used even without using
LOTUS.
The screening methodology for organic emissions from storage tanks is based on
emission factors in EPA's AP-42, "Compilation of Air Pollutant Emission Factors"
(U.S. EPA, September 1985). The following categories of tanks are addressed:
• Fixed roof tanks
• Floating roof tanks
• Variable vapor space tanks.
Open tanks should be assessed using the methodology for storage impoundments.
i
The screening methodology for particulate matter releases from wind erosion of
storage piles and batch dumping and loader activity on the pile is based on emission
factors in EPA's AP-42 (U.S. EPA, September 1985). The screening methodology for
particulate matter releases from wind erosion of flat, exposed, contaminated
surface areas is based on emission factors in EPA's "Fugitive Emissions from
Integrated Iron and Steel Plants" (U.S. EPA, March 1978). The EPA-OAQPS is
currently developing guidance regarding particulate emissions for treatment,
storage, and disposal facilities.
2.1 Overview
The air release screening assessment methodology involves applying emission rate
and dispersion results to estimate long-term ambient concentrations at receptor
locations for comparison to health-based criteria. The methodology consists of five
steps as follows (see Figure 2-1):
• Step 1 - Obtain Source Characterization Information: This information
(e.g., unit size, operational schedule) is needed to define the emission
potential of the specific unit.
2-2
-------
FIGURE 2-1
SCREENING METHODOLOGY OVERVIEW
RFA
1
Stem
Obtain Source
Characterization
Information
Step 2
Select Release
Constituents and
Surrogates
Calculate
Emission Estimates
Step 4
Calculate Concentration
Estimates
StepS
Compare Results to
Health-Based Criteria
1
Input to
Decision on Need for
RFI-Air Media
2-3
-------
• Step 2- Select Release Constituents and Surrogates: The primary
approach involves using the actual physical/chemical properties for all
unit-specific constituents for emission modeling purposes. The
alternative (scaling) screening approach uses a limited set of constituents
or surrogates to represent a wide range of potential release constituents.
This surrogate approach significantly simplifies the screening assessment
process.
• Step 3 - Calculate Emission Estimates: The primary approach involves the
use of emission rate models based on unit-specific source conditions.
Modeling results of emission rates for a wide range of source conditions
are also presented in Appendices C through Q. As an alternative
approach, these modeling results can be interpolated to estimate an
emission rate specific to the unit.
i
• Step 4 - Calculate Concentration Estimates: Emission rates from Step 3
are used to calculate concentration estimates at receptor locations of
interest. The primary approach involves the application of dispersion
models based on site-specific meteorological conditions. As an
alternative approach, dispersion conditions are accounted for by use of
modeling results available in Appendix R for typical annual
meteorological conditions.
• Step 5 - Compare Concentration Results to Health-Based Criteria:
Concentration results from Step 4 can be compared to constituent-
specific health-based criteria provided in the RFI Guidance.
For some applications, Step 4 (Calculate Concentration Estimates) will not warrant
the use of emission models because it can be assumed that all the volatile wastes
handled will eventually be emitted to the air. This assumption is generally
appropriate for highly volatile organic compounds placed in a disposal unit like a
surface impoundment. In these cases, the air emission rate can be assumed to be
equivalent to the disposal rate, so that an emission rate model may not be required.
This assumption is valid because of the long-term residence time of wastes in the
disposal units. In open units like surface impoundments, a substantial portion of
the volatile constituents will frequently be released to the atmosphere within
2-4
-------
several days. However, for more complex situations (e.g., storage or treatment
units where total volatilization of the constituents is not expected), air emission
models can be used to obtain a more refined long-term release rate.
Results from the air release screening assessment, using the above steps, will
provide input to decisions on the need for an RFI for the air media. They can also be
used to prioritize air emission sources at a facility (i.e., by identification of the major
onsite air emission sources) as well as to prioritize the total release potential at
candidate facilities.
2.2 Step 1 • Source Characterization Information
Implementation of the air release screening assessment methodology involves
collecting source characterization information, as illustrated in Figure 2-2.
Specifically, this involves completion of Column 2 of unit-specific Emission Rate
Estimation Worksheets (included in Appendix S) as specified in Figure 2-2.
Parameters in Column 2 of the worksheet represent standard input used by the
CHEMDAT6 air emission models or input to the AP-42 emission equations. Source
characterization information should be available from the RFA but it may be
necessary to request additional information from the facility owner or operator on
an ad hoc basis.
Additional worksheets should be completed for each unit to be evaluated. Similar
units can be grouped together and considered as one area source to simplify the
assessment process. For example, several contiguous landfills of similar design could
be evaluated efficiently as one (combined) source.
Completeness and quality of the source characterization information are very
important and, as previously stated, directly affect the usefulness of the screening
assessment results. Certain source characterization parameters are considered
critical inputs to the screening assessment. These critical input parameters are
needed to define the total mass of constituents in the waste input to the unit being
evaluated or the potential for release of particles less than 10 microns. These
parameters have been identified in the unit-specific worksheet (Tables S-1 through
S-13 for VO sources and Tables S-14 and 15 for participate sources).
2-5
-------
FIGURE 2-2
STEP 1 - OBTAIN SOURCE CHARACTERIZATION INFORMATION
RFA
Complete Column 2 of Unit-Specific Emission Rate Estimation Worksheet:
• Disposal impoundment - Table S-1
• Storage impoundment/open tank-
Table S-2
• Oil film on storage impoundment -
Table S-3
• Mechanically aerated impoundment
Table S-4
• Diffused air system - Table S-5
• Land treatment (emissions after
• tilling)- Table S-6
• Oil film surface on land treatment
unit-Table S-7
Closed landfill-TableS-8
Open landfill-TableS-9
Wastepile- Table S-10
Fixed roof tank - Table S-11
Floating roof tank-Table S-12
Variable vapor space tank -
Table S-13
Storage pile (participates) -
Table S-14
Exposed, flat, contaminated area
(particulates)TableS-15
Complete Column 2 of additional worksheets for each unit to be evaluated
(similar units can be grouped as one area source).
Select typical and/or reasonable worst-case values specified in Appendices C-M if
values for input parameters are not available.
Disposal impoundment - Table C-1
Storage impoundment/open tank-
Table D-1
Oil film on storage impoundment -
Table E-1
Mechanically aerated impoundment -
Table F-1
Diffused air system - Table G-1
Land treatment (emissions after
tilling)-Table H-1
Oil film surface on land treatment
unit-Table 1-1
Closed landfill-Table J-1
Open landfill-Table K-1
Wastepile-Table L-1
Fixed roof tank-Table M-1
Floating roof tank -Table N-1
Variable vapor space tank -
Table p-1
Storage pile (particulates)-
Table P-1
Exposed, flat, contaminated area
(particulates)- Table Q-1
T
Step 2-
Select Release Constituents and
Surrogates
2-6
-------
Unit-specific values for some of the source characterization parameters may be
difficult to determine. For example, air porosity values of the fixed waste are
needed for evaluating emissions from open landfills, closed landfills, and
wastepiles, and total porosity values of the fixed waste are needed to evaluate
emissions from open landfills and wastepiles. However, unit-specific data are
typically not available for these parameters. If unit-specific values for input
parameters are not available, typical and/or reasonable worst-case values should be
selected from the range of values specified in Appendices C through Q.
Selection of source scenario input data should be based on realistic physical and
chemical limitations. For example, the waste concentration value for a constituent
should not exceed the constituent-specific solubility in water.
2.3 Step 2 • Release Constituent Surrogates
The primary approach involves using the actual physical/chemical properties for all
unit-specific constituents for emission modeling purposes. The alternative
screening approach (scaling) uses a limited set of constituents or surrogates.
A limited set of surrogates is used to represent the constituents of concern in this
alternative screening method to represent a wide range of potential release
constituents. This significantly simplifies the screening assessment process since the
list of potential air release constituents included in the RFI Guidance is extensive.
Selection of appropriate source release constituent surrogates is illustrated in
Figure 2-3. Table B-3 presents the appropriate surrogate to be used for each
constituent of concern. This step is not used in screening for particle emissions from
storage piles and exposed areas.
Table B-3 of Appendix B, presents the appropriate surrogate to be used for each
constituent of concern. Two subsets of surrogates are presented in Appendix B. The
first subset is applicable to emissions that can be estimated based on Henry's Law
Constant (i.e., applicable for low concentrations, less than 10 percent, of wastes in
aqueous solution). Surrogates based on Henry's Law Constant are appropriate for
units like storage and disposal impoundments. Henry's Law Constant surrogates are
presented in Table B-1.
2-7
-------
FIGURE 2-3
STEP 2 - SELECT RELEASE CONSTITUENTS AND SURROGATES
Source Characterization Information
Impoundments
(Organic Releases)
Select
appropriate'
surrogate
subset.
Other Units
(Organic Releases)
Surrogate subset
based on Henry's
Law Constant (see
Table B-1)
Paniculate Releases
Surrogate subset
based on Raoult's
Law (see Table B-2)
Select
appropriate
constituents to
represent
release.
Alternative Approach
Primary Approach
Use all constituents to
evaluate unit.
Limit evaluations to release
constituent(s) that represent
reasonable worst-case
conditions.
Identify surrogates which
correspond to release
constituents
(Table B-3).
Step 3-
Calculate
Emission Estimates
2-8
-------
The second subset is applicable to emissions that can be estimated based on Raoult's
Law. Raoult's Law predicts the behavior of most concentrated mixtures of water
and organic solvents (i.e., solution with over 10 percent solute). Surrogates based
on Raoult's Law are appropriate for units like landfills, wastepiles, land treatment
units and storage tanks. Raoult's Law surrogates are listed in Table B-2.
It is also necessary to select surrogates from the appropriate subset (i.e., from the
Henry's Law Constant or Raoult's Law subset selected) to represent release
constituents of interest. The primary approach is to use all surrogates from the
appropriate subset to evaluate the unit. This approach will provide a
comprehensive data base for the screening assessment. An alternative approach is
to select release constituent(s)/surrogate(s) that represent reasonable worst-case
conditions. Release constituents having the most restrictive health-based criteria
and those having high volatility are frequently associated with these reasonable
worst-case (long-term) release conditions.
i
2.4 Step 3 • Emission Estimates
Two approaches for calculating emission estimates are identified in Figure 2-4. The
primary approach involves the calculation of unit-specific emission rates based on
available models (e.g., CHEMDAT6, et cetera). This approach is recommended for
most applications.
The alternative approach involves the calculation of emissions by applying scaling
factors to emission modeling results presented in Appendices C through Q for a
limited set of source scenarios. This approach is appropriate when a rapid
preliminary estimate is needed and modeling resources are not available. However,
the primary approach will provide more representative unit-specific emission
estimates.
Specific instructions for implementing the alternative emission estimation approach
are presented in Figure 2-5.
Emission rate modeling results for a wide range of source scenario conditions are
presented in Appendices C through Q to facilitate implementation of the
alternative emission estimation approach. These available modeling results can be
2-9
-------
FIGURE 2-4
STEP 3 • CALCULATE EMISSION ESTIMATES
Source Characterization Information/Constituent Surrogates
Primary approach -
calculate emissions by
using models available
from the following:
• CHEMDAT6
• Hazardous Waste
Treatment Storage and
Disposal Facilities
(TSDF)- Air Emission
Models
• AP-42
• Other standard EPA
technical documents
Alternative approach -
calculate emissions by
applying scaling factors to
emission modeling results
available for limited set of •
so u rce see n a r i os (see .
Figure 2-5). '
Step 4-
Calculate
Concentration Estimates
2-10
-------
FIGURE 2-5
STEP 3 - CALCULATE EMISSION ESTIMATES (ALTERNATIVE APPROACH)
Source Characterization Information/Constituent Surrogates
Obtain Emission Rate Estimation Worksheets (as selected in Step 1):
• Disposal impoundment-Table S-1
• Storage impoundment/open tank -
Tablel-2
• Oil film on storage impoundment-
Table S-3
• Mechanically aerated impoundment -
Table S-4
• Diffused air system - Table S-5
• Land treatment - Table S-6
• Oil film surface on land treatment
unit-Table S-7
Closed landfill-Table S-8
Open landfill-TableS-9
Wastepile-Table S-10
Fixed roof tank-Table S-11
Floating roof tank - Table S-12
Variable vapor space tank -
Table S-13
Storage pile (particulates)-
Table3-l4
Exposed, flat, contaminated area
(particulates) - Table S-15
I
Select the source scenario for each modeling parameter (identified in Col. 1 of
worksheets) that best represents unit-specific conditions from available cases
(appropriate alternative case numbers are identified in Col. 3 of the worksheet
and case specifications are presented in Appendices C-Q):
• Disposal impoundment-Table C-1
• Storage impoundment/open tank-
TableD-1
• Oil film on storage impoundment -
Table E-1
• Mechanically aerated impoundment-
Table F-1
• Diffused air system - Table G-1
• Land treatment - Table H-1
• Oil film surface on land treatment
unit-Table I-2
Closed landfill-TableJ-1
Open landfill-Table K-1
Wastepile-Table L-1
Fixed roof tank-Table M-1
Floating roof tank-Table N-1
Variable vapor space tank -
Table 0-1
Storage pile (particulates) -
Table p-i
Exposed, flat, contamianted
area (particular
f^l I VW I I I I M I I VW W
es) Table Q-1
Compute parameter-specific scaling, factors by completing Cols. 4-11 (12 for
Raoult's Law surrogates) of the worksheet or Co|. 4 for particulate worksheets
based on modeling results presented in Appendices C-Q (computational
;ented with each v
instructions are presentee
worksheet):
• Disposal impoundment - Table C-2
• Storage impoundment/open tank-
TableT>-2
• Oil film on storage impoundment -
Table E-2 s
• Mechanically aerated impoundment -
Table F-2
• Diffused air system - Table G-2
• Land treatment - Table H-2
• Oil film surface on land treatment
unit -Table I-2
Closed landfill-Table J-2
Open landfill-Table K-2
Wastepile - Table L-2
Fixed roof tank -Table M-2
Floating roof tank - Table N-2
Variable vapor space tank -
Table O-2
Storage pile (particulates) -
Table"P-2
Exposed, flat, contaminated area
(particluates) Table Q-2
Complete unit-specific emission rate, which accounts for unit-
specific scaling factors (last line item on each worksheet based
on instructions presented with each worksheet).
Step 4-
Calculate Concentration Estimates
2-11
-------
interpolated to estimate a unit-specific emission rate. The process for calculating
emission rate estimates for application to a specific unit (i.e./unit-specific
application) is summarized in Figure 2-5.
Calculating emission rate estimates is accomplished by completing an Emission Rate
Estimation Worksheet, included in Appendix S. A separate worksheet is provided in
Appendix S for each unit category. Column 2 (unit-specific values for each modeling
parameter) of the worksheet should already have been completed during Step 1.
The alternative emission estimation approach presented in Figure 2-5 also involves
scaling the emission rate modeling results available in Appendices C through Q to
represent unit-specific conditions. This is accomplished by first computing
individual parameter-specific factors and then combining the results to calculate a
unit-specific emission rate for each surrogate of interest. Therefore, it is necessary
to select the appropriate source scenario that best represents unit-specific
conditions for each modeling parameter (identified in Column 1 of the worksheet).
Column 3 of the worksheet identifies the appropriate candidate scenario cases for
each parameter. The source scenario case specifications (i.e., values of the modeling
parameters for each case) are presented in Table C -1 (disposal impoundment), D-1
(storage impoundment), E-1 (oil film on storage impoundment), F-1 (mechanically
aerated impoundment), G-1 (diffused air system), H-1 (land treatment), 1-1 (oil film
surface on land treatment unit), J-1 (closed landfill), K-1 (open landfill), L-1
(wastepile), M-1 (fixed roof tank), N-1 (floating roof tank), O-1 (variable vapor space
tank), P-1 (storage piles), and Q-1 (exposed, flat, contaminated areas).
It is also recommended that a second scenario case be selected for each parameter
in order to bracket source conditions. The selection of a second scenario is
appropriate if unit-specific source conditions are different than those presented in
the source scenario case specifications (Appendices C-Q).
Parameter-specific scaling factors are computed by following instructions in each
worksheet and by completing Columns 4-11 (12). (Column 12 is needed for RaouIt's
Law surrogates.) Information needed to complete Columns 4-11 (12) is available in
Appendices C through Q. Information needed to complete worksheets for
particulate emissions are available in Appendices P and Q. Instructions for
2-12
-------
computing unit-specific emission rates based on applying scaling factors are
included in each worksheet.
The last set of three source scenario cases for unit-category modeling results
presented in Appendices C through Q represents the following:
• Reasonable best-case emission rate for unit category (for a typical source
surface area or tank size)
• Typical emission rate for unit category (for a typical source surface area
or tank size)
• Reasonable worst-case emission rate for unit category (for a typical
source surface area or tank size)
Frequently these cases can be used to rapidly estimate typical and extreme emission
rates. However, they should not be considered as absolute values. These scenarios
generally represent the range of source conditions identified in the Hazardous
Waste Treatment. Storage and Disposal Facilities (TSDF) Air Emission Models (U.S.
EPA, December 1987). But frequently this information was incomplete, and
subjective estimates were postulated instead. Therefore, the emission rates for
best, typical and worst case source scenarios should only be used as a preliminary
basis to compare and prioritize sources.
At times one of the source scenario cases presented in the Appendices may be
representative of the modeling parameters for the unit scenario being evaluated.
For these situations, it is not necessary to implement all of the intermediate
computational steps otherwise needed to complete the worksheet. Instead, the
modeling results presented in Appendices C through Q can be used to directly
represent unit-specific emission rates. However, it may be necessary to scale these
results to account for the unit-specific surface area and waste constituent
concentrations. (Scaling can be accomplished by the approach specified in each
worksheet).
2-13
-------
2.5 Step 4 - Concentration Estimates
Emission rate values from Step 3 are used as input to calculate concentration
estimates at receptor locations of interest. Dispersion conditions are accounted for
by use of available modeling results for typical annual meteorological conditions. A
summary of this process is included in Figure 2-6. Dispersion models can be applied
to directly estimate concentration. This primary approach is recommended for most
applications. The EPA-lndustrial Source Complex (ISC) model is generally
appropriate for a wide range of sources in flat or rolling terrain. Alternative models
are identified in the Guideline On Air Quality Models (Revised) (U.S. EPA, July 1988).
An alternative approach to obtain concentration estimates (for flat terrain sites)
involves the application of dispersion factors presented in Appendix R. A
Concentration Estimation Worksheet (Table R-1) is used as the basis for
concentration calculations. This approach is appropriate when a rapid preliminary
estimate is needed and modeling resources are not available. However, the primary
approach will provide more representative site-specific concentration estimates.
Specific instructions for implementing the alternative concentration estimation
approach are presented in Figure 2-7.
Concentrations should be estimated at locations corresponding to receptors of
concern (pursuant to RFI Guidance). Receptor information may also be available
from the RFA. Column 2 of the worksheet should be completed to define distances
to receptors as a function of direction.
Ambient concentrations are influenced by atmospheric dispersion conditions in
addition to emission rates. Atmospheric dispersion conditions for ground-level non-
buoyant releases (as is the case for surface impoundment, landfill, land treatment
unit, and wastepile applications) can be accounted for by the use of dispersion
factors. Appropriate dispersion factors based on Figure R-1 should be used to
complete Column 3 of the worksheet The dispersion factors presented in Figure R-1
include individual plots for a range of unit-surface-area sizes. Instruction regarding
the use of these plots to determine unit- and receptor-specific dispersion factors is
included with Figure R-1.
2-14
-------
FIGURE 2-6
STEP 4 - CALCULATE CONCENTRATION ESTIMATES
Emission Estimates
Primary approach -
calculate concentrations by
using dispersion models:
• ISC
• Other models identified
in Guideline on Air
Quality Models
(Revised)
Alternative approach -
calculate concentrations by
applying dispersion factors
(see Figure 2-7).
Step 5-
Compare Results to
Health-Based Criteria
2-15
-------
FIGURE 2-7
STEP 4 - CALCULATE CONCENTRATION ESTIMATES
(ALTERNATIVE APPROACH)
Emission Estimates
RFA Receptor
Information
Obtain Concentration
Estimation Worksheet
(Table R-1).
Define receptor locations of interest
(complete Col. 2 of worksheet to
define distances of receptors as a
function of direction).
Determine dispersion factor (Chi/Q)
values for appropriate source area
and receptor downwind distance
based on Figure R-1 (complete Col. 3
of worksheet).
Assume annual downwind frequency
of 100% for each receptor (complete
Col. 4 of worksheet).
Calculate long-term ambient
concentrations based on Equation 1
of worksheet (complete Cols. 5-13).
Step 5-
Compare Results to
Health-Based Criteria
2-16
-------
The dispersion factors presented in Figure R-1 are based on the assumption that
winds are flowing in one direction (i.e..toward the receptor of interest) 100 percent
of the time on an annual basis. This conservative assumption of a wind direction
frequency of 100% for each receptor of interest should be used if Figure R-1 is used
as the basis to estimate dispersion conditions for Column 4 of the worksheets.
The information entered into Column 3 and 4 of the worksheet, plus the emission
rate results calculated during Step 3, provides the required input to calculate
ambient concentrations. Specifically, Equation 1 presented in the worksheet should
be used to obtain ambient concentrations for each surrogate and receptor location.
Equation 1 of Table R-1 includes a safety factor of 10 which is applied to all
concentration estimates based on the scaling approach. This factor accounts for the
inherent uncertainty involved in the scaling approach. This safety factor is
applicable to all concentration estimates based on emission rates obtained via the
scaling approach. These results should be entered into Columns 5 through 13 of the
worksheet.
I
2.6 Step 5 • Health Criteria Comparisons
Concentration results from Step 4 can be compared to constituent-specific
health-based criteria provided in the RFI Guidance (see Figure 2-8). To facilitate this
comparison, it is recommended that the appropriate reference toxic and
carcinogenic criteria be entered in the space allocated in the Concentration
Estimation Worksheet.
Interpretation of the ambient concentration estimates should also account for the
uncertainties associated with the following components of the assessment:
• Inaccuracies in input source characterization data will directly affect
concentration results.
• Emission rate models have not been extensively verified. However,
OAQPS states, "In general, considering the uncertainty of field emission
measurements, agreement between measured and predicted emissions
generally agree within an order of magnitude." (U.S. EPA, April 1987).
These verifications have been for short-term emission conditions. Model
2-17
-------
FIGURE 2-8
STEP 5 - COMPARE RESULTS TO HEALTH-BASED CRITERIA
RFI Guidance
Health Criteria
Concentration Estimates
l
Compare annual concentrations
to health criteria:
• Toxic criteria
• Carcinogenic criteria
Consider modeling/screening
methodology uncertainties and
background concentrations.
Consider variations in emission
rates/concentrations for various
exposure periods
Input to Decision on Need for
RFI-Air Media
2-18
-------
performance is expected to be better for long-term emission rate
estimation (as used for this screening assessment).
• Inaccuracies associated with use of the alternative emission estimation
approach presented in Figure 2-5.
Source conditions for the unit of interest may not be the same as
those for the source scenarios presented in Appendices C-Q.
Therefore, scenarios should be selected to bracket the unit-specific
conditions in order to obtain a range of emission rate estimates.
The use of scaling factors for each source parameter may yield
somewhat different emission rate values compared to those based
on direct use of a model with unit-specific inputs. These differences
are attributed to the interrelationships of source parameters which
may not be linear. A comparison of direct modeling results versus
scaling estimates is presented in Exhibit 2-1.
• Atmospheric dispersion models for long-term applications (as used for
this screening assessment) typically are accurate within a factor of ± 2 to
3 for flat terrain (inaccuracy can be a factor of ± 10 in complex terrain.
Therefore, "safety factors" commensurate with these uncertainties should be
applied to concentration estimates for health criteria comparisons.
The calculations of emission rate and concentration estimates obtained have been
for a 1-year period. Some units, such as closed landfills, will have different average
emission rates for longer exposure periods for certain constituents. The air pathway
health-based criteria included in the RFI Guidance are based on a 70-year exposure
period. Appendices C through Q each contain a set of scenario cases for 1-, 5-, 10-,
and 70-year exposures for information purposes. However, only inactive units are
expected to have an average 70-year emission rate that is significantly different
from the 1-year rate. All of the emission results presented in Appendices C through
Q are assumed to be active with the exception of closed landfills (Appendix J). Air
concentrations for each one-year period within the reference 70 year exposure
period should be less than those associated with constituent-specific health criteria.
2-19
-------
< UJ
i"
2 in
Ul i2
CD
X,
o 2
P £
< —
ee S
scenarios
iable Best CaseA/Vorst Case Emission Rate 5
o
m
ac
VHVLB
HHHB
VHVHB
II"
II"
HHMB
MVMB
QQ flfl
II
II"
II1
Henry's Law Surrogates: MHLB
Raoult's Law Surrogates: HVHB
Particle Case: Particle
Unit Type
•'
00
o
s
o
00
vO
0
—
o
s
o
s
o
CO
o
CO
o
o
o
o
o
"
prt
O
~
0
o
*~
3
"~
s
8
8
[Disposal Impoundment
i
S
s
£
0
o
*~
o
o
^
o
m
in
^
cn
O
m
^
CM
in
*~
O
in
o
o
in
(Storage Impoundment
o
o
IN
^
2
00
o
"
o
^~
o
o
o
5
o
o
in
o
o
"
en
0
tn
o
5
nSlo
3
8
m
01 —
^11 ^^
-lo
ILand Treatment (after
tilling)
(N
d
(N
cn
O
00
o
~
0
"
o
o
p.
o
00 1 0^
C^I^D
Sj^p
1^5
sqs
mi-
en «N
INJOO
cn UN
.,-
[Oil Film Surface on Land
(Treatment Unit
IN
"
00
00
^""
2
^™
—
^
9
•*"
iS
00
~
o
o
o
"
en
en
o
fN
0
(N
O
"""
O
"™
o
*~
8
o
*"
o
[Closed Landfill
00
cn
O
cn
o
0019
en cn
did
o
o
en
o
m
-------
3.0 EXAMPLE APPLICATIONS
Two case studies have been selected to demonstrate the application of the
alternative (scaling) air assessment screening methodology based on the technical
aids presented in Appendices B through S. The first example involves a storage
impoundment and the second a closed landfill.
3.1 Case Study A
Case Study A involves a storage impoundment located close to a small community.
The closest resident lives 0.2 mile south of the unit. The impoundment has a surface
area of 1 acre, a depth of 0.9 meter, and a typical storage time cycle of 1.2 days.
Wind data from the nearest National Weather Service station indicate that
northerly winds occur 10 percent of the time annually. Waste records for the un^t
indicate the frequent appearance of carbon tetrachloride. Limited waste analyses
indicate that a 1,000-ppm concentration of this constituent in the impoundment is a
reasonable assumption. The object of this example screening assessment is to
estimate the ambient concentrations at the nearest residence. Following is a
summary of this example application.
Step 1 -Obtain Source Characterization Information
The appropriate Emission Rate Estimation Worksheet for this case study is Table S-2
for storage impoundment units. The unit information provided above is sufficient
to complete Column 2 for Lines 1-4 of the worksheet (see Exhibit 3-1) pursuant to
Instruction A of the Worksheet (Table S-2).
i
Step 2 - Select Release Constituent Surrogates
Based on Figure 2-3, it is apparent that the Henry's Law Constant surrogate subset
(Table B-1) is appropriate for a storage impoundment unit. Evaluation of Table B-3
indicates that the following surrogate is applicable to Case Study A:
3-1
-------
LU
i
X
UJ
a
1
a
I
O
1=
>
LU
S
o
LU
_
•-
"o
u
o
•—
"o
u
«••
Q
U
00
3
^
3
V0
"5
u
LA
3
^
T5
u
3
IN
S
3
V
=
GO
5
X
CO
• 5
«« >
^ ^".
O
t? «
^^ ^j
Si
._ ^
10
u
.. ** eo
LJ W 5
*.• Me
c :t >
** Q.
3 ^
^ 91 *
> •* X
£| 1
3
Vrt a)
C >
v «°
O S
I
00
X
^^
X
lo
clil
O w ttf 4^
•2 £ Q-"o -.
s 8- IN
m IN
0
^f
m
i
S.
Af ^3
oe oT t
^rf afb
1 "ul
Typical Surrogate-Specific Em
(Case 18). 106g/yr
Calculate Unit-Specific Emissii
(multiply lines # 2 x #3 x #5 x
rx oo
Ol
oe
o
'%A
«/l
1
•o
41
(N
a
O
V
m
E
o
O
x
^
Q.
CL
O
•o
(O
(N
1/1
C
J
,0
01
V «-
5*2 s
-• o "
|s =
.£ u. —
•
«
« «
3-2
-------
Constituent
• Carbon tetrachloride
Surrogate No.
Surrogate Code
HHLB
Step 3 - Calculate Emission Estimates
This step involves implementing Instructions B-D of the Worksheet (Table S-2).
Instruction B involves selection of representative cases from Table D-1 which best
match actual unit values in Column 2. A review of Table D-1 indicates that Case 1
(based on a depth of 0.9 meter) best estimates the depth of the example case (also a
depth of 0.9 meters has been specified for Case Study A). Table D-1 also indicates
that Case 5 (based on a retention cycle of 1 day) best represents the example case (a
retention cycle of 1.2 days has been specified for Case Study A).
Implementation of Instruction C involves determination of surrogate-specific
i
scaling factors. For this example this involved completion of Column 5 for lines 2
and 3 of the Worksheet (Table S-2). Emission rates for Cases 1 and 5, and a typical
emission rate (Case 18) were obtained from Table D-2 as follows:
Case
Casel
CaseS
Case 18
Emission Rate (1Q6g/yr)
Carbon Tetrachloride
22.5
161.5
39.2
Column 5 of the worksheet (for carbon tetrachloride) was completed via the
following computations (Case 18 represents a typical emission rate for the source
category of storage impoundment):
*Line2:
Case 1 Emission Rate (from Table D-2)
Case 18 Emission Rate (from Line 7 of the Worksheet)
0.57
3-3
-------
*Line3:
Case 5 Emission Rate (from Table D-2) 161.5
Case 18 Emission Rate (from Line 7 of the Worksheet) 39.2
= 4.1
Implementation of Instruction D of the Worksheet (Table S-2) involves completion
of Lines 5-6 and 8 as follows:
*Line5:
Unit-Specific Area (from Column 2 of the Worksheet) 1.0
Case 18 Area (this value is identified in the Worksheet 0.4
instructions for Line 5)
*Line6:
Unit-Specific Concentration 1,000
= 2.5
= 1.0
Case 18 Concentration 1fOOO
*Line8:
Emission Rate = Line 2 x Line 3 x Line 5 x Line6x Line?
= 0.57x4.1x2.5x1.0x39.2
= 229.0 x106g/yr
- 229.0 Mg/y
Step 4 - Calculate Concentration Estimates
This step involves use of the Concentration Estimation Worksheet (Table R-1).
Application of the Worksheet involves implementation of Instructions A-D included
in Table R-1. The example Concentration Estimation Worksheet for Case Study A is
presented in Exhibit 3-2. Implementation of Instruction A involves input of the
distance of the receptor from the downwind unit boundary for sectors of interest.
Notice that the receptor distance of 0.2 mile (Column 2) corresponds with the south
(downwind) sector. This is because the frequency of northerly winds obtained from
the National Weather Service (as stated at the beginning of 3.1) represents the
3-4
-------
a.
2
X
UJ
d
uT
O
z
OSEDLA
cJ
ec
O
V9
UJ
^f
Z
UJ
5
ee
O
ATIONW
1
UJ
O
ICENTRATI
O
U
-
5
0
u
i
o
•3
u
91
3
00
3
0
u
3
in
3
i
en
3
IN
0
u
0
t.
I
«J
1
w
.£
C
5
5
(D
§•£
Ul
|
1
O
m
J5T
!
A V
istruction 1
on Estimal
3
I
0
e
3
1
G
1
1
W
5
u
c
w
S1
^
«
a
E
o
w
1 Instruction
•
S,
6
^
«
g
i
n
•w
'G
10
W
J
w
1
!•.
3 .
e *
1 £
1 ^
Jl
.- £
c o
X K
II II
2
: o>
ll
" i
ioi
s 1
555
* 5
||
i S
ioi
i s
5 OX
sj
x5x
•
2*1.
I ? C S
1 5i52.§
fslil!
a i 1 ?Z 1
< < Q i o S
ml
tructionA:
s.
||5 5
=155 i*£
.S li
M 2 1 1
ii
z
Ul
Z
u
u
u
Ul
Ul
Wl
Ul
u
u
n
1
1
in
^
X
IN
O
„,
3
5
I
5
Z
I
1
z
2 3
0
Criteria
'£'=
£ v
u 2
•Sa
£3
m* (I
I =
^«
.2!
u g
!l
ISI
I
•o
S!
I
o
u
I
•= fe
§ I
^5*' •
"
3-5
-------
direction "from which the wind is flowing." This is standard meteorological
terminology. Therefore, northerly winds affect receptors south of the unit.
Implementation of Instruction B involves determination of the appropriate
dispersion factor for the downwind distance selected. The dispersion factor
obtained from Figure R-1 for this example is 6.4 x 10-5 sec/m3 (entered in Column 3
of the Concentration Estimation Worksheet). This value is applicable to a receptor
0.2 mile downwind from a 1-acre area source.
Implementation of Instruction C involves entering the downwind frequency for the
sector of interest in Column 4 of the Worksheet. The downwind frequency
(conservatively assumed to be 100 percent if Table R-1 dispersion factors are used)
fora receptor located south of the unit is entered in Column 4 of the Worksheet.
Implementation of Instruction D involves computation of air concentrations based
on Equation 1 of the Worksheet (Table R-1). The concentration estimate for carbon
tetrachloride was calculated using Equation 1 of the Worksheet as follows:
• Worksheet estimate:
Concentration (pg/m3) = Col. 3 x Col. 4 x Emission Rate x (unit conversion =
3.17 x 102) x (Safety factor « 10)
= (6.4 x 10 -5) x (100) x (229.0) x (3.17 x 102) x (10)
= 4600 vig/m 3
Step 5 - Compare Results to Health Criteria
Available health-based criteria from the RFI Guidance were entered into the
Concentration Estimation Worksheet (see Exhibit 3-2). These results indicate that
carbon tetrachloride concentrations at the nearest receptor significantly exceed the
carcinogenic health-based criteria. Based on the expected carbon tetrachloride
concentrations, this unit is a prime candidate for unit-specific emission rate and
dispersion modeling to confirm the need for an RFI for the air media.
3-6
-------
3.2 Case Study B
Case Study B involves a closed landfill of 7 acres with a waste-bed thickness of 25
feet and a cap thickness of 6 feet. Benzene is believed to be a primary constituent
of the waste (approximately 10 percent). The closest resident lives 1 mile east of the
unit. The prevailing winds (which occur 20 percent of the time annually, based on
available facility data) are from the west (i.e., these winds will affect the downwind
sector east of the unit). Following is a summary of the screening assessment for
Case Study B.
Step 1 -Obtain Source Characterization Information
The appropriate Emission Rate Estimation Worksheet for Case Study B is Table S-8
for closed landfill units: The unit information provided is sufficient to complete
Column 2 of the worksheet, with one exception (see Exhibit 3-3): the air porosity of
the fixed waste is not known. Therefore, typical conditions [i.e., 25 percent a^
represented by Cases 14 and 22 (see Table J-1) will be assumed for this assessment].
Step 2 - Select Release Constituent Surrogates
•
Based on Figure 2-3, it is apparent that the Raoult's Law surrogate subset (Table B-2)
is appropriate for a closed landfill unit. Evaluation of Table B-3 indicates that the
following surrogate is applied to Case Study B:
Constituent Surrogate No. Surrogate Code
Benzene 1 HVHB
Step 3 - Calculate Emission Estimates
The calculational inputs for the Emission Rate Estimations Worksheets for Case
Study B are presented in Exhibit 3-3 and 3-4. Scenario Case 1 (Exhibit 3-3) and
Scenario Case 2 (Exhibit 3-4) were selected to bracket the actual waste-bed thickness
for the example unit. Scenario Case 1 is associated with a waste-bed thickness of 15-
feet and Case 2 with a 30-foot bed thickness. The actual waste-bed thickness is 25
feet. The resulting benzene emission rate estimates range from 46.4 x 106g/yr to
83.4x106g/yr.
3-7
-------
a.
UJ
iZ
O
!
§
o
m
O
i
1
rM
^
^5
u
^
o
u
o
«•
•5
u
o\
0
CO
•5
u
£
u
tfi
"5
u
m
"3
u
3
"o
u
rsi
0
U
0
u
4>
c
00
>
I
>
CO
I
I
>
J
i/i
o 3?
*3 ^>
U. — I
Oi
^ OQ
3 >
u v 2
c E
§* I
* i 2
vi *i
c ^
""* O^ JHB
§ I
*c
c
1 2
OJ ^
+•* T"
? Ol
3 ^L ^ ^ ^*
c * I T -o
^~ ^ — ^ 4^
« "oJ J! w
^~ 2 *i •"•"
t/{ (3 bZ S
. ,
^J ^f *^» Al
* S .« E *
Ot •* %£ LU w
fc-8. Z 'i ^^
5 | 5 a 5"k
C ^^ Lrt W W
o « ' 5
\* *w ^ ^^ 9% *f ^^
S Is " ^i i
li h u
c | 2g S-a
2 j) S a> s5
li M 31
vc r» co
Oi
j£
U
c
•o
c
; —
01
•D
01
W
ec
c
o
£
E
LU
15
'5.
i~
^
•o
o>
T3
5
A
i
in
,E
'o
h.
*
T]
E
LU
*^
ac
c
o
£
£
LU
^ ^
M
0>
a
a
<(
E
o
w
in
PSI
i
'_j
5
1
"§
VI ^
V *
— 4>
(0 T3
ilf
III
'C S rsi
U i/i rsi
«
3-8
-------
a.
UJ
Q
UJ
(/i
UJ
o
5?
(N
"5
u
"
o
u
0
0
u
a\
"5
u
00
"5
u
r.
0
u
10
"o
u
m
"5
u
<•
"5
u
/Mb
n
"o
u
(N
3
0
u
0>
'^
03
>
I
>
CO
I
>
I
•
I*
2 1
vJ ^!
(0 >
LL. -1
91
.E ffi
fB -^
3 >
W u
§5
ai m
•5 8. 5
3«<» >
is * 3
«• 5
5 &
2 T
| >
«^ 5
or
c
E 3
& >
« T
o> -1-
o
m
S
>
fti
Uit
>
X
OI '
> UL
-1|2
5SI1
~ S 9- -o -.
w k- &. c 01
5 g- ™* ^
Sc
0)
S g i
g * „ a a
18 ml •*• oi mi
M «Nl U9I H (Ml
«
01
a 1 -
aJ *- 5 >, o»
"S *U J « "o 'is S
. ti | i{ I|
s 13 J || ?1
< $5 5 as <£
^- fN no ^ i*n
s
(N
ON
^
^
UJ
in in
m o
"
m
UJ
=)
^•J
<
y
00
Q
00
b
u.
O
UJ
2?
UJ
00
o»
m
0
I-
s
0
oc
ce
m
^
^
d
•^
^
^ ^
i « S x E *
£ S « S UJ X
8? 1 * P
u sl S| «^^
•5 (0 S, »» ,1 „,
1= I* 1.1
«i ss st
§ 2 S * =5
II M 31
vo r- oo
O>
*/1
(0
u
c
0)
•o
0>
c
o
(O
w
'o.
(N
I
ITJ
E
Si
13
tr
O
o»
Q.
0.
<
O
in
_c
'Jj
5
I
E
- 01
(0 "O
* w.
•C
U (X
3.Q
-------
Step 4 - Calculate Concentration Estimates
The example Concentration Estimation Worksheets for Case Study B are presented
in Exhibits 3-5 (Scenario Case 1) and 3-6 (Scenario Case 2). The resulting benzene
concentration at the nearest receptor is estimated to range from 69 ug/m3 to 124
ug/m3.
Step 5 - Compare Results to Health Criteria
A review of results presented in Exhibits 3-5 and 3-6 indicates that the estimated
benzene concentrations of 69 wg/m3 to 124 ug/m3 are approximately 1000 times the
carcinogenic criterion of 0.1 ug/m3. A toxic criterion is not available for benzene.
Based on the results presented in Exhibits 3-5 and 3-6, this unit is a prime candidate
for an air release RFI.
3-10
-------
S
|0
U
•§
IV
§
t/>
s
2
^
Ul
o
J
o
Ul
• A
EXHIBIT 3-5
TABLE R-1
HT CATEGORY: CLO!
^
i
Ul
Ul
4/t
^
i
3
§
5
<
p
i/>
Ul
$
CONCENT
m
3
IN
•3
Wl
*•
•3
U
e
0*
3
to
3
fNt
3
1
c 3
^ 2
II H
5
iol
. s
i.x
x 5>
I i
§52
IM
1 >
x. xs
MHMB HHMB 1
or or <
MVHB MVMB 1
1,1
_ 00
x3x
•
3 I S.
is^sli
^ ? -
ililii
I1! *
lib'
la!*!
<
.§
1
i
•
• *s
J (i
h'sil
?
Z
Ul
Z
Ul
£
Ul
s
s
s
M
r>
i
s
I
$
i
,
2
•
•
12
i
5
* w
•?. c
Toxic CriU
Carcinogc
Sf •
ii
H
«!Z
11
a
X
r»
m
II
3
S
1^
S
•g
1
2
3
H
3-11
-------
0)
M
U
0
'w
m
c
41
^
£.
UJ
—I
X
Ul
_j
tZ
O
!
a
Ul
o
_i
U
^
oc
.
1
|
3
m
w
w
3
*
re
e appropri
w
w
•o
C
re
i
Ol
o
3
C ">
I I
;• >
J 5
-. .*
11
II II
00
_J
i o>
09
IJ
X O>
£ QD
isl
oa «
!5!
00 J*
2 5
T «5
x 02
11
555
09 a
_ 09
!J
X OX
•1
00 03 V
is! sl*
1 Assume
Annual
tructionB:
**
c
I
3
5
-
a
lg s
Q i'oS
*§ i
§ | s «
S'5u?£
wt
* 5
s II
5211
ll
Z
UJ
Z
UJ
UJ
s*
fll ^
« U.
5 0
€^
U
I
a
31
o
5
m
c
o
U
o
U
« re
§ |
2 |
1 1
o "o
E ?
-?|.
ri!Li
tl
c
«
3 i 2 5 re U
§• re £ •= „
uj £. Q u O w
• = If
« • < Z 2
3-12
-------
4.0 REFERENCES
U.S. EPA, September 1985 (and subsequent supplements). Compilation of Air
Pollutant Emission Factors, Vol. I. Washington, DC 20460.
U.S. EPA, June 1974. Development of Emission Factors for Fugitive Dust Sources,
Research Triangle Park, NC, 27711.
U.S. EPA, March 1978. Fugitive Emissions from Integrated Iron and Steel Plants. EPA
600/2-78-050, Washington, D.C.
U.S. EPA, July 1988. Guidelines on Air Quality Models (Revised). EPA-450/2-78-027R,
Office of Air Quality Planning and Standards, Research Triangle Park, NC 27711.
U.S. EPA. December 1987. Hazardous Waste Treatment Storage and Disposal
Facilities (TSDF) Air EmissiorTModels. Office of Air Quality Planning and Standards.
Research Triangle Park, NC 27711 (CHEMDAT6).
U.S. EPA, 1989. RCRA Facility Investigation (RFI) Guidance. Office of Solid Waste,
Washington, D.C. 20460.
Turner. D.B. 1969. Workbook of Atmospheric Dispersion Estimates. Public Health
Service, Cincinnati, OH. ,
I
4-1
-------
Appendix A
Background Information
-------
A.O BACKGROUND INFORMATION
The air release screening assessment methodology has been developed based on
use of available air emissions models applicable to facilities for treatment, storage,
and disposal of hazardous waste, and on results of atmospheric dispersion
modeling. The emission models were used to calculate emission rates for a wide
range of source scenarios. (An emission rate is defined as the source release rate for
the air pathway in terms of mass per unit of time.) These modeling results have
been summarized in this document so that they can be easily used by Environmental
Protection Agency (EPA) Regional and State Agency staff to estimate emission rates
for facility-specific and unit-specific applications. These source-specific emission
rates can be used in conjunction with dispersion modeling results, representative of
typical annual conditions, to estimate long-term ambient concentrations at
locations of interest. (Ambient concentrations are defined as the concentrations of
the released constituent downwind from the source.) The emission rate and
atmospheric dispersion modeling approaches used to develop the screening
methodology are discussed in the subsections that follow.
A.1 Emission Rate Models
The air release screening assessment methodology has been based primarily on
application of air emission models (available on a diskette for use on a
microcomputer) developed by EPA's Office of Air Quality Planning and Standards
(OAQPS) to estimate organic releases for hazardous waste treatment, storage, and
disposal facilities (TSDFs) (U.S. EPA, December 1987). Computer-compatible air
emission models (referred to as CHEMDAT6 models) are available for the following
sources:
• Surface impoundments, which for modeling purposes include quiescent
impoundments, aerated impoundments, and open-top tanks
Disposal impoundments
Storage impoundments
Oil films on storage impoundments
Aerated impoundments
A-1
-------
• Land treatment
Soil emissions subsequent to waste tilling
Oil film surfaces
• Closed landfills
• Open landfills
• Wastepiles
Since the results presented in this document are based on the December 1987
version of CHEMDAT6, subsequent modifications to any of these models may
require revisions to this screening methodology
The available models for CKEMDAT6 provide a basis to estimate emissions for
numerous unit categories (e.g., surface impoundments, landfills) as previously
listed. Therefore, the CHEMDAT6 models will be applicable to a wide range of air
release screening assessments. CHEMDAT6 (December 1987 versions) does not,
however, include models for the following sources: '
I
• Land treatment-waste application
• Fixation pits
• Containerloading
• Container storage
• Containercleaning
• Stationary tank loading
• Stationary tank storage
• Fugitive emissions
• Vacuum truck loading
However, guidance for estimating organic emissions from these sources is available
from OAQPS (U.S. EPA, December 1987).
In addition to the CHEMDAT6 model, emission equations from EPA's AP-42,
"Compilation of Air Pollutant Emission Factors" and "Fugitive Emissions from
Integrated Iron and Steel Plants" have been used for estimating organic emissions
from storage tanks and paniculate matter emissions that are less than 10 microns in
diameter from storage piles and exposed areas which result from wind erosion and
activities on storage piles.
A-2
-------
A.2 Source Scenarios
A wide range of source scenarios were evaluated as a basis for developing the air
release assessment methodology. This involved identification of a limited set of
surrogates to represent the numerous individual potential air release constituents
of concern. This also involved evaluating of the sensitivity of the input parameters
used by the CHEMDAT6 air emission models and the AP-42 emission equation input
parameters.
A.2.1 Release Constituent Surrogates
A limited set of surrogates was required to simplify the air release assessment
methodology since the list of potential air release constituents included in the RFI
Guidance (U.S. EPA, 1988) is extensive. The set of surrogates selected for this
application was the same list developed by OAQPS for assessment of organic
emissions from TSDFs (see Appendix B).
Two subsets of suhogates are presented in Appendix B. The first subset is
applicable to air emission modeling applications based on the use of the Henry's
Law Constant (Table B-1) and the second subset is based on use of Raoult's Law
(Table B-2). Raoult's Law accurately predicts the behavior of most concentrated
mixtures of water and organic solvents (i.e., solutions over 10 percent solute).
According to Raoult's Law, the rate of volatilization of each chemical in a mixture is
proportional to the product of its concentration in the mixture and its vapor
pressure. Therefore, Raoult's Law can be used to characterize potential for
volatilization. This is especially useful when the unit of concern entails container
storage, tank storage, or treatment of concentrated waste streams.
The Henry's Law Constant is the ratio of the vapor pressure of a constituent to its
aqueous solubility (at equilibrium). This constant can be used to assess the relative
ease with which the compound may vaporize from the aqueous solution and will be
most useful when the unit being assessed is a surface impoundment or tank
containing dilute wastewaters. The potential for significant vaporization increases
as the value for the Henry's Law Constant increases; when it is greater than 10E-3,
rapid volatilization will generally occur.
A-3
-------
The surrogates presented in Appendix B span the range from very high volatility to
low volatility (frequently classified as semi-volatiles). Biodegradation potential has
also been accounted for in the surrogate specifications. Therefore, a cross-
reference of constituents has also been provided in Appendix B (Table B-3). This
listing provides the basis for the identification of the appropriate surrogate for
individual air release constituents of interest. Instructions for use of Appendix B
data are provided in Section 2.
A.2.2 Sensitivity Analyses
Sensitivity analyses of the input parameters used by the CHEMDAT6 air emission
models emission rate relative to output were evaluated to determine the feasibility
of developing a source characterization index. The object of the source
characterization index was to define a simple relationship between the primary
source description parameters and the emission rate of the release. This evaluation
was accomplished by modeling a series of source scenario cases for each unit
category (i.e., categories such as surface impoundments and landfills). Each of these
source scenario cases represents long-term (i.e., annual) emission conditions. A base
case representative of typical source conditions was defined for each unit category.
These typical conditions were specified based on TSDF survey results and on
guidance presented in the OAQPS air emissions modeling report (U.S. EPA,
December 1987). This base case provided a standard for comparison to results of
parametric analyses. The parametric analyses consisted of varying (one at a time)
the input values for the most sensitive modeling parameters. These input
parameter values were varied over a range of expected source conditions. In
addition to the parametric analyses and the typical (base-case) scenario, a
reasonable best-case (minimum emission rate) and a reasonable worst-case
(maximum emission rate) source scenario were also modeled. The most sensitive
modeling parameters and their associated range of values were determined by
considering model sensitivity results and TSDF source survey information presented
in the OAQPS air emission modeling report (U.S.EPA, December 1987), as well as
other judgmental factors. A similar sensitivity analysis was performed for the three
tank types.
A-4
-------
A summary of the air emissions modeling parameters, input values, and modeling
results (emission rates) is presented in Appendices C through Q. Evaluation of these
results indicates that emission rates are highly dependent on numerous sensitive
source parameters. Therefore, these complex relationships are not conducive to
development of a source characterization index (i.e., defining a simple relationship
between the primary source description parameters and the emission rate of the
release). However, the modeling results presented in Appendices C through Q
provide data which can be interpolated to estimate unit-specific emission rates with
minimal guidance. The methodology for application of these data is discussed in
Section 2.
A.3 Atmospheric Dispersion Conditions
Atmospheric dispersion conditions affect the downwind dilution of emissions from
a source. Available EPA dispersion models can be used to account for site specific
meteorological and source conditions. For this screening assessment, modeling
results are presented which represent typical dispersion conditions (neutral stability
and 10-mph winds) in the United States.
Dispersion modeling results to be used for the screening assessment (assuming flat
terrain) are presented in Appendix R (Figure R-1) and are applicable to ground-level
sources with non-buoyant releases (this assumption is valid for surface
impoundments, land treatment units, landfills, waste piles, tanks, and exposed
areas). These results are presented in terms of dispersion factors. Dispersion factors
can be considered as the ratio of the ambient concentration to the source emission
rate. Therefore, dispersion factors facilitate the calculation of ambient
concentrations if emission rate estimates are available.
The dispersion factors presented in Figure R-1 were developed from similar
dispersion graphs presented in a standard technical reference (Turner, 1969). These
dispersion factors are applicable to long-term (e.g., annual) conditions. It has been
assumed that dispersion factors (and, thus also ambient concentrations) decrease as
a function of downwind distance but are uniform in the crosswind direction within
a 22.5 degree sector (22.5 degree sectors correspond with major compass directions
such as N, NNW, NW, etc.). The dispersion factors presented in Figure R-1 also
account for the initial plume size, which corresponds to the surface area of the
A-5
-------
source (Turner, 1969). Results presented in Figure R-1 are expected to be similar to
results from the EPA-approved Industrial Source Complex dispersion model.
A-6
-------
Appendix B
Release Constituent
Surrogate Data ,
I
-------
00
D
O
u
1
*-y*
eo>
o:
Q.
O
OC
D
* 0
500
TJ j-«g
IS) ^^
*>» ~
f fy
c ts
o> c
x o
u
Characteristics
6
z
•o
0
un
o
1
LU
rsi
(N
r\i
C
0
(TJ
.aw, low biodegra
^^
v»
E
X
E
"TJ
,
u.
C
o*
X
.c
en
IE
m
09
X
X
f^
o
LU
00
LO
^
C
o
(TJ
•o
medium biodegra
TJ
— J
>
C
0)
X
>
_0
o»
00*
03
X
_J
LO
0
LU
00
o
'if
c
o
nj
-o
«j
0>
.aw, medium biod
«^
i/i
^
c
o>
X
E
_3
^Q
s
h_
C
X
5
Q
^^
03
X
X
_J
in
o
LU
O
00
\a
c
o
"i
O)
Oi
T3
o
en
IE
.
E
Oi
X
E
_3
^n
Oi
E
^
O3
X
X
nn
O
1
LU
00
m
in
c
O
^^
, high biodegradal
(TJ
\
*>,
C
X
JZ
en
!E
r-
03
X
X
X
O
E
EP
£T
i O LJJ
\J
c ./»
>.
^
c
oi
O» O>
a*
B-1
-------
01
«/)
CO
00 CC
111 '
UJ
o.
O
t£.
O.
Ul
O
cc
cc
^
3
S ^— ,
H Vj
i" °
_.
o*-*
Q.
(T3
Characteristics
6
z
a;
•o
0
u
-
vO
o
, high biodegradation
>.
=
s
^=
fn
"6
j=
01
IE
CM
CO
X
VO
in
(N
, low biodegradation
>«
^=
(Q
"o
_£
o>
IE
m
CO
X
^^
o
CN
C
o
ility, high biodegradat
*•*
"o
^
1
^j
0)
E
CO
X
i
(N
q
rvi
C
o
'™
flj
•o
jm
ility, medium biodegrc
•+•
"5
^
E
3
^^
0)
E
in
| MVMB
^—
CT*
rsi
C
0
ility, low biodegradati
*-
"o
^
E
TD
a;
E
vo
CO
o
o
o
d
c
o
medium biodegradati
•
^
*V
"o
^
§
^
00*
CO
o
o*^
CO
«"
c
o
tility, high biodegrade
a
o
f~
at
IE
^^
k.
>
^
o*
| VHVHB
o
no
O
fN
O
tility, low biodegradal
TJ
O
j£
IE
^
w
>
rs
| VHVLB
« -
•2E o 5>
§.2 >^
-
B-2
-------
TABLE B-3
LISTING OF CONSTITUENT-SPECIFIC SURROGATES
Constituent
Acrylamide
Acrylonitrile
Aldicarb
Aldrin
Aniline
Arsenic
Benz(a)anthracene
Benzene
Benzo(a)pyrene
Beryllium
Bis(2-chloroethyl)ether
Bromodichloromethane
Cadmium
Carbon tetrachloride
Chlordane
1-Chloro-2,3-
epoxypropane
(Epichlorohydrin)
Chloroform
Chromium (hexavalent)
DOT
Dibenz(a.h) anthracene
1,2-D«bromo-3-
Chloropropane (OBCP)
1,2-Dibromoethane
1,2-Oichloroethane
1,1-Dichloroethylene
Dichloromethane
(Methyl ene chloride)
CAS
No.
79-06-1
107-13-1
116-06-3
309-00-2
62-53-3
7440-38-2
56-55-3
71-43-2
50-32-8
7440-41-7
1.11-44-4
75-27-4
7440-43-9
56-23-5
57-74-9
106-89-8
67-66-3
7440-47-3
50-29-3
53-70-3
96-12-8
106-93-4
107-06-2
75-35-4
75-09-2
Henry's Law
Constant
Surrogate Code
7
4
8
3
8
0
9
1
9
0
5
3
0
3
6
6
3
0
3
9
6
3
3
3
I
Raoult'sLaw
Surrogate Code
4
1
9
7
5
0
7
1
8
0
5
7
0
3
7
3
3
0
7
7
6
3
3
3
1
B-3
-------
TABLE B-3
LISTING OF CONSTITUENT-SPECIFIC SURROGATES (Continued)
Constituent
2,4-Dichlorophenol
2,4-Dinitrophenol
2,4-Dinitrotoluene
1,4-Dioxane
1 ,2-Diphenylhydrazine
Endosulfan
Ehtylene oxide
Heptachlor
Hexachlorobenzene
Hexachlorobutadiene
Hexachloroethane
Hydrazine
Isobutyl alcohol
Lindane (gamma-
Hexachlorocyclohexane)
3-Methyl-cholanthrene
4,4-Methylene-bis-{2-
chloroaniline)
Methyl parathion
Nickel
Nickel (refmtry dust)
Nickel subsuifidt
2-Nitropropan*
N-Nitroso-N-methyl urea
N-Nitroso-pyrrolidine
Pentachlorobenzene
Pentachlorophenol
CAS
No.
1 20-83-2
51-28-5
121-14-2
123-91-1
122-66-7
115-29-7
75-21-8
76-44-8
118-74-1
87-68-3
67-72-1
302-01-2
78-83-1
58-89-9
56-49-5
101-14-4
298-00-0
1440-02-0
7440-02-0
12035-72-2
79-46-9
684-93-5
930-55-2
608-93-5
87-86-5
Henry's Law
Constant
Surrogate Code
8
9
9
6
9
9
4
3
6
3
9
q
7
9
6
3
6
0
0
0
6
5
2
3
9
Raoult's Law
Surrogate Code
5
3
6
3
7
7
10
7
7
6
6
3
4
7
7
6
6
0
0
0
3
9
2
6
7
B-4
-------
TABLE B-3
LISTING OF CONSTITUENT-SPECIFIC SURROGATES (Continued)
Constituent
Perchloroethylene
(Tetrachloroethylene)
Styrene
1,2,4,5-
Tetrachlorobenzene
1 , 1 ,2,2-Tetrachloroethane
2,3,4,6-Tetrachlorophenol
Tetraethyl lead
Thiourea
Toxaphene
1 , 1 ,2-Trichloroethane
Trichloroethylene
2,4,5-Trichlorophenol
2,4,6-Trichlorophenol
CAS
No.
127-18-4
100-42-5
95-94-3
79-34-5
58-90-2
78-00-2
62-56-6
8001-35-2
79-00-5
79-01-6
95-95-4
88-06-2
Henry's Law
Constant
Surrogate Code
3
3
3
6
9
3
6
3
6
3
6
6
Raoult's Law
Surrogate Code
3
6
6
6
6
6
3
6
3
3
6
6
B-5
-------
Appendix C
Emission Rate Estimates
Disposal Impoundments
(Quiescent Surfaces)
-------
Q
Z
O
1/1
o
o
S
u.
u
UJ
Z
ui
vj
O
o
3
UJ
a
o
3
i/»
2
CASE NUMBERS
«
m
4
«
«
«
00
»••
«
«
P-.
S0
in
^
• g o»
121
C 3 v*
20$
S •" •£
5 a £
S w£
i £ ***
w tu P "?
.
•O
£
i_
2
I
£
c
c
•s o
Hj c
~ *
S
o
I I
0>
i
3
a
"Z
3
^£ =
o
C-1
-------
0
o
(V
2
*
O
w>
*
UI
UI
t/l
I/I
00
u
sr-
S
<9
VJ
s
8
10
u
S
»•
in
S
*
u
S
u
S
3
Henry's Law Constant Surrogate
00
«
2
IN
m
2
z
S
00
o
en
2
R
ea
z
in
^
S
Z
'
00
00
0
9>
2
a
CO
i
CO
00
0
91
2
a
OB
Z
o
*
OB
Z
00
e
2
3
1 MHHB
00
o
*
2
m
i
z
5 —
£*
«• ^
3 ji
> S
> 8
— u
1
Jt
«
0
S
10
w
Henry's Law Constant Surrogate
IN
m
S
IN
m
e
3
i
m
(0
2
IN
e
z
!
S
2
e
'
S
2
IN
in
e
i
IN
m
\o
2
IN
m
e
z
*
00
*
e
i
s
VO
IN
2
IN
m
e
| MHHB
IN
m
2
IN
m
e
z
2
_
2
10
u
Jt
,3
Henry's Law Constant Surrogate
1
IN
m
e
S
z
S
IN
m
e
OB
z
te
Z
e
'
S
m
IN
m
e
| MHMB
co
00
m
IN
e
CD
»
e
oa
Z
IN
m
e
| MHHB
oo
S
IN
O
00
z
C-2
-------
Appendix 0
Emission Rate Estimates
Storage Impoundments/Open Tanks
(Quiescent Surfaces)
-------
Q
Z
O
a.
5
ae
O
O
g
eo;:
IN
nable Best
S
5
01
at
f
«
urce area)
2
<0
W
a
pi
assumin
§
u
al Emission
w
1
•-
;
<
ource area)
w«
"5
.summg typi<
n
4A *
c
o
'3
E
LU
3
X
Q
"s
it
u
«A
>nable Wor
5!
01
oc
;
«
it is recommended ili.il Ct IEMDAT6 be used to calculate emission estimates direi
S
above scenai
1
|
>,
nidcant
?
7
3k
ilues var
ual input vt
••
«
^
o
Z
D-1
-------
UJ
O
IU
I
O
!!
1
1
w
1
u
S
55"
1
5
w
_
j
1
$
j
j
j
[
OR
ta
o
m
v
85
i
rs
i/i
p
S
«•-
i
Ot
IN
m
tft
!£
-
S
«N
M
IN
M
§
e
(•t
IN
at
S
m
IN
m
in
|
09
IK
S
OB
m
e
3
S
e
S
|
OB
^
S
rx
*
at
S
IN
m
m
IN
IN
Z
p
r
o
IN
O
IN
IN
IN
p
IN
~
|
09
00
s
>0
I/I
i
K
s
IN
IN
i
i
09
J
IN
«
kO
£
-
S
S
(N
IN
Z
S D
8u
* di
sl
> *
-.
« «
"5
i
^
a
s
I
9
I
|
w
U
!
i
o
p
p
w
p
l*»
^
S
p
p
01
IN
m
IN
m
IN
«
IN
m
at
S
IN
S
e
Z
IN
IN
IN
m
IN
m
p
*
IN
m
p
1
S
m
_
m
r
S
S
P
I
09
m
S
V
m
m
in
*
S
p
1
o
p
IN
e
IN
P
IN
2
m
P
IN
P
Z
5
3
fK
m
P
|K
5
ei
m
X
IN
"^
IN
a
IN
m
IN
S
Ot
S
IN
P
.
i
9
~
*
S
I
J
*
1
1
Cooslant
1
i
!
P
p
oa
Z
IN
i
IN
S
p
p
OB
4
P
IN
P
P
|
m
5
o
S
0
p
MHMB
•-
i
i
P
e
i
z
ijO
p
21
p
p
00
z
OB
fN
S
p
p
i
IN
i
a
p
e
Z
0-2
-------
Appendix E
Emission Rate Estimates
Oil Films on Storage Impoundments
-------
O
tc
O
O
O
s
UJ Ul
fflvn
<
«
«
CO
«
«
r*
VD
in
v
rn
IN
-
0
u>
00
r^
us
in
«
m
rx
-
Modeling
Parameters
^
O
V
o
^
0
«•
o
V
0
V
o
^
o
«
o
V
o
V
0
V
o
V
0
V
o
9
0
V
0
*
o
»
e
^
o
«
0
[Area (acres)
o
UJ
IN
P*.
IN
0
UJ
rst
r^
S
UJ
fN
r»*
rj
O
UJ
rN
rx
(N
o
UJ
CN
r*».
*N
0
UJ
<
o
o
o
o
o
o
o
o
o
o
o
o
o
0
o
o
o
0
o
1 Wind speed (mph)
-
-
-
a
r%
O
in
-
-
-
-
-
-
-
-
-
-
-
-
-
Icalculational period (yrs)
fN
oS
"~ ii
U
Input assumptions.
- Oil (fraction of waste)
- Molecular weight of oil
«
_
o
— 01
01 i?
w Of
18 w
.
-------
t/t
o
I
(L
I
35
<
UI
UI
<
1
!'
S
s
£
*
5
m
91
u
W
~
I Raoult't Law Surrogate
„
o
o
s
o
m
e
OB
OB
IN
en
*
en
o
o
0
3
z
(N
o
o
00
o
en
o
i
IN
en
en
m
S
0
0
o
5
OB
(N
O
o
00
in
o
3
m
0
i
IN
3
i
S
0
e
UJ
3
OB
i
o
o
OB
in
e
S
o
in
OB
«
a
e
0
9
UI
3.
i
i
o
o
OB
in
e
en
oo
m
O
r*.
in
S
m
s
S
0
e
0
UJ
| MVMB
o
0
2
o
0
5
s
3
o
e
1
OB
I
i
o
o
2
2
3
—
S
UJ
o
2
2
S
'
o
0
OB
in
0
1*1
0
I
IN
3
en
m
en
3
e
o
UJ
| VHVHB
o
o
00
m
0
0
OB
oo
IN
3
en
m
en
3
o
o
UJ
| VHVLB
•Q ~
lOYedis
(Case IS)
y» «T
5 I
m w
1 s
-. "
-
s
,3
0
S
37
u
1 Raoult't Law Surrogate
,
o
en
m
o
en
o
m
in
OB
en
o
en
o
I
Z
„
o
»
0
a
S
in
00
en
3
o
en
o
OB
i
e
3
o
S
e
m
in
OB
i
m
e
en
o
CD
i
e
m
0
m
0
en
m
OB
OB
I
I
O
en
o
I
Z
S
o
i
e
en
S
a
i
^
in
3.
i
e
en
o
| MVMB
„
0
o
S
o
3
S
00
I
~
o
en
o
OB
I
,
o
2
2
2
LU
O
2
-
o
IN
^
in
2
0
S
'
S
e
o
e
m
in
OB
en
m
O
en
O
| VHVHB
e
o
e
1
Wl
CO
en
IE
e
en
e
| VHVIB
2
S
u
2
w
S
j
1 Raoult'tLaw Surrogate
-
to
00
I
o
2
s
I
«n
i
«
o
2
I
| HVMB
m
-
v
o
S
S
OB
3
S
e
2
S
OB
S
O
2
i
| MVMB
in
m
o
s>
1
OB
0
UJ
r*.
2
UJ
O
o
m
^
§
en
en
0
2
I
| VHVHB
S
en
S
o
S
UJ
8
I VHVIB
E-2
-------
Appendix F
Emission Rate Estimates
Mechanically Aerated Impoundments
-------
o
z
o
a.
u
z
Z
O
O
o
2
o
l/l
l/l
CASE NUMBERS
«
rf\
'N
«
«
IN
IN
«
«
IN
O
IN -
C*
00
^
>£
m
v
m
IN
-
O
0<
CO
^.
VO
in
*
m
(N
-
Modeling
Parameters
^
O
^
o
«
0
«
0
9
o
«
o
«
o
V
0
w
o
.
I
e
3
M«
c
f
5
7
v»
i
LW
?
i
c
able Best Case (m
§
I
9
«
^
«
5
o>
5
2
•«
w
Q.
^.
O>
C
1
1
i
Emission Condili
15
V
a.
H-
J
«
rce area)
=
"S
w
Q.
9
sumin
I
*t
c
o
7
v«
C
UJ
?
i
X
i
able Worst Case (
§
i
Of
*
•
-3
•3
O
Z,
:enarios i
x
01
1
^
o>
~
O
*^
>*
V
1
*
1 input values var
i»
3
O
Z
F-1
-------
I/I
z
O
a.
a
u
H
!=!
t/1
Ul
S
l/t
ae
O
2
00
-
r*.
u»
If
(Case 5)
(Case 4)
a
w
^
u
«
u
Henry's Law Constant Surrogate
r»
«J
§
en
m
00
o
IN
rsi
3
§
»
00
z
2
en
*
3
kO
m
PS
IN
in
$
S!
5
z
•"
^
2
5
OB
Z
CO
*
S
i
rx
VO
IN
00
3
00
| MHMB
en
*
00
en
en
(N
m
fN
fN
CO
en
*
en
| HHMB
00
en
O
"
00
X
00
«
en
en
VO
(N
O
00
\f>
en
?
| MHHB
en
*
CD
en
to
io
(N
IN
00
*
en
00
Z
s
U
a
u
a
u
V
-
u
§
u
en
«
S
Henry's Law Constant Surrogate
on
8
UJ
i
«
*
o
•
z
00
*
S
CO
en
CO
S
en
$
3
e
OB
z
o
(N
•"
00
en
IN
^J
0
'
rs
0
3
(N
en
0
S
3
o
| MHMB
S
cn
"
e
o
en
m
$
S
3
0
z
_
00
e
UP
IN
£
co
S""
e
1
CO
«n
en
»
00
CO
m
on
en
*
O
I
S
en
=
I
oo
cn
^
O
00
z
V
u
a
u
(Case 21)
70 Years
(Case 20)
> S
2 w
C o7
n "~
in u
* Of
-J
Henry's Law Constant Surrogate
o
en
o
1
o
en
^
§
00
i
2
9
0
i
"
3
O^
oo
i
IN
•"
o
o
o
cn
2
•
—
OB
Z
;
rr
IN
O
S
en
5
5
| MHMB
uS
00
(N
o
00
en
9
en
'
9
Z
i
CO
o
o
U1
oo
9
9
CO
Z
Z
9
cn
0
cn
en
cn
cn
en
en
CO
I
3
a
o
a
e«
9
e«
3
a
cn
5
z
F-2
-------
Appendix G
Emission Rate Estimates
Diffused Air Systems
-------
O
ui
V/l
O
u>
u.
5
W1
u
LU
Q.
tt
O
I/I
Q
O
K
t/t
I
CASE NUMBERS
4
«
«
«
^
IN
«
«
«
*N
fN
«
e
m
p
LU
r^
VO
nn
p
LU
^
VO
m
p
\u
r*.
<0
m
p
LU
rx
ve
9
LU
rx
ve
m
p
iU
r»
>e
m
p
LU
P»
ve
m
p
UJ
^
ve
m
p
LU
^
vO
n^
p
LU
r«
V0
9
UJ
(^
vb
m
p
LU
rv
vO
m
p
LU
r^
VD
m
p
LU
^
VO
1 Area (acres)
J3
^
IN
«
V
«
^
V
•»
^
«
^
*
»
«r
v
«
v
»
VO
in
w
IN
1 Depth (m)
rw
*
vO
V
V
V
V
V
V
V
«
«
V
V
V
ve
in
»
m
V
9
V
*
Retention time
(hours)
§
o
^
o
o
o
o
o
§
I
!
i
I
i
i
I
«
IN
I
O
1
I
1
1
1
I
I
1
Constituent
concentration
(ppm)
ji
3
O
*T
O
O
m
O
0
V
O
O
3
O
3
O
§
a
in
o
e
l/l
3
0
3
o
(*l
o
o
s
0
s
0
«
o
o
s
o
S
0
s
o
3
o
3
0
5
o
3
e
3
o
3
a
Submerged air
ilow(m)/sec)
ut
IN
1/1
IN
LO
IN
U1
r*
vn
IN
in
(N
in
IN
Lfl
IN
VTI
(N
in
(N
in
(N
LA
IN
U1
IN
in
(N
in
(N
in
fN
in
P*
in
w 2
II «l V
3 £« *
fsll
3£el
a. O J£ £
•"
"»
S S 6
111
•3
z
o
c
1
iffi I
€ S oi£ "I
£r c c LU —
l.|o
S|l i
.7 =11
. _ I ~ 5
H " S j
Ol
>.
5
"!• I
III I
« 41
41 41
41
i
G-1
-------
1/1
I*
o
Ul
«
w
p*.
s
2
w
S
u
^
2
RT
w
I
|
1
Constant
1
m
on
01
m
0)
*
01
OB
m
OB
5
3
0
m
3
e
1
in
en
m
01
00
fN
fN
in
§
,
o
in
o
IN
in
e
i
I
a
Z
in
e
S
0
S
vu
00
i
o
a
o
i
e
|
IN
VO
a
o
ve
VO
2
ve
MHMB
VO
(N
vn
in
-
e
in
in
in
in
in
in
in
vn
OB
fl
O
in
m
o
e
S
UJ
in
e
in
m
o
o
o
e
|
ve
m
00
|
e
OB
00
OB
00
i
i
oo
s
oo
in
o
en
OB
IN
en
00
IN
en
00
IN
en
00
(N
Z
G-2.
-------
Appendix H
Emission Rate Estimates
Land Treatment
(Emissions After Tilling)
-------
<
t/1
o
G?
5
UJ
£
O
«/»
O
SH
*&
UJ
t/t
s
O
K
o
«/»
O
o
l/l
t/t
I
UJ
CASE NUMBERS
Modeling
«
•
«
m
rx
•
«
•
IN
IN
•
•
IN
O
rx
O<
OB
r^
ID
I/I
V
m
rx
-
0
CO
rs
V0
in
«
m
rx
-
Parameters
IN
10
rx
US
rx
VD
IN
10
rx
VB
IN
vO
IN
vO
IN
VO
rx
lO
rx
vO
rx
10
rx
10
rx
10
rx
VO
rx
16
(N
vO
rx
10
rx
VO
rx
IO
rx,
IO
rx
VO
rx
vO
rx
VO
Aiea (acres)
o
o
a
I
|
1
§
«
§
a
o
§
8
eo
§
GO
1
|
|
|
|
|
I
|
|
|
I
I
I
I
Annual waste
(oil & water)
throughput (10*g/yr)
o
m
o
IN
O
o
o
o
o
o
0
o
o
o
o
o
o
o
o
o
3
o
IN
o
IN
Oil content of
waste(K)
s
1
fN
§
m
rx
rx
IN
IN
0
IN
rx
IN
IN
rx
IN
IN
o'
1/1
1
1
1
rx
1
IN
Constituent of
interest content of oil
(ppm)
in
wo
o
in
m
v
o
U1
o
in
o
I/I
s
o
in
s
s
o
in
in
SO
I/I
in
s
m
v
s
s
s
s
o
in
S
s
s
| Soil porosity (%)
in
o
IN
in
vC
o
IN
o
rx
O
IN
O
IN
in
vO
O
V
8
l/l
O
IN
g
O
rx
O
rx
0
IN
O
rx
O
rx
g
O
>
. oi-o
« > S "« <• S O
- w d. -3 « 2 >
a_2 «, S oa s >
II 1
s
«
o
H-1
-------
O
v/i
2
UJ
ZQ
!!
i
S
S
»
u,
„
S
u
S
0
*•
I
S
w
.
a
•
i
1 Raoult'i Law Sun
I
i*
in
00
e
e
„,
S
e
!
i
e
„
o
S
o
i
s
s
0
m
e
„
0
§
8
o
£
b
0
i
z
*
i
e
S
m
0
o
m
O
8
b
&
OB
>
3
fN
0
S
o
„
e
e
m
«N
e
m
in
e
S
o
i
OB
i
s
3
o
Z
o
„
o
e
m
m
in
b
e
01
e
8
!
§
in
2
e
09
m
0
in
S
e
00
m
S
e
$
b
S
OB
I
m
O
hU
S
m
O
*
V
3
i
s
«r
EO-309E
9
9
3
S!
OB
8
i
o
s
e
b
00
IN
i
0
8
b
e
| VHVHB
£
i
e
S
m
O
e
S
2
e
0,
b
e
e
OB
I
to
s
5
jj
w
i
u
*
5
w
I
I
s
I
S
1
•
*
I lUoull't Law Sun
I
in
IN
e
e
10
o
e
S
b
—
e
m
O
i
z
m
m
e
in
^
e
b
in
«n
b
in
b
in
O
i
ot
u"t
3
e
in
m
O
S
m
O
m
e
£
b
S
b
S
o
OB
i
o
e
§
e
in
b
b
IN
b
S
e
S
o
OB
i
m
S
e
2
IN
b
o
IN
b
N
o
in
IN
b
S
e
^
e
1
§
m
£
0
s
o
m
m
b
r.
3
e
b
S
e
01
S
o
OB
I
S
i
s
i
—
I
9
IN
m
e
Ul
IN
in
IN
m
O
5
s
;
OB
S
^
2
O
m
e
b
b
o
,0
b
$
e
1
m
s
e
in
m
O
s
m
e
3
m
b
m
e
3
e
m
O
^
IN
*
*•*
IN
S
u
-.
I
(Cat* 20)
2
a
«
1
2
1
1 Raoull's Law Sur
OR
00
5
o
r»
o
e
m
e
I
b
S
e
3
e
I
i
00
n
o
00
o
o
m
e
e
S5
e
in
O
1
N
a
m
o
00
e
o
S
0
0>
in
m
O
S
b
in
m
e
oa
i
i
~
§
o
,
o
s
o
s
o
s
o
8
o
1
j
V
IN
O
o
o
«n
IN
O
01
IN
e
£
b
IN
O
| MVMB
»
r»
S
o
^
0
0
s
m
b
s
o
OB
m
b
m
e
OB
I
IN
O
IN
S
vu
9
UU
0>
9
9
S
i
s
i
OB
S
00
CD
VO
1/1
e
oo
0
o
g
b
o
in
b
in
m
e
| VHVHB
IN
00
e*
00
s
m
o
00
o
o
s
m
O
S
m
e
S
o
m
e
I VHVLB
H-2
-------
Appendix I
Emission Rate Estimates
Oil Film Surface on Land Treatment Units
-------
UJ
oe
O
z
«/»
Ul
«/»
O
K
z
Ul
*
Ul
U
«
§
O
i
o
f
I
(/>
CASE NUMBERS
«
«
«
•
2
•
«
«
«
«
•
^
>e
in
«
m
IN
-
O
»
oe
rv
(0
in
*
m
IN
-
Modeling
Parameters
(N
1C
IN
16
IN
10
IN
10
IN
10
IN
.10
IN
10
IN
10
IN
10
IN
10
IN
10
IN
10
IN
10
IN
10
IN
10
(N
10
IN
U>
IN
10
-------
h-
o
UJ
~
*
U
U
*
U,
,«:
u
"**
— .
S
u
— „
_
1
s Law Sun
^
•
"*
,
^
o
00
IN
(N
in
S
5
m
1
ea
IN
,
^
O
oo
S
^^
.«
is
_ I*
—
—
.<
—
s
0
10
"
_
1
ft Law Sun
«•»
*
^
*•
*
r
m
e
in
CO
on
e
r»
in
<0
.
X
*
^
**
*
«
on
e
in
m
e
S
.
to
ea
i
^
*•
^
2
r
**
o
tn
S
m
e
v»
ea
i
co
oa
00
o
ea
m
J^
IN
ft
3
1
0
e
e
O4
e
04
Pk
OK
0>
s
N
1
*N
(N
£
r-4
m
lift
S
ea
o
UJ
*"
m
O
UJ
on
O
UJ
in
m
UJ
in
^
S
o
0
m
Ul
IN
a
in
S
on
IN
oa
*
•«•
,
*~
*
•»
on
O
in
S
on
e
-
VO
VHVHB
*
^
^
*
"
on
O
in
eo
on
O
in
-
VHVIB
en
J
2
S
c
i!
u
•
i
3
in
1
1
04
O
ve
-
09
i
s
p*
*
-
i
VO
s
VO
en
-
01
VO
s
s
1
i
IN
X?
MVMB
m
in
ea
S
aa
i
09
O
O
IS
-
3
Ui
IN
i
§
in
•e
m
VHVHB
r*
i
in
1C
m
VHVL8
I/I
UJ
I
QC
O
1-2
-------
Appendix J
Emission Rate Estimates
Closed Landfills
-------
«
Q
Z
UJ
O
\
Q
O
"7 "*
-*tu
UJ >
ec
O
O
3
LU
o
O
<
O
I
CASE NUMBERS
«
•
«
«
nn
r*4
«
«
•
IN
IN
<
•
IN
O
IN
en
00
r*
VO
in
v
m
IN
-
O
en
00
r*
*c
in
f
m
IN
-
Modeling
Parameleis
HI
m
in
1*1
U!
m
_m
1*1
in
m
in
m
in
m
in
m
in
m
in
m
in
m
in
m
in
m
in
m
4
in
r^
O
in
in
IN
in
in
IN
in
IN
in
(N
in
IN
in
IN
in
PM
in
IN
in
IN
in
(N
m
IN
in
(N
in
IN
IAir porosity of fixed
waste (%)
PW
IN
(N
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
»N
1 Waste liquid density
(y/crr>J)
00
00
oo
oo
00
eo
00
00
00
00
oo
00
00
00
00
00
00
00
00
00
00
00
00
| Cap air porosity (•/•)
^
^
«
«
«
«
5
^
«T
^
^
^
V
«
«T
V
5
5
v
v
«
»
v
(Cap total porosity ("i)
m
in
in
in
in
in
m
in
in
in
in
in
m
m
in
in
in
in
in
vn
in
in
in
(Temperature beneath
cap('C)
!•«
0
«^
o
m
o
^n
e
m
o
m
O
1*1
O
m
O
pn
O
m
O
m
O
1*1
O
m
O
m
0
m
o
m
o
1*1
S
m
O
m
o
1*1
o
m
O
m
O
m
0
1 Typical barometric
pressure (mb)
T
V
V
V
V
V
V
^
«J
^
^
«
«
«T
^
T
1
V
^
»T
V
«
«
1 Typical barometric
pressure drop (mb)
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
in
5 — =
U o •- ~
0 o = ^
13 II 4* C 3
0^ ^ ^
s -3 - ?
u c S * '
us S S ' 5
t- § TJ -5 _
50y C =" -
2 ^ ~ —
5 < O 5 5
. , ,
w
C
Ol
o
b
tl
** C
$ 0
Oy 5
C nM^ ^
"5 o TIT
fcl
3 O 0<
5 S 9
^ 2 ^2 w
|?2|
- * !1 5
a " 5"
)% of the organics in wai
light percent oil in waste
light percent water in w<
omelric pumping time =
ilecular weight oil = 147
o 5 5 <5 S
—
"^ i
a/ "
c ^
u ^
i 2
S 3
w ^
5. "? *
M ^ ?
o ~
C Oj =
III
"* i 2
Q >• a.
S §"5
C -»*
111
.enable Best Case (minim
cal Emission Conditions (
.onable Worst Case (max
S 5. 5
4l >* 41
ae h- oc
« * «
* * *
_
*
*"*
i*
j:
I
5
i
-
•=
—
*
~
z
~
i
-5
S
I
_I
^
E
5
u
5
—
S
5
c
Oy
v
i
o
.2
Oy
S
£
5
i
w
il input values vary signif
'W
3
J
o
Z
J-1
-------
a
z
o
UJ
V/l
o
3*
fflvo
<0
o
i
CO
01
•a
u
_
S
ij?
OJ
in"
S
O
^
|
m
s
u
_
s
u
—
3
«
5
in
15
1
£
0
LU
US
IN
0
UJ
IN
O
UJ
O
UJ
a
IN
IN
o
UJ
O
UJ
a
o
UJ
S
0
+
5
IN
a
X
X
0
vu
IN
0
o
UJ
§
IN
S
LU
IN
IN
O
UJ
vO
IN
IN
O
UJ
O
UJ
O
^
O
UJ
0
+
UJ
S
fN
a
o
LU
1^
O
UJ
VO
*
O
LU
g
O
UJ
en
in
IN
0
UJ
IN
O
LU
3
O
UJ
S
a
o
+
S
*
a
X
0
UJ
a
0
UJ
in
IN
O
UJ
in
0
UJ
8
UJ
O
8
m
S
UJ
IN
a
o
«
«
8HAVMI
0
LU
r*.
O
UJ
a
o
UJ
a
en
0
UJ
VO
8
vu
m
a
8
UJ
"
o
?
O
9
a
>
o
UJ
vO
O
vu
O
UJ
a
o
a
o
UJ
s
a
8
UJ
8
vu
IN
en
IN
O
O
UJ
in
O
S
a
a
>
in
O
LU
S
UJ
in
O
LU
in
m
in
O
UJ
vO
O
UJ
in
O
s
UJ
en
1*1
in
in
o
UJ
m
a
IN
s
in
in
a
_j
IN
O
UJ
O
3
UJ
IN
IN
O
UJ
IN
0
UJ
in
**
IN
0
UJ
en
IN
o
UJ
*
IN
O
UJ
in
IN
IN
e
*
UJ
a
>
>
IN
O
UU
IN
O
UJ
en
IN
IN
o
LU
VO
IN
IN
O
VU
IN
O
m
0
£
IN
O
UJ
c?
en
IN
o
UJ
IN
O
+
UJ
IN
a
>
>
to
OJ
in
S
^
x
19
u
m
2
s
$
—.
1
o
i
.
u
S
*
v*rt
1
1
"
o
vu
O
UJ
'
o
UJ
IN
8
UJ
in
a
o
UJ
5
0
UJ
IN
O
UJ
«N
O
•*•
UJ
<"N
03
X
0
UJ
S
O
UJ
O»
O
UJ
if
fN
(N
8
vu
a
0
LU
O
UJ
^
O
UJ
IN
O
UJ
(N
*N
m
r
r*
O
UJ
IN
O
S
a
o
UJ
O
r*.
O
UJ
in
o
vu
o
O
'
o
UJ
3
0
+
*
a
X
S
UJ
O
IN
O
UJ
in
IN
a
o
LU
O
vu
O
vu
in
O
UJ
in
0
vu
in
O
UJ
in
a
>
§
UJ
O
O
UJ
0
LU
0
UJ
O
IN
0
UJ
a
en
o
UJ
m
a
?
o
vu
m
a
i
8
UJ
8
UJ
in
0
UJ
a
o
a
o
vu
IN
O
UJ
a
o
S
UJ
S
a
o
o
a
o
UJ
IS
a
a
>
in
O
UJ
in
O
UJ
m
a
IN
in
O
UJ
in
vn
VO
LU
en
IN
in
in
0
UJ
in
in
in
o
vu
in
in
in
U.'
Lfl
in
O
UJ
un
ut
CO
^
r*
O
UJ
3
UJ
in
IN
IN
O
UJ
en
S
UJ
*
IN
O
LU
en
3
3
"
3
UJ
3
"
1
0
UJ
m
a
vO
IN
O
UJ
in
*
IN
O
LU
vO
IN
O
u.
en
3
LU
in
vO
IN
O
UJ
'N
O
UJ
(N
0
UJ
(N
Ik/HVLB
3
*^
Co
* o;
>- »*»
^ J?T
o> v
£ a
10 •"
> s
m u
5 —
2 s
P* /j
i
S
(•A
I
|
4
IN
O
LU
S
o
UJ
rs
8
UJ
U9
0
UJ
S
o
UJ
IN
O
UJ
IN
O
UJ
IN
a
X
IN
0
UJ
IN
IN
O
vu
IN
(N
8
vu
r»
vo
0
vu
a
o
vu
*""
o
UJ
IN
IN
O
UJ
IN
IN
a
X
IN
0
UJ
vn
en
O
UJ
O
0
UJ
™ *
0
UJ
O
UJ
!f?
o
!
o
UJ
O
*
a
X
8
LU
IN
IN
en
o
UJ
in
O
LU
0
UJ
in
O
UJ
in
o
LU
in
o
LU
in
e
i
8
LU
a
o
UJ
CO
en
IN
0
LU
CO
en
O
LU
§
o
UJ
m
0
UJ
g
pn
O
UJ
a
en
m
IMVMB
o
LU
O
"
O
LU
9
O
O
LU
IN
O
fN
0
UJ
5
ao
O
UJ
§
a
o
vu
§
a
o
LU
a
a
i
o
£
uT
O
UJ
to
O
LD
U*
O
u.
O
LU
ijTl
S
vu
in
in
in
O
LU
in
in
a
-t
o
^
o
£
O
^
O
o
UJ
^
(N
0
UJ
00
O
UJ
^
a
X
o
LU
O
vn
IN
0
O
IN
O
=
rs.
= '
- i
*N
UJ
IN
O
LU
0
o
LU
fN
IVHVLB
J-2
-------
Appendix K
Emission Rate Estimates
Open Landfills
-------
Q
O
O
tsi
Wl
O
(SI
V9
Q
O
5»
VI
i
S
UJ
O
2
UJ
i/i
5
«
«
•
«
nn
U)
in
»
m
IN
-
Modeling
Parameters
m
m
in
m
in
m
in
m
>fi
m
LA
m
1^
m
LT1
r«
1/1
m
ifi
m
in
«
i/i
m
1/1
m
in
m
in
m
in
m
in
m
in
m
in
m
in
m
in
m
in
m
in
m
1 Area (acres)
-5
rn
in
r%
m
in
r»
in
^
in
r»
in
r*
in
^
in
rx
in
^»
in
^.
in
^
in
p^
in
r»
in
rv
in
i»
in
^
in
^
in
r«
S
in
in
K
m
1 Waste-bed thickness
|(ft)
§
o
v
o
o
^
o
V
o
«
o
V
5
o
n
0
§
§
§
s
s
§
s
§
0
§
e
»
o
0
(Constituent content
of waste (H)
o
in
m
IN
in
in
IN
in
rx
in
IN
in
(N
in
rN
in
IN
in
IN
in
IN
s
in
m
in
rs
in
in
(N
in
IN
in
rx
in
IN
in
IN
in
(N
1/1
fN
in
IN
IAir porosity of fixed
waste (%)
o
o
in
in
^
S
o
in
2
o
in
in
r**
S
in
IN
o
o
in
O
in
S
a
s
a
s
s
a
a
o
in
a
Total porosity of fixed
waste (%)
r*i
f*4
rN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
1 Waste liquid density
(g/cmJ)
in
IN
m
IN
in
IN
in
IN
in
fN
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
in
rs
in
IN
m
IN
in
IN
u-
IS
in
IN
in
IN
in
IN
in
IN
in
IN
in
IN
1 Air temperature (*C)
-
-
-
o
rx
O
in
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Cakulational period
(yrs)
i
TS 1
f
2 e
i S
*fl ^
?sl £
3 ?r ^ W
HI
«• y S
sal
5 Tl 5
51
o
j:
o
i&Hi t
flTr5«-.i^fl 3
Hlsii! 1
lilcI'U 1
Si'S^.SwOs I
•|i|ll|i| t
Q.— fl vi fl OB J> > 3
I Ifif11| s
fl ?Z ? o c -a e
ual
o
K-1
-------
O
z
i
o
s
u
S
u
3
u
1
3
u
IN
1
*
?
Law Sur
««
g
a
m
§
1
1
i
s
IN
S
1
^
*
i
g
0
OB
S
2
m
2
5
m
2
OD
i
91
ve
IN
I
oa
i
s
m
oe
oo
1*1
00
oa
I
«r
S
§
in
3
i
wt
^
5
oa
i
s
ve
3
o
3
i
3
•»
3
I
S
S
in
3
S
in
.
r»
oa
>
0
o
o
e
o
o
e
in
oa
in
I
IN
00
in
IN
01
m
IN
oe
m
IN
m
IN
^
f*
VHVHB
oo
S
HI
in
PS
pn
IN
p*
in
p»
oa
S
0
PS
VHVIB
K-2
-------
Appendix L
Emission Rate Estimates
Wastepiles
-------
I
o
u
z
u
ui
a.
t/»
S
ear
O
2
UJ
O
O
CASE NUMBERS
r*
«
«
•
IN
IN
«
«
IN
O
IN
(Ti
00
rv
ifi
ui
v
m
IN
=
o
e*
00
r*
to
in
^
m
IN
-
Modeling
Parameters
o
o
o
e
o
o
o
o
o
o
o
o
o
0
o
0
o
0
o
0
0
o
e
1 Area (acres)
o
m
r^
0
IN
U1
O
^
O
^
o
V
o
^
o
^
o
^
o
«
o
V
o
V
o
V
§
o
§
o
V
o
V
o
(N
in
§
m
10
m
O
m
^
I Turnover rale (per
|year)
9
a
«
o
o
V
§
o
«
?
o
«
o
«
o
«
o
§
o
«
o
«
i
s
s
s
o
§
o
V
§
3
1 Constituent content
of waste (%)
o
m
in
IN
in
m
IN
m
IN
in
IN
in
IN
in
IN
1/1
rsi
U1
IN
in
IN
O
in
in
m
i/i
IN
1/1
in
IN
in
IN
i/i
IN
1/1
fN
Ul
IN
in
IN
Ul
IN
in
IN
IAir porosity of fixed
waste (•/.)
0
o
in
in
r^
O
Ul
O
Ul
o
Ul
o
in
Ul
r^
O
in
Ul
IN
o
o
Ul
o
Ul
0
Ul
s
o
Ul
o
Ul
s
o
irt
S
S
o
Ul
o
Ul
1 Total porosity of fixed
| waste (%)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
| Pile height (m)
1^4
rs(
fN
r«*
rsj
<%
fN
fN
rs*
(N
fN
f%
fN
(N
(N
(N
fN
fN
fN
(N
fN
rw
fN
1 Waste liquid density
(g/cml)
t/i
IN
in
IN
Ul
» *
o»
c
E
3
**•>
r»
v*
C
O
1^
E
^ ^
c
E
easonable Worst Case (maxi
tt
«
*
«
0*
"5
C
£
o
5
£
-
0
f£
>w
4i
O
•^
0.
5
s
*s
Q
C
actual input values vary sig
;S
2
o
z
L-1
-------
LU VU
5)
j
S
S
u
«
Si
<*
w
in
il
w1
i
w
s
II
w
^
1
^
1
•
1
s Law Sur
1
PS,
a
o
PM
m
V0
IN
S
S
VO
(N
IN
<»
m
£
Z
01
*N
^
IN
8
m
i
m
«»
2
IN
i
10
in
ill
IB
^
ve
5
m
I
PS
PS
r*
a
§
p*l
p».
IN
PS
;
^»
m
3
eo
i
*
Hi
m
40
PM
m
in
ft*
-/
m
s
i
i
m
pn
in
PH
dt
g
o
IN
ft
m
m
•A
^
»
eo
i
m
OB
m
m
in
IN
IN
in
in
r«
5
m
S
ea
>
m
0
o
o
0
o
o
e
V
e
a
^
o
§
—
i
1
$
pn
ps
00
^r
t
§
«^
i
VHVHB
o
(N
oo
IN
1
IN
in
•
vo
IN
^
S
in
IN
in
a
g
m
4*
i
a
^
Z
VB
I
W
in
«
I
y
i
u
m
2
w
(Case 12)
I
3
I
9
I
•
1
s Law Sur
M
v
r»
*
ft
IN
IN
Ifl
°
m
PS,
s
ft
«
1
p.
S
en
IN
2
in
S
~
i
§
i
a
a
i
IN
tn
a
e
r»
r*.
r«
S
in
«
P*
S
in
r.
i/>
IN
r>
IN
in
a
a
i
m
ps
O
S
PM
tn
ft
10
m
in
10
(N
e
a
i
en
m
0<
e
rs
«r
3
m
**
pn
W
g
V0
0
|
o
(x
in
in
IN
in
rs
IN
»
W
tn
s
a
e
a
i
o
IN
O
0
0
e
e
o
e
o
a
1
0i
IN
m
m
m
1
|
»
P*
s
m
§
g
r>
m
«
IN
VHVHB
a
3
in
IN
in
a
1.
m
r»
r*
*
in
IN
a
m
K
ft
IN
in
ft
VHVIB
m
IN
«
3
u.
IN
IN
^
U
(N
S
0
(Case 20)
0t
i
9
i
s
i
•
?
$ Law Sur
1
r»
i
a
S
«N
«•»
«a>
ft
i
S
«
VO
IS
i
»
I
PM
i
tN
i
IN
on
IN
HVMB
^
IN
m
IN
^»
^
IN
-
IN
IN
(N
r«
^.
IN
1*.
IN
IN
o
p»
IX
IS
a
|
00
1"!
S
ve
IN
O
10
(N
«
fS|
10
IN
o
10
IN
i
2
a
i
s
e
S
e
IN
S
IT»
S
a
>
?t
e*
vn
o<
vn
fN
e
in
IN
in
IN
in
IN
in
in
IN
a
^
>
p*.
i*n
O
§
0
e
e
e
F*
e
a
1
vo
0i
V
ve
S
rs,
u-i
!E
s
Ps
1
PR
1
VHVHB
^
*•••
«
E
IN
in
CD
Ifl
/N
£
fSJ
£
IN
in
eo
rN
m
eo
VHVIB
I
o
L-2
-------
Appendix M
Emission Rate Estimates
Fixed Roof Tanks
-------
u.
O
O
ee
O
UJ
X
1/1
O
u
-
eo<
O
UJ
o
o
ee
CASE NUMBERS
«j
S
•
s
o
2
00
-
£
1C
2
nn
fN
-
o
.
00
-
v0
I/I
*
m
IN
-
Modeling
Parameters
0
o
IN
I
o
(N
3
§
§
§
§
1
8
8
§
§
§
§
§
§
§
1
1
S
§
o
IN
0
Tank diameler (f I)
o
o
o
0
o
o
o
o
o
o
o
o
S
o
in
S
o
m
o
0
o
§
9
o
§
a
•9
[Tank height (It)
S
fN
CC
5
IN
IN
s
IN
IN
5
5
~
-
s
IN
1C
R
0
-
fN
PS
VD
i
VO
IN
Turnovers (per yr)
o
o
3
VO
3
ve
m
3
o
o
§
U"t
8
§
o
1/1
o
VO
VO
m
VO
3
VD
3
VO
VO
VO
VO
3
Throughput ( 1 01' gal/yr)
-
-
-
o
o
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Calculcitional period (yrs)
^r
01 -1
M
» =
•*£ 2
^ •.
"c =
w &
3. -
at * 5
= 5 S
E-g 2
1 2 3
|lf
1 i ±.
£!§
3 8 E
11
^ ~ 91
SI "e J
<3 w 5
£ §|
03 ? S
Reasonable
Typical Emii
Reasonable
•
• • «
IN
«T
o.
i
•o
S
c
S
o
w
Oi
£
^
0
13
c
K
41
g
"5
ti
c
5
c
o
c
9
£
ut valu
II aculal inp
HI
o
z
M-1
-------
o
o
o
UJ
X
o
LO
on
1^
0)
3
i?
«
10
u
in
u
u
m
S
w
^
u
*
^
|
f
i
a
r
1
1
«
5
0
r
o
{
?
5
*
•
r
1
1
*
•
r
1
1
^
™
r
2
$
|
:
•
f
i
f
•
r
I
r
«
| •
21
JJ
IHVHB*
B
Z
B
>
8
*
UJ
8
UJ
*"
8
*
00
*•*
0
UJ
f**.
a
8
+
LU
en
m
8
»
LU
in
8
IU
m
8
UJ
03
8
LU
r*
O
IU
m
r-.
8
Ul
O
Ul
U3
m
S
LU
S
en
S
uu
en
S
>u
""
i
o
o
*
LU
Ul
IN
8
Ul
**•
8
10
^
0
IU
OB
8
>
LU
Ln
m
8
+
tu
OB
8
Ul
in
m
8
UJ
00
8
UJ
in
O
tu
•»
8
UJ
fN
o
tu
m
5
IU
in
00
S
en'
S
tu
in
OB
3
•U
•^
a
>
§
+
0
m
8
IU
IN
IN
8
Ul
tn
m
8
*
Ul
via
—
8
•*•
UJ
r»»
O
+
Q
—
8
IU
•*
8
Ul
IN
m
8
*
in
8
Ul
—
8
Ul
in
~
5
IU
in
ua
8
LU
O
00
8
uu
IN
O
IU
en
IN
ea
i
un
o
LU
en
1*1
0
LU
~
in
O
Ul
O
UJ
3
f.
3
Ul
m
•"
O
IU
in
3
IU
^
S
IU
in
8
Ul
3
UJ
V
VO
0
IU
in
'
3
IU
IN
m
in
UJ
i»i
3
0)
w
UJ
m
in
o
dbJ
**
SB
.BHAHA!
[VHVIB*
*
^^
in
19
U
^
«
***
m
u
—
1
§
19
U
3?
H
»
=
4
?
\
»
•
l
1
I
iaj
•
?
>
?
r
•
«
r
1
1
*t
m
?
I
1
!**
—
"
f
*
^
r
•
?
i
1
2
1
I
;
!•
Raoult'f L
Surroga
i
z
as
i
§
+
UJ
«n
§
09
—
8
UJ
33
8
UJ
OP
~
8
cn
*
8
+
a
—
8
Ul
m
8
LU
*
e
UJ
8
UJ
OB
•-
8
Ul
en
^*»
8
S
(N
8
IU
m
8
a
8
•*>
eX
«»i
8
*
3
^
I
§
^
UJ
a
8
co
—
8
IU
r»
•»
8
LU
OB
—
8
IU
in
*
8
»
IU
OB
—
8
OB
IN
8
*
LU
09
O
in
S
Ul
oa
••
8
IU
in
^^
8
4>
?4
8
*
tu
in
m
8
oa
8
IU
in
m
8
IU
in
^
i
O
+
UJ
*™
8
(N
m
0
tu
tn
8
UJ
m
8
LU
en
8
*
Ul
^4
m
8
IU
in
8
Ul
f*
8
IU
8
UJ
m
8
Ul
r
8
tu
tn
m
8
Ul
l**
8
m
8
Ul
f+
8
•u
f»»
|
o
LU
m
m
O
in
—
3
tu
OB
IN
m
O
IU
tn
—
s
LU
"™
m
O
Ul
in
—
s
3
•"
m
e
IU
in
jg
tu
IN
a
tu
in
—
3
Ul
m
«
Ul
•^
3
tu
m
3
in
3
Ul
m
m
0
•u
^
|
IVHVHB*
•
2
^
tj
m
IN
X
U
/N
—
fj
f
en
|
|
„
-
1
a
e
r
«
r
1
1
w
«
r
^
?
*
•
r
M
|
1
K
•
r
1
I
*
•
"
r
X
i
?
:
•
r
I
1
i
1
1
1
5
2.
Raoult'f L
Surroga
ea
i
09
Z
09
I
-
^
O
m
8
*•
LQ
UJ
8
*
UJ
C*
m
8
*
LU
oa
—
o
en
IN
O
tu
OB
en
8
en
8
n.
*
8
m
8
UJ
a
•-
8
Ul
8
Ul
a
8
Ul
m
8
*
o
tu
m
8
LU
09
••
ea
o
LU
09
Ul
in
m
8
2
5
tu
IN
8
*
Ul
a
*~
i
i
o
LU
UJ
in
0
—
8
Ul
•»
8
m
8
r*«
•"
o
a
—
8
IU
8
*
Ul
t*N
m
8
*
Ul
8
Ul
m
8
IU
"
8
Ul
m
8
Ul
f**
8
5
tu
IN
8
*
IU
IN
m
ea
>
o
UJ
O
—
o
r^
in
3
LU
—
O
in
—
in
o
m
tn
e
LU
in
oa
3
tu
m
~
e
IU
in
3
IU
m
e
UJ
uv
—
3
Ul
ft
e
LU
in
—
3
Ul
O
in
3
Ul
m
•»
O
IU
in
••
^
IVHVHB*
.81 AH A]
5
*»
$
S.
|
•s
«A
5
F
8
1
w
S
e
1
|
M-2
-------
Appendix N
Emission Rate Estimates
Floating Roof Tanks
-------
o
o
oe
o
u
O
a
o
CASE NUMBERS
«
•
«
00
r*
«
«
fs.
rsi
•
*6
fN
un
rsi
v
i%
r*i
r*
r*
m
vO
m
vO
no
vO
nn
vO
3
m
vo
m
vO
m
VO
no
vO
no
VO
no
VO
no
vO
m
vo
m
VO
no
vO
no
vO
Tank throughput
(I0*gal/yr)
-
-
-'
-
o
r^
O
in
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
1 Calculational period
(yrs)
01
•5
•o
K
— * X
^r ~
_ j^
c
"c 2
— . V
3 |
a~ ~
*- *• ?
?.* §
l!i
i^I
nj vj w*
T a.§
S|;i
si's
i 6S
"26
C *^ 3
3 "rr S
E * »
III
st Case (m
ion Condi
arst Case (
z*$
III
I
5
c
5
•s1
c
—
c
gi
^
>>
5
I values v
3
Q.
C
"fl
w
t
N-1
-------
2
u.
O
o
o
ee
O
vn
vO
s
u
if>
«T
•
a
^
x
rs
o
S
m
IN
O
UJ
IN
ve
o
S
m
rx
O
3
m
IN
9
UJ
IN
u»
S
UJ
00
IN
9
0
m
4
VO
S
ui
US
IN
9
3
IN
IN
O
UJ
IN
to
3
8
•
IN
e
3
IN
3
S
VO
3
Ul
in
3
Ul
•
(N
3
^
vb
3
Ul
(N
m
IN
e
3
IN
3
Ul
IN
<0
3
Ul
IN
i
I
fN
O
UJ
vO
in
IN
O
UJ
fN
ve
o
vu
IN
e
3
Ul
»
IN
a
.^
i
£
s
IN
O
UJ
IN
ve
vn
a
vu
t
%
IS
0
vu
IN
ve
S
^
ve
S
S
IN
e
VU
IN
ve
^
vu
IN
m
S
S
IN
O
UJ
IN
ve
^
ibi
a*
••
^
Ul
e
*j
o
SI
VO
s
Ul
IN
S
3
fN
O
Ul
IN
vO
r»
O
Ul
»
»
I
•
>
>
«
•j
•>
IN
41
^
U
S
IS
w
e
1
u
O«
J
i
r»
S
3
•Jl
!-
r
5
^
I
I
§
r
«
I
i
r
I
i
i£
<
l_
I
«£
«
|
1
!
!
^
1
*f
-J
•
OD
>
•
0)
>
•
a
^
»N
O
S
I'n
(N
0
u*
PN
un
§
*
s
IN
o
S
m
§
Ul
4N
t£
S
!
IN
e
VU
O
m
e
vu
f*
8
S
3
vu
o
m
3
vu
IN
VO
s
»
$
IN
9
Ul
O
m
3
vu
IN
ve
S
•
Ul
a
IN
a
vu
e
m
IN
9
vu
IN
UJ
o
«
VO
OB
i
rsi
O
UUI
«
f^4
f-St
0
UJ
r*
tn
O
UJ
*A
0*
*N
O
LU
00
r*
8
UJ
IN
vB
o
vu
vn
»>
S
£
IN
9
VU
1*1
9
vu
vn
**
IN
O
g
IN
3
UJ
F*
ve
o
»
«>
3
3
IN
IN
9
UJ
IN
vO
S
s
a>
IN
O
s
IN
IN
9
vu
r*
vO
9
vu
r»
ut
a)
i
rs)
p
UJ
NO
in
(•S
O
UJ
^»
in
§
vu
WN
IN
o
UJ
U9
VA
§
VU
IN
VB
§
*
UJ
^
(N
0
3
u"t
0
w
r*
S
$
IN
9
vu
ve
w>
IN
o
VU
««
U9
8
VU
ff>
IN
O
3
vn
IN
9
vu
IN
ve
S
*
*
3
vu
VO
in
IN
a
vu
IN
u»
§
*
UJ
CD
«t
i
o
o
vu
O
IN
O
UJ
r^
Ul
vn
O
vu
vn
m
S
S
§
UJ
fN
vO
Wl
o
u
U">
fn
9
vu
a
o
vu
m
in
9
vu
in
m
S
vu
O
IN
9
vu
IN
U9
vn
o
vu
vn
m
S
S
3
Ue
m
S
3
3
«
OD
w
5
N-2
-------
O
UJ
I
O
u.
O
o
te.
O
I/I
Ul
2
tt
o
i
00
V
8
u
s
u
s
I
}
I
w
«M
"ft
f
1
§
J
1
E
£
I
1
i
|
1
1
f
i
1
f
I
i
(Surrogates
1
z
•
ca
>
03
Z
O
UJ
O
o
UJ
9
UJ
IN
9
o
UJ
IN
9
UJ
m
3
S
m
o
UJ
§
UJ
e
IN
O
3
m
9
UJ
S
IN
e
3
m
(S
9
3
«n
S
m
3
UJ
09
S
3
CO
i
o
UJ
00
IN
O
UJ
9
in
IN
9
UJ
ca
IN
9
UJ
e
UJ
en
3
S
IN
o
UJ
o
9>
3
UJ-
CO
IN
e
UJ
>
•
ca
«j
z
>
IN
U
m
IN
3
-------
3
O
u.
O
O
<
Case 28
S
u
Case 26
IN
3
9
1
§
K
5
4*
I
i
!
|
§
f
s
u.
1
i
111
4 3
i
z
0)
>
IN
O
g
s
UJ
S
IN
9
S
IN
a
IN
00*301
IN
9
3
m
9
UJ
IN
IN
e
UJ
IN
e
a1
m
3
UJ
IN
10
S
OB
i
O
s
IN
00*3*1
§
Ul
e
S
IN
IS
O
Ul
IN
0
UJ
2 8E-02
a
UJ
IN
IN
O
(N
9
IN
3
UJ
IN
10
9
Ul
91
| MVMB
5 6E 02 1
8
Ul
§
Ul
I/I
m
IN
9
i/i
6 2E-02
00*361
2
yi
9
UJ
Ft
IN
9
l/i
S
s
<0
00*361
^
UJ
O
8
*
u*
S
UJ
m
1
IN
0
UJ
IN
UJ
m
1
9
UJ
IN
s
s
3
UJ
r*
10
9
Ul
I/I
m
^
|
0)
s.
i
i
f
i
««
N-4
-------
TABLE N-3
TANK RIM SEAL CLASSES
DESCRIPTION
External Floating Roof Tank:
Metallic shoe seal
- primary seal only
- with shoe mounted secondary seal
- with rim mounted secondary seal
Liquid mounted resilient seal
- primary seal only
- with weather shield
- with rim mounted secondary seal
Vapor mounted resilient seal
- primary seal only
- with weather shield
with rim mounted secondary seal
Internal Floating Roof Tank:
Liquid mounted resilient seal
- primary seal only
- with rim mounted secondary seal
Vapor mounted resilient seal
- primary seal only
- with rim mounted secondary seal
CLASS
E (E)*
C (D)*
A (B)*
C
B
A
H
G
F
A
A
B
A
*For riveted tank
TABLE N-4
TANK SHELL CONDITIONS
CLASS
A
B
C
DESCRIPTION
Light rust
i
Dense rust
Gunite lined
N-5
-------
Appendix 0
Emission Rate Estimates
Variable Vapor Space Tanks
-------
*
z
<
UJ
oe
O
a.
03
<
Q£
2
i
(/)
O
1
u.
U
UJ
'O.
l*»
, UJ
O
s
<
UJ
U
IA
UJ
U
K
O
I/I
LU
a
o
5
ui
5
2
LU
CO
2
o
2
UJ
Wl
5
«
«
«
UI
•
«
^
«
(N
_
*~
0
en
00
r«»
U9
Ut
V
ro
«N
-
Modeling
Parameters
o
^
o
o
o
o
0
o
o
0
0
0
01
«N
o
UI
hput(1Q6gal/yr)
Ol
^
h-
O
Lft
fN
O
10
O
10
O
V0
o
10
o
VC
s
o
LO
(N
O
(N
O
«
fn
s
s
s
s
w
%
jT
2
w
-
o
r«.
o
in
^
-
-
-
-
-
-
-
tional period (yrs)
(0
3
w
5
oj
w
E
o
0)
0)
(Q
w
•o
01
wn
•o
0»
'oT o
.a S
<« w
c ~
2 -3
_ u
11
111
.£ *• e
e-g 3
3 s «•
c a-5
o >. S
» o. E
p c ui
II
0»
V
1
10
u u «
«
1
«l
a.
-------
UJ
a
J
6-6
I
o
00
10
VJ
8
u
VO
3!
ft
in
3!
u
,
s
10
en
8
<0
vj
(N
3!
3
S
i5
$ -
Raoult's La
Surrogate
o
Ol
00
o
Ol
en
«N
o
Ol
O
en
o
Ol
m
O
Ol
m
^™
o
Ol
***
O
Ol
q
9
00
^
00
z
o
Ol
(N
o
Ol
m
(N
o
Ol
V0
~
O
01
'N
(N
O
Ol
IN
*™
O
Ol
V0
o
Ol
o
Ol
00
en
§
Ol
O
"
8
Ol
o
o
•f
Ol
i
Ol
"*
§
Ol
***
§
Ol
IN
o
Ol
00
| VHVLB
in
Ol
10
u
*
01
(0
u
8
(0
u
rs
«
IQ
U
^
3!
%
u
o
4*
U
en
3!
10
$ »
IRaoult's La
Surrogate
IN
O
Ol
"
0
Ol
O
f*»
O
Ol
0
m
o
Ol
O
m
0
Ol
O
O
Ol
o
m
o
Ol
O
r*»
00
z
(N
0
Ol
"
O
Ol
10
IN
0
•¥
Ol
V0
IN
O
Ol
VO
IN
O
Ol
VO
(N
O
Ol
O
*
Ol
VO
^
CD
Z
IN
O
Ol
in
IN
O
Ol
en
in
o
Ol
en
in
o
Ol
cn
in
0
Ol
en
in
O
Ol
en
in
o
Ol
en
in
00
Z
§
Ol
IN
O
i
Ol
in
0
Ol
in
o
Ol
in
9"~
Ol
in
9"~
Ol
in
o
Ol
in
00
I
o
o
Ol
IN
*"*
O
1
Ol
rr>
in
0
01
rn
in
o
1
Ol
m
in
9~"
Ol
m
in
O
i
Ol
m
in
O
Ol
in
00
i
§
Ol
**
O
o
Ol
§
4-
UJ
§
UJ
^™
8
Ol
~
§
Ol
§
Ol
00
i
in
o
Ol
00
in
o
•
Ol
en
in
o
Ol
en
in
o
1
Ol
cn
^
in
o
Ol
en
~
in
o
Ol
en
in
o
Ol
en
00
^
(N
O
Ol
VO
(N
O
Ol
in
(N
0
UJ
in
3
Ol
in
*~
IN
O
Ol
in
^
3
Ol
in
3
Ol
in
| VHVHB
ro
O
Ol
m
"
O
Ol
(N
«
-------
Appendix P
Emission Rate Estimates
Particles from Storage Piles
-------
o
o
ee
ee
O
Z!
Ul
O
O
i
3?
i/i
2
oe
UU
to
z
Ui
U
| Modeling
91
00
IN
X
s
m
rs
IN
s
IN
IN
IN
e
2
QD
£
•a
in
2
=
IN
-
0
91
-
K,
m
*
«
(N
-
s
I
in
in
in
in
in
in
in
in
•in
in
in
in
in
in
in
in
m
in
in
m
-
in-
m
m
"
m
-
-
IArea of surface of pile
(acres)
o
in
in
in
in
in
in
in
in
in
m
in
in
in
in
in
in
i/i
in
in
in
in
^
in
in
m
s
0
in
~
[Sill content (H)
s
o
in
o
o
o
o
e
o
o
o
o
o
o
o
o
o
o
0
e
o
in
IN
in
0
in
O
O
e
e
% of lime windspeed
[exceeds 12 mph
s
3
§
3
3
3
3
3
3
3
3
3
3
3
3
3
3
8
S
3
8
3.
3
3
3
3
3
3
Days of precipitation
201 inch per year (see
| Figure P-l)
2
o
-
0
o
o
o
o
e
e
0
e
o
e
jr
0
*
o
e
e
e
o
o
0
o
o
a
o
o
JMean windspeed (mph)
o
m
o
-
in
O
in
e
in
O
in
e
in
e
in
o
in
O
10
m
-
in
O
in
b
in
b
in
O
in
b
in
b
in
b
in
b
in
O
m
O
in
b
in
b
in
b
in
0
in
b
in
O
[ Moisture content (H)
»
'
*
"
*
«c
*
o
'
IN
*
'
«r
'
*
*
*
*
*
"
»
-------
Table P-2. Emission Rate Estimates (106 g/yr) - Particles from Storage Piles*
Case
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Wind Erosion**
8.1E-07
2.0E-06
4.0E-06
8.1E-06
3.1E-06
6.2E-06
9.0E-06
1.5E-05
6.9E-06
6.2E-06
5.2E-06
5.0E-06
6.2E-06
6.2E-06
6.2E-06
6.2E-06
6.2E-06
6.2E-06
6.2E-06
6.2E-06
6.2E-06
6.2E-06
6.2E-06
6.2E-06
6.2E-06
6.2E-06
8.8E-07
6.2E-06
2.3E-05
Batch Dump***
1.1E-06
2.8E-06
5.6E-06
1.1E-05
8.7E-06
8.7E-06
8.7E-06
8.7E-06
8.7E-06
8.7E-06 '
8.7E-06
8.7E-06
5.1E-06
8.7E-06
1.2E-05
8.7E-06
2.1E-06
2.3E-07
5.9E-08
8.7E-06
8.7E-06
8.7E-06
8.7E-06
8.7E-06
8.7E-06
8JE-06
4.2E-07
8.7E-06
1.6E-05
Vehicle
Activity****
1.4E-07
3.6E-07
7.1E-07
1.4E-06
1.1E-06
1.1E-06
1.1E-06
1.1E-06
1.2E-06
1.1E-06
9.3E-07
8.6E-07
1.1E-06
1.1E-06
1.1E-06
1.1E-06
1.1E-06
1.1E-06
1.1E-06
6.5E-07
1.1E-06
2.1E-06
1.1E-06
1.1E-06
1.1E-06
1.1E-06
3.1E-07
1.1E-06
1.7E-06
P-2
-------
Table P-2(Cont'd)
'Particle size of 10 microns assumed (emission rate particle multiplier of 0.5 used,
based on pg. 4-7 of Control of Open Fugitive Dust Sources. U.S. EPA, September
1988). Constituent concentration of 1ppm assumed.
**
Emission rate estimates for wind erosion based on Equation 3, p. 11.2.3-5 of
Compilation of Air Pollutant Emission Factors. Vol.I. (U.S. EPA, September 1985).
***Emission rate estimates for batch dump operations were calculated using
Equation l.p. 11.2.3-3 of Compilation of Air Pollutant Emission Factors. Vol. I. (U.S.
EPA, September 1985). Drop height of 21.9 feet and dumping device capacity of
6.375 yd3 assumed.
****Emission rate estimates for vehicle activity were calculated using Equation 1, p.
11.2.1-1 of Compilation of Air Pollutant Emission Factors. Vol. I. (U.S. EPA,
September, 1985) assuming one vehicle in continuous operation for 2,080 hours per
year at speed of 3 mph (this low speed assumed to account for loading/unloading in
immediate vicinity of the waste pile.) Minor adjustments in emission rates should
be implemented if unit-specific vehicle speeds and/or total vehicle miles traveled
per year are higher than these assumptions.
P-3
-------
CM
«?
0_
0»
3
o-
c
o
'S
2
'a
G
ai
w
a.
75
a
-------
Appendix Q
Emission Rate Estimates
Particles from Exposed, Flat, Contaminated Areas
-------
a
O
u
o
x
ui
O
3
u
u
IM
l/t
5
o
i
(SI
in
O
cs
CASE NUMBERS
:
«
in
ee
p^
10
in
*
m
r*
-
e
«»
OB
K.
*
in
«
m
IN
••
Modeling
Parameters
\fi
in
in
in
in
in
in
in
in
in
in
in
in
in
in
in
in
in
in
in
in
in
in
in
in
Area of exposed
area (acres)
0
(N
m
in
m
in
in
in
in
in
in
in
in
in
in
in
in
in
in
in
in
in
s
o
in
IN
| Silt content (H)
s
IN
pv
V
ee
m
px
V
r%
«»
r«
V
r*
«
r»
«
r*
«
p»
»
r»
«
f«.
«
r^
»
^
«
r»
*
r^
«
s
.
Ol
c
•o
•a
fl
w
i
2
|
u
I
i
o
£
o
.a
S 5
« *
til *!
ijf Il
III i.*
III Ig
III M
III il
<3 u a "3 >
III !;
lii n
hi I
Q-1
-------
TABLE Q-2
EMISSION RATE ESTIMATES (106 g/yr) PARTICLES FROM EXPOSED AREAS*
Case
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Estimated Emission Rates**
(106 g/yr)
4.8E-08
1.2E-07
2.4E-07
4.8E-07
2.9E-07
4.3E-07
6.7E-05
1.0E-06
1.7E-06
9.1E-06
1.0E-06
3.6E-07
9.1E-06
4.0E-08
1.8E-07
3.6E-07
5.5E-07
9.1E-07
3.6E-07
3.6E-07
3.6E-07
3.6E-07
3.4E-08
3.6E-07
1 .46-04
* Particle size of 10 microns assumed (emission rate particle multiplier of 0.5
used, based on p. 6-9 of Control of Open Fugitive Dust Sources. U.S. EPA,
September 1988). Constituent concentration of 1 ppm assumed.
** Emission rate estimates for particles from exposed areas were calculated
using Equation 8, p. 4-2 of Fugitive Emissions from Integrated Iron and
Steef Plants (U.S. EPA, March 1978).
Q-2
-------
TABLE Q-3
SOIL ERODIBILITY FOR VARIOUS SOIL TEXTURAL CLASSES*
Predominant Soil
Textural Class
Sand
Loamy sand
Sandy loam
Clay
Silty clay
Loam
Sandy clay loam
Sandy clay
Silt loam
Clay loam
Silty clay loam
Silt
Erodibiiity,
tons/acre/year
220
134
86
86
86
56
56
56
47
47
38
38
* U.S. Department of Agriculture July 1964. Guide for
Wind Erosion Control on Cropland in the Great Plains
States, Soil Conservation Service.
Q-3
-------
o
°3»
'5
Q
IB
2
v/»
I
x
•o
c
o
a
m
a
2
3
Q-4
-------
Appendix R
Dispersion Estimates
-------
K
LU
UJ
X
1/1
ec
O
z
o
^
1
UI
O
Z
<
ce
UJ
U
O
u
^
0
0
u
0
u
o
0
vj
"5
u
00
3
u
^
0
u
to
3
^j
U1
3
u
3
u
i
,"Sjj
3
u
3
u
«
c
o
^
3
o-
UJ
C
O
1
CO
I1
~
O — '
r "
s.i
^Jj Q
c x
~ 2
c
41
^
g
*•
41
1
|
U
J
w
3
!
1
w
.£!
^
1
|
-2
M*
"•-
"5
o
ce.
I
Q
U
$
_2
I
•
1
1
91
2
i
2
Q.
O
a
a
fl
_ai
w
*Q
1
I
C
c
1
c
O
| Instruct)
t
J
T5
3
|
CD
C
O
1
1
1
V
«
2
3
^ 4;
1 o
c -
0 =
u >^
$ $
^t
I *
X x
II II
00
: wZ
: o>
oa
1. 1
x o>
i i
552
oo 3
551
i i
x 55
1 1
is!
00 _
5 2
5 OX
2 1
xox
a>
D CO ~
• ^ - J!
i 5x52.3
1> 0
$ £ *^ ^*
c ••! O 4f
J i O ^
>! "o 3
c § i
S '5 w *
a 5^s£
;
J^
I &1
5 2 OE £
Downwind
Sector
Z
UJ
Z
UJ
Z
UJ
Z
i
1
UJ
UJ
UJ
UJ
$
1
,
1
5
1
z
i
J
1
3
2
s
u
4f §
U §
= 5
O *Q
i- U
~ 41
.I]
41 iT
Health Cm
Based on Rl
3
U
O
U
I
g 3
O "3
W *•
£ S
l§ I
" i-§
»:
C5 Q
R-1
-------
Figure R-1. Atmospheric Dispersion Factors for Typical U.S. Meteorological
Conditions (Neutral Stability and 10-MPH Wind Speed)
R-2
-------
Appendix S
Emission Rate Estimation
Worksheets
-------
O
2
O
5
uj V/l
2 *
eo ce
O
P
I
P
ui
<
ec
O
I
^
0
u
o
"5
u
01
"5
u
oo
"o
u
!*•
O
u
vO
"5
u
in
"5
U
"5
u
3
(N
3
"5
u
w
_c
03
X
X
I
03
. X
*l/» ^
o " 2
fl
u. m
C X
w
CD
U .« §
c — X
= ^ oa
tx — 2
C fl T
— o> J
o 2
w
^ 03
8 1
e -1
w
^ GQ
Qi
i
X
00
X
s
V '
> u
.. — x ay
oo 2 ^5 c
c c c ^
o a» * a,
•5 £ g- -o -
W w U. C 1l
| c? £ r_3
^ 7 J w
3Sh2 s
C-- w
|l!|
< 5
s
o»5
II
o.
t
'
'
j
!•
w
(0
«
(N
-------
Q
Z
1
UJ
O
ee
Z
O
I/I
ui
1/1
t/i
_
_
o
u
o
•g
u
0
u
00
"o
u
r^
o
u
*
0
u
U*l
0
u
„
"o
u
"o
u
IN
0
u
"5
VJ
_C
CD
I
X
X
CO
X
tf i
o • 2
u. GO
i i
"5
.. ^ m
u£ 5
11 i
w 0.
2 *i 0°
i^ *•' ^2
C ** T
o 2
w
w
^ CO
oi S
£ X
E -•
5
^ CO
X
X
CO
i
01 '
> 0
•• '5 * S
03 2-g. £
c c c —
•2 2 &3
*•* ^
ao
m
UJ
D
5
rf
>
U
u.
U
o
00
m
UJ
a.
UJ
h»
a
O
00
OO
0
ae
3
in
rsi
^
(N
m
O
m
! I 1
«w _ o ai O
fQ JJ ^^ .^ ^B
T3 « .2 — • « . r-
c 25* * ^
Q " 2 » C ^ *
z f " c o .2 « *
n1" * *c S c if
^ * 2 c 8 1 9 *
^» S Ql «*T 3% *^ ^ ^*
z-| 3 !£ s. 5*
C *^^ U. """^ L/% W «^
1 il 1i U 45
rf * ^ ^ C^ ,ir *^ j«*
^ .ii ^ •— ® o 2 *
| | § 3- S® If
u '? w '? 5. m ^3
w w ^^ A c
U> U9 f*. 00
S-2
-------
o
z
UJ
a
O
o
^
£
5
ee
O
O
l/t
Ul
o
^
(N
"o
U
^™
"o
u
o
*"•
"Q
u
_
w*
"5
u
00
•o
u
r>-
"5
U
V0
"5
U
Lfl
"o
u
^
"o
u
M
"o
u
(N
0
u
"5
u
'•=
00
X
00
z
X
•
O v
*•* ^
W ^
U_ — •
£
^= 00
ro -J
W ^i
U u 2
c rr
• - v S?
w °- *
1? ^
m
— o> -.
o ™
^ ^L
5 >
01
1 2
* X
o
00
^
I
00
X
>
X
fly '
> UJ
. . ** X (U
03 2 =5 C
c c c '-r
O * 9» OJ
.• ^ O «•*
*^ ^ i* ^^ ^"^
a a< ^§ $
£ * 1 ^-o
^ »W ^ jjy ^
— * S ju
^ O H- 3*
. ,
• i
,, <9 Q.
E -oo.
Depth of Oil
Film*
Retention Time*
^ m *
Constituent
Concentration*
Q
?
3
^1
2
LU
D
«j
^
U
^
U
UJ
Du
*?
LU
^
a
o
QC
ec
LO
m
on
O
S
^^
^^
LU
O ^
— 0
—
^J
A
on
cl
00
on
0
&i
on
d
S
on
o
-------
o
I
5
< UJ
-SB
cc
O
O
ui
<—
"o
u
,_
"5
u
_
o
u
00
0
u
1X1
"o
u
0
u
in
0
u
"o
u
"o
u
- Q. c Oi
ff^S
_
E * o 2 •"
"t * w
i 3! S j;
Of (Q
1£
ff
,
1
i
.
,
i
;
'
S
(0
*(0
2
*~
i
•
•
,
,
;
;
*t co
*~ in
& E
c 10 a
c -o a.
«
I
1 i
2 3
* '5 t;
-C c ~
& « S
^
co
i/o
L4J
o^
in
un
00
^
>
U
u
oo
0*
UJ
o.
wo
UJ
<
O
ce
or
r«.
S
no
30
O
0%
—
"3? £ ^
i- O >a 2?
* s s- S 2-
c ^ "" ae oT ^
X <0 ° ii - — Ifc
Q " c «
~ I*? n o ec x
? * *- 2 £ -5 x
Is 5 la i ^x
^^ 4l J*J (Q (Q y W f^
ZTT «w w ii w ^ —^
^ *i w *^ ii Q^ .tr ^»
1 w^ 2x it S.<^
u S 2 c § ^ •£ 5"%
^ (0 O w g, * ± *
-1 fi 15 s*
eg. eg. — S ^^.
§ 'S § % \* -5 1
* ^ * *
S-4
-------
ae
O
si
i
IS)
I
•-
"o
u
o
•5
u
"o
u
00
o
u
r*.
"o
u
"o
u
un
"o
U
*T
"o
u
m
"5
U
o
'•* x *
i 2^6.5
c c c ~
•x S Q- -o ^
3 a < 3 jg
- ?|5|
"5 i fc i«k \— * ^.
§ 13 01 w "•' ,. 01 £
^.(0 5W O> S *•
t ^ w ^ 2 O ^ C
1| |'| |g |l
•j *fl •• *^ fg ^ ^B *^
ll ll IS 3l
• - - -
a;
cc
5
a.
•
rs
6
0)
c
O
£
LU
a
X
C
Ol
a
a
E
o
•3
rst
O)
.£ 00
.
•f> ^~ ^
* Srs,
_ 0) CM
.= u.
•s
•c 3
u L« -a
S-5
-------
V9
ec
LU
V/l
Z
O
ui
Si
1!
o
2
O
P
I
O
fN
^~
^r
U
~
^
"5
o
"o
u
G\
u
00
"o
u
r*.
"o
U
VO
"o
u
0
u
'
o
u
"o
u
(N
"o
u
"o
u
*
••=
oa
^
x
00
X
I
•
I 1'
<0 >
u. — '
Ol
C
.= m
flS ^^
U*> P
v^ ^
c r=
o y oo
'5 a 5
3 ^^ ^^
^ Q) ^5
tx •- *
c ^
<- X
5 >
^ 2
If
1 1
5 r
Q
03
2
X
03
X
^
X
ay '
> X
.. '•= x w
oo 2 ••$ .E
c = c •=
3 Q.< 3 £
k- HI _ — <9
S oe E - w
— fl 2 x "2
^ "a") — vJ
55^1
1 1* |
| g 1|
|£
wt
^^
O ^
^2 ^
i
1
,
j
!
(
(
I
•
;
j
^ oo — ^
55oo
! m ^ »— ^
*~ in *" .
i - —
•^ c c
O> 41 91
•* % E ¥
a
a
^
E
W
hi il ii I !
in u if 1 1
< •£ >E o'o u 8 5 K
•- §
o - >«
1" ^ff
"^ II ^^
^^ II ^ '
* 15 1 «•*
^ C ^ *f* «» ^ ^k
O (a > n 5 w _
7 «A — O O 06 X
2 9» 2 3 '~ o5"6
u •- = ^ J '5 "
of S .ii §* « ,| *
if 11 1 5*
1 *« -«2 w X «
u j. a * 5, a|N
II ft II
c c 5C ^ * "a
g - 87 35
8 c a S ^|
«O P*. 00
U
C
•o
II
c
oc
o
a
•a
01
•
•
rs
ac
O
X
x
'•B
a.
a.
o
in
fN
j
O
is?
.si J
"5 21 *
«
« «
S-6
-------
<
3
2
O
m
ui
ee
O
7>
>
3
VO
"5
U
in
"5
VJ
**
"o
u
"o
u
(N
"o
u
"o
u
c
'-J
00
-J
I
-*
00
X
X
•
1 i"
m >
u. — '
01
C
;- 00
ro —1
v ^
U w 2
C *tl
o w oa
••• o» ^
I? 2
c
Ot
c
1 s
*•* T
^ ^"
a
00
^
X
00
X
^
X
00 1 ^ *
1 1 S-1-
^ 01 _ 3 3
V ^ 'w **
" in
S E
w.
Of
1
a
w
CL
^
. S. 1 J
2 a *P "3- <3 1
« a; — a o» o
< 0 iZ < > u
^ IN m v
g
o
Concentrati
in
jj
13
.^
^
>
VJ
u_
O
LU
0.
tn
—
3
O
S S S 2
r< ° e? ,,,-P
* II C S *
> u .0 oe x
>n g c "5 c Jg
i 5 i ^ i ^ *
> ex ^
* * S o ^> ^» *0 "
J;(0 fPw O> «!•«»
^« w« 20 'E*
If 1! If !{
li 31 la 31
in - io r-» oo
Of
c
•a
££
O
a
.2
•o
0)
•5
>
•5
CM
0;
ae
^"
5
a
a
<
o
01
IM W
4) w -^
1*3
.z u. ••*
U m U
«
S-7
-------
\
o
ui
O
U
ui
Ul
5
"I
I
Ui
UI
fN
"o
U
—
"o
u
0
"o
u
CT>
"o
u
CO
"o
u
rs.
"o
u
US
0
u
in
0
u
2
u
m
"o
u
(N
"o
u
"o
u
c
CD
^
X
>
OD
X
X
•
I 1-
<"" ?r
m <
U. — 1
c
:= CD
fo — '
«3 >
vn ^
*-* —
c £
O ^ CD
5 Q. 2
3^^ ^i
t ^
a * 5
i^ ^J
— o» m
o 55
^ JK
3 >
01
E 2
™
S x
Q
CO.
>
X
CO
X
X
0) i
•• 5 "* 9>
a c "1 •-
o * * "5
'5 5 S-O-ST
3 O.4 ^ •*
Sfifif 3
— ^ ^ "^ *
o{ j{ ^ «J
4/^*9^;
^ W H- tft
§i^s
1 ^ !••«
I-'/'>
t/t
O> >-
C **
^ *rf
i 1
?i
si
1
,
*
1
1
•
!
!
;
j
fN *
« °° - X
o o ® 5
I PO ^ H *~
"" in *~ """.
c^ ^
s
V
1
percent
percent
*
w< fO
91 ui _* 91 O '7
•98 .a 3 « 2
• 2 c 5 •= S o
2 S^ S. s| 5-
< $5 5 38 <
~- IN fn v in
T
VO
fN
ffl
—
LU
in in
in o
^*
LU
3
>
u
00
0
LL.
U
LU
a.
i
00
nt
O
kU
h-
a
0
ec
in
^
O
3
o
r».
*
^
CM
fN
^
fN
5= 5«
Of O) i^
2 ^ CD ^^
w> (O ^5 _p ^
^3 • QJ rtT ^O
X C **"* C « **
2 ^ ^ •§ * in
ii 5 s i?
|i 3 i ««
3 ro| ifc i£
w Ja ^ O^ '
< u §** •= S
<2 'O 5 ^ ~
** s ^ ?? " —
s a- Tg s « .2-
0 ± u « 3 S
a H si
. ID r^ oo
u
c
o
•o
(
ir
o
•o
Ol
•o
o
w
4)
i
O
x
^
o
tn
I
2?
.55 |
S.?|
U (X fN
«
« «
S-8
-------
X
«/)
-
o
2
O
CM
"o
u
^*
«••>
o
u
o
"o
u
o\
w<
•5
u
00
"o
Pv
"5
u
\o
"o
u
m
"5
u
v
o
u
r^
"o
u
"o
u
"5
u
.£
CO
>
X
>
ea
X
^P
X
•
*£ ca
r, 5
w ^
LL. —*
Ol
:r 00
nj —I
«3 >
u« *
c E
2 * £
w 0- 1
^ ^^ ^^
1 i 5
^* *"
^ fl
— o» _
2 x
3 >
in 2
41
C
1 3
n» ^^
w T1
^ •••
Q
CO
5
X
03
X
^
X
> ^
**^ x 4>
d S^s
1 1 ll_
3 a< ^£ 3j
t« ae £ «- w
•Si " 2 ^ "S
vj "*• 41 «J
_* S 3 *
»^>
— 5 >. 41 -s 41
?•-, £o Is 2S
•9S =- o> o>
*= 5 g 2-0 5-o
i«-J5 <04»
- x « x
§r* w O •— ;.• O i^
- U w < ^ 1- 2^
^~
in
ul
U
LU
d
lO
m
UJ
h-
^<
a
o
^
_
in
^
^
T
^
00
!2
^
00
(N
LO
LC
^^ w
"$ "B> —
W yfl J"
4t O ^^
•o oe « **
li ; 1 Is
h c « E - x
U j iu <" ^
24. P! .« 1 =*
^ flj 41 ^ *" X
if S I 552
o la *? ._ 4> x
U ^i ™ dl •• ^^ ^4
<0 a» 2 >« ^ ..7
4J w 75 ^J u? *
< w §** '? S
35 §2 §£
^ « 3- * >«
c a _ PJ ^ a
a ~ w 41 35
li M 31
ve P«. ao
•o
a/
av
T3
a/
'S
tr
c
o
a
T3
Oi
_5
~
^3
>—
o
a<
I
•^
<^
LU
«
c
o
•a
4>
a
a
<
o
C
'Jj
o
41
3 -o
•= S IN
U >X IN
S-9
-------
(L
Ul
(/I
Ul
Ul
z
1/1
"j z
«0
< H
I- <
I/I
I/I
I
Ul
rsi
*~
^
u
*™
"o
u
0
"o
u
ON
"o
u
00
"5
u
^
"o
u
vO
0
u
1/1
o
u
2
u
PO
"5
u
(N
"o
u
"o
u
0>
'.J
CO
_J
>
I
>
CO
X
X
«
t*
2 1
u. — '
C
"m — i
o^ s
c rE
w °- P
T ^^ ^^
1 i 2
«/» ti
C" 'w
2 g
I >
1/1 2
41
C
'i g
5 >
S X
CD
2
X
00
X
X .
V >
'— x "^
^"
acres
per year
C
!
percent
percent
«
«
ts
~ i >."* -fT*1
- £ *o • 9 m S $
> .i c 25 | 5
1 If 11 I* Is5
*~ °* m * ^
(/>
=5
j
>
U
^
LU
O.
—
rsi
00
no
ff!
(N
O
tft
tri
(N
ci
rvi
t—
S
0
p
CN
^
^
0
j^
fN
CT>
,g.
oo
(N
^
^
fS
^^ w
>*" * — .
^
w ^ ^^ ^»
xi 5 | 5 is
z£ " I |J?
2 S 8 | § *
C -- s ,1 'S *
*2 « £ I*
L_ rtl <^ .•• **• ^
K -s 2 S! w «*»
1 §• «3 S. 5 *
o "^ i k. S IN
* 5 Jii !£ *
,5 * fc * ^ =
1 9- -5 S j2 !9-
8 •? '5. a ^ "=
-------
O
o
oe
a
UJ
X
UJ
UJ
Z
«/1
< Z
I- O
<
1/1
UJ
UJ
ee
IM
"5
U
»•»
^
"o
u
0
"o
u
CT>
"o
u
00
0
^
—
u
1C
5
Ift
"o
u
9
"Q
u
"o
u
fS
*~
"o
Ol
.3
•
c
w
l«5
u.
c
-Z
J
.. *"
VJ .^
c ^
O ••
V c
«• ^
— i
00
_j
I
03
X
X
00
S
-J
1
00
^
5
00
^
2
~ S> ffl
w
^
V?
Ol
c
1
a
a)
o
*
X
1
00
X
00
2
X
00
X
^
X
§5
InstuctionB:
Select a Representati
Case from Appendix
Table M-2 (underline
selected case)
<
§
«*
w
3
tyt
_C
1
C !S S
-^1
2 ^^ ^*
O-vrt
CTi
.E *<
— 01
"§1
* cw
*" ^* ^ w* */*
O w> O ^ ^
_i O -J O O
C O^ C O^ C3^
2 .£ 5 .£ .£
<-• -i ~ jc li
Oi O Oi O O
w > w ^ ^
CO > CD > >
^
^~
<£ ~ O
o 1 i£
O Wl
t
Ol |£
S 2
*~ -
c *
com
.2 '"» o^
•^ .— e
•- i o
o E u] •§
^5 a § 1
Z. f Q _J LU
o £ -• o> a
g'-= c | 3
CC * Q 2 C
i_. ^-•••» W w -^
'^ .^1 ^^ W ^rf
ee.
o
'7
«•
I* ^
i ?
* .£ °
>*•£-«. a»
"^IfclfcwjISl
5* O^ »^k O^ *C X *G M *" S^
i ^B i ^9 Ql f^t ttl ,^_ C i^k
** ^ 1S.I ** ^^ ^^ ^ ^^ ^^ ^
Sj* &* *i*^ *1*^ ^5
2S 2S || |.c o%
.5 * .2 S> 0|>, 01 >. 414;
'5.S '.5.5J J3 3| J^
i— oc t— ' oc vj -£- u -3> vj -2.
o
0)
^
u
c
•o
c
"oi
•a
a;
Q£
c
o
^
^
E
LU
.
h-
.a"
•a
a>
•
fNI
_«»
^
r3
I—
5
a;
c
rf»
ee
c
o
'3
2
x
•6
I
a
o
i
01
«v» w
n, a,
— V
5-0
*^ o
3 2
•t: m *
w ., C
u tX =
S-11
-------
t/l
o
o
O
z
o
2
O
p
I
u
O
£
"o
u
~
~0
u
o
"o
u
"o
00
Q
u
^
"5
u
IA
o
in
"5
U
V
"5
u
m
"o
U
(N
"o
"o
VJ
0)
_c
'.J
.
VI
2
^
u.
a
c
"rt
SJ
VJ i
C ~
as
™j ^H
*" dJ
W* **
C *
— a
0
w.
"C
c
'§
£
^"M
0
1
InstuctionB:
<
struction
£
CO
_1
X
03
I
X
•*
CO
S
^
i J
00
_J
2
1
1
00
X
i
00
^
^
X
00
^
X
00
X
X
(
> 2
avi«(.i 0 nepr civilian
Case from Appendfx
Table N-l (underline
selected case)
t
^^
£ ^
c ^
=5 *
^g C
o 2
5 a
„
o
^J
E
K
00
hi.
w\
•
"5
01
P
g
^
1
-
VI
O
_i
1
w
•o
_^;
$
O
O
en
1
€ ••*
S ^
•2 £
IN
VI
O
_i
i
w
•o
•*
§
VO
o
01
*
T3
I
01
o
_J
1
^
^"
«^
i
in
VI
VI
O
O>
C
'2
uZ
VO
!
,
1
;
i
i
[
'
V
1
[
(
;
j
;
If
a
3
O
K
^
u^
UJ
— 1
>
U
u.
U
LU
a.
i
UJ
1
ac
1/1
RUCTION 0:
Lin^^ftanH 17- IS
1— 01
w* 5
2 J
£
V.
i
i
«
3
i Q.
j—
J
|
g
O
^
a
: ! ;
• * i
• t ,
i s i
0 0 §
o ° S
** ^*
"5 ^ O "S
(Q >il *"* i
U "be 5 ut ^
Q. 'G 'w '0 w
||||*|J|
f 10*^* W 10^ —
.t; -^ * -s^ 5! -s^ * 3
J. i^!i *Q, K"!i >3
8 °* 2 - -
^.
^
S
*~
n"
•^
f8
Irawal Loss Emission R
S? 5
l\
11
w T?
"5 '
S3
2
^
"S^
^M
O
«^
IQ
ee.
o
x 2^
tr» .5
•I
11
•— —
"5 3
^.u
2
w
Of
«J
*- .2
J S
w ^
"5 3
.§3
-
^
—
*
*
5
^
•Q
3
.
2
•^
a
15
S-12
-------
IU
s!
on
iu
a
S
<
•
K
Ul
O
I
1/1
O
W1
IN
"o
u
"-
"o
u
o
0
u
o>
"o
u
00
"o
u
f^
"o
u
VO
3
in
"o
u
o
u
rn
u
M
3
3
91
c
oo
^r
X
>
CO
X
^
X
«
•
1 1
u. J
c
— OD
yj ^
^ ^
2
^ w
c E
O w Q3
w a- 2
«• v^ ^^
Si 2
c ^
~ o1 2
I >
^ ^—
c
E <=
I >
gj —
o
CO
2
X
CO
X
X
01 '
.1 o
** X ^
oo 2 -5 —
c = c ~
O 41 01 4>
W w ^fc • Q^
3 ^L ^f 3 ^
W At ^^ f^
- 2 J ° *
J! S 3 *
01 *9 (Q MI
wn U t— S
< i
ff -^ w
llll
- 2 a5->
l£
o» £
II
22
9 n
o 5
on f"*
IN" vo"
«
" in
"!5
2' £
X »|
1. 2
1 '!
•§• js
S 1l
^ 10 C
5 ^2
- «.
o
CN
nn
o
un
••
LO
0
LU
^
LU
D
^ _
> -
i.
^
u m
S in
•0 o
LU
<
o £
K °
2
^
in
-
(N
O
l!
I
*r
^t § I
s I -s 1
9 ^ •- S
P * E a -.
y * w .g w
-J T- w C ^
8C ^ IS X
II I^ E*
ll* 8-5
*0 *"* i ^|
O) ^^ ,*•* A^
5 a» 3 3
J? S * >•
»* (0 •* •—
•= u * Q-
S * 3 '5
-Q.2 «"3
.
m *
I
01
^3
ce
o
•o
0)
T3
5
•5
.2
(0
>—
O
^
o>
oe
o
O
X
1
V
I
•S «
•= 3 S
«
« «
S-13
-------
TABLES-14
EMISSION RATE ESTIMATION WORKSHEET- PARTICLES FROM STORAGE PILES
Col 1
Col 2
Col 3
Col 4
Line
Modeling Parameters
Instruction A:
Input Unit -
Specific
Values
Instruction B:
Select a Representative Case
from Appendix P - Table P-1
(underline selected case)
Instruction C
Determine
Scaling Factor
1 Area
2 Silt content*
3 Silt content*
4 Silt content*
5 % of ti me wi nd speed
exceeds 12 mph*
6 Days precipitation
(> .01 inch/day)
7 Mean wind speed*
8 Moisture content*
9 Vehicle weight*
10 Vehicle wheels*
11 Throughput*
12 Mass fraction of
contaminant
acres
1,2, 3 or 4
5, 6, 7 or 8
wind erosion
batch dump
vehicle activity
wind erosion
days
mph
tons
tons/yr
ppm
9, 10, 11 or 12 wind erosion
vehicle activity
13, 14or15 batchdump
16, 17, 18oM9 batchdump
20, 21 or 22 vehicle activity
INSTRUCTION D: Complete Lines 13-15 and 19-22
13 Account for Area
[unit-specific area/(Case 28 area = 5 acres)]
14 Account for Vehicle Wheels
[square root (vehicle wheels)/square root (Case 28 wheels = V4)]
15 Account for Throughput
[unit throughput/(Case 28 throughput = 50,000 tons/yr)]
16 Typical case emission rate • wind erosion
(Case 28), 106 g/yr
17 Typical case emission rate - batch dump
(Case 28), 106 g/yr
18 Typical case emission rate - vehicle activity
(Case 28), 10* g/yr
19 Calculate Unit-Specific Emission Rate-Wind Erosion, 106 g/yr
(multiply lines #2 x #5 x #6 x #13 x # 16)
20 Calculate Unit-Specific Emission Rate - Batch Dump, 106 g/yr
(multiply lines #3 x #7 x #8 x # 15 x # 17)
21 Calculate Unit-Specific Emission Rate - Vehicle Activity, 106 g/yr
(multiply lines #4 x #7 x #9 x #14 x #18)
22 Calculate Total Emission Rate, 106 g/yr
(add lines #19 f #20 •*• #21)
SURROGATE-SPECIFIC VALUES
6.2 x 10'6
8.7x10-6
1.1 x 10'6
Critical input value
Scaling factor determined for Lines 2-12 from Appendix P
m Case 28 (see lines 16.17, and 18).
Emission Rate Estimate from Table P-2 divided by Typical Emission Rate defm
S-14
-------
<
a
LU
a
LU
2
X
LU
O
u
I/I
o
o
<
o
I
^>
O
u
en
"5
U
(N
0
u
"5
u
«
c
g»
^J
c
o
vj
5
=
u.
TS
W •
w
.£ u ' ' '
I^L
5
4»
o
„ •: _
" S
^ Jjj (^
03
O
instruct
•Mi , - !
2 .1. u v. o o
c Qj; o a ^
9» « 4* : ^ - - : :
S | S rJ • "X *
S C )£ LQ «—
u Q^ C ^
3f Q- "5 LH Lft"
- — 0. o> -
u < -0
4> e c
— C 3
w rt ^^
11
._ >-
o
Instructi
'0 4^
J-S 2
'£ "* S "3« p
„ w .p § >a Q.
3 (fl 0^ — 0^ &
a
c
v -o
Tf, « 4)
* V 01
S * Q.
§ :=••» — >N
* 3 1 °. "55
1 i i ii IS i
s 4>m";ivC
1 % ! s IMl i
1 1 = 1 5= |i s
^ ^ »/^ dp /\ AW *•*
«- rn f*> ^ i/> ^O
v/l
LU
r>
_
>
u
O
LU
a.
LU
H-
a
o
X
ae.
D
o _
^ »A ^
"i 25
• • S u C
O
1
u
ae
i/>
^m
** (0 •=
* un S
s » s
S a O
(O O —
(N (N K
? 5 s
C Af ^
o — c
•v (Tl LLJ
?
x
eo
x
X
L0
X
2
(unit-specific mass fract
Typical Case Emission R.
Calculate Unit Specific 1
(multiply line #2 x #3 x
0* »
CM
a
ai
J3
t—
E
o
a
E
LU
4j
o
X
0
ll
¥
**• •
2 a;
Is
c.
O- 4*
< 3
E c
o E
i -a
21
fl
3 i
o£
itical input values
aling Factor determined
/ided by Typical Emission
U vX ^
« «
S-15
-------
APPENDIX H
SOIL LOSS CALCULATION
EXCERPTED FROM
U.S. EPA. Final Draft Superfund Exposure Assessment
Manual. September, 1987. Office of Emergency and
Remedial Response, Washington, D.C. 20460
H-1
-------
APPENDIX H
SOIL LOSS CALCULATION
•
Introduction
Many of the organic substances of concern found at Superfund sites are
relatively nonpolar, hydrophobic substances (Delos et al., 1984). Such substances
can be expected to sorb to site soils and migrate from the site more slowly than will
polar compounds. As discussed in Haith (1980) and Mills et al. (1982), estimates of
the amount of hydrophobic compounds released in site runoff can be calculated
using the Modified Universal Soil Loss Equation (MUSLE) and sorption partition
coefficients derived from the compound's octanol-water partition coefficient. The
MUSLE allows estimation of the amount of surface soil eroded in a storm event of
given intensity, while sorption coefficients allow the projection of the amounts of
contaminant carried along with the soil, and the amount carried in dissolved form.
Soil Loss Calculation
Equation 2-20 is the basic equation for estimating soil loss. Equations 2-21
through 2-24 are used to calculate certain input parameters required to apply
Equation 2-20. The modified universal soil loss equation (Williams 1975), as
presented in Mills et al. (1982), is:
Y(S)E = a(Vrqp)0.56KLSCP (2-20)
where
Y(S)E a sediment yield (tons per event, metric tons per event).
a a conversion constant, (95 English, 11.8 metric).*
Vr = volume of runoff, (acre-feet, m3).
qp = peak flow rate, (cubic feet per second, m3/sec).
Metric conversions presented in the following runoff contamination equations
are from Mills etal. (1982).
H-2
-------
K = the soil credibility factor, (commonly expressed in tons per
acre per dimensionless rainfall erodibility unit). K can be
obtained from the local Soil Conservation Service office.
L «. the slope-length factor, (dimensionless ratio).
S = the slope-steepness factor, (dimensionless ratio).
C = the cover factor, (dimensionless ratio: 1.0 for bare soil); see
the following discussion for vegetated site "C" values).
P a the erosion control practice factor, (dimensionless ratio: 1.0
for uncontrolled hazardous waste sites).
Soil erodibility factors are indicators of the erosion potential of given soils
types. As such, they are highly site-specific. K values for sites under study can be
obtained from the local Soil Conservation Service office. The slope length factor, L,
and the slope steepness factor, S, are generally entered into the MUSLE as a
combined factor, LS, which is obtained from Figures 2-4 through 2-6. The cover
management factor, C, is determined by the amount and type of vegetative cover
present at the site. Its value is " 1" (one) for bare soils. Consult Tables 2-4 through 2-
5 to obtain C values for sites with vegetative covers. The factor, P, refers to any
erosion control practices used on-site. Because these generally describe the type of
agricultural plowing or planting practices, and because it is unlikely that any
erosion control would be practiced at an abandoned hazardous waste site, use a
worst-case (conservative) P value of 1 (one) for uncontrolled sites.
Storm runoff volume, Vr, is calculated as follows (Mills et al. 1982):
Vr » aAQr (2-21)
where
a » conversion constant, (0.083 English, 100 metric).
A a contaminated area, (acres, ha).
Qr ~ depth of runoff, (in, cm).
Depth of runoff, Qr, is determined by (Mockus 1972):
Qr = (Rt - 0.2Sw)*/(Rt + 0.85*,) (2-22)
H-3
-------
Slope Length, Meters
20 30 40 60 80 100 150200 300 4OO 600 800
40.0 •
70 100 200 4OO 600 IOOO 2000
Slept Length, Fett
Figure 2-4. Slope Effect chart Applicable to Areas A-1 in Washington, Oregon,
and Idaho, and All of A-3: See Figure 3-5 (USDA 1974 as Presented
in Mills etal. 1982).
NOTE: Dashed lines are extension of LS formulae beyond values tested in
studies.
H-4
-------
Figure 2-5. Soil Moisture-Soil Temperature Regimes of the Western United
States (USDA 1974).
H-5
-------
20
3.S 6.0 10
Slope Length, Meters
20 40 60 100
200 4OO 60O
10 20 40 6O IOO 200 400 600 1000 2000
Slope Length, Feet'
Figure 2-6. Slope Effect Chart for Areas Where Figure 3-5 Is Not Applicable
(USDA 1974).
NOTE: The dashed lines represent estimates for slope dimensions beyond the
range of lengths and steepnesses for which data are available.
H-6
-------
TABLE 2-4
C" VALUES FOR PERMANENT PASTURE,
RANGELAND, AND (OLE LAND
Vegetal canopy
Type and height
of raised canopy*"
No appreciable canopy
Canopy of tall weeds or
short brush
(0.5m fall height)
Appreciable brush or
brushes
(2 m fall height)
Trees but no appreciable
low brush
(4m fall height)
Canopy
covert
(%)
25
50
75
25
50
75
25
50
75
Cover that contacts the surface/Percent groundcover
Typed
G
W
G
W
G
W
G
W
G
W
G
W
G
W
G
W
G
W
G
W
0
0.45
0.45
0.36
0.36
0.26
0.26
0.17
0.17
0.40
0.40
0.34
0.34
0.28
0.28
0.41
0.42
0.39
0.39
0.36
0.36
20
0.20
0.24
017
0.20
0.13
0.16
0.10
0.12
0.18
0.22
0.16
0.19
0.14
0.17
0.19
0.23
0.18
0.21
0.17
0.20
40
0.10
0.15
0.09
0.13
0.07
0.11
0.06
0.09
0.09
0.14
0.085
0.13
0.08
0.12
0.10
0.14
0.09
0.14
0.09
0.13
60
0.042
0.090
0.038
0.082
0.035
0.075
0.031
0.067
0.040
0.085
0.038
0.081
0.036
0.077
0.041
0.087
0.040
0.085
0.039
0.083
80
0.013
0.043
0.012
0.041
0.012
0.039
0.011
0.038
0.013
0.042
0.012
0.041
0.012
0.040
O.U13
0.042
0.013
0.042
0.012
0.041
95-100
0.003
0.011
0.003
0.011
0.003
0.011
0.003
0.011
0.003
0.011
0.003
0.011
0.003
0.011
0.003
0.011
0.003
0.011
0.003
0.011
Source: Wischemier 1972.
a All values shown assume: (1) random distribution of mulch or vegetation and (2) mulch of appreciable depth
where it exists.
b Average fall height of waterdrops from canopy to soil surface: m = meters.
c Portion of total-area surface that would be hidden from view by canopy in a vertical projection (a bird's-eye
view).
<* G: Cover at surface is grass, grass! ike plants, decaying compacted duff, or litter at least 5 cm (2 in.) deep.
W: Cover at surfact is mostly broadleaf herbaceous plants (as weeds) with little laterial-root network near the
surface and/orundecayed residue.
H-7
-------
TABLE 2-5
'C VALUES FOR WOODLAND
Stand condition
Well stocked
Medium stocked
Poorly stocked
Tree canopy
percent of area'
100-75
70-40
35-20
Forest litter
percent of area*>
100-90
85-75
70-40
Undergrowth^
Managed0
Unmanagedd
Managed
Unmanaged
Managed
Unmanaged
"C" factor
0.001
0.003-0.011
0.002-0.004
0.01-0.04
0.003-0.009
0.02-0.09*
Source: Wischemier 1972.
a When tree canopy is less than 20 percent, the area will be considered as grass land or cropland
for estimating soil loss.
b Forest litter is assumed to be at least 2 in. deep over the percent ground surface area covered.
c Undergrowth is defined as shrubs, weeds, grasses, vines, etc., on the surface area not
protected by forest litter. Usually found under canopy openings.
<* Managed - grazing and fires are controlled.
Unmanaged - stands that are overgrazed or subjected to repeated burning.
« For unmanaged woodland with litter cover of less than 75 percent, C values should be derived
by taking 0.7 of the appropriate values in Table 3-4. The factor of 0.7 adjusts for much higher
soil organic matter on permanent woodland.
H-8
-------
where
Rt s the total storm rainfall, (in, cm).
Sw - water retention factor, (in, cm).
The value of Sw, the water retention factor, is obtained as follows (Mockus
1972):
-10 a (2-23)
where
Sw » water retention factor, (in, cm).
CN a the SCS Runoff Curve Number, (dimensionless, see Table 2-6).
a a conversion constant (1.0 English, 2.54 metric).
The CN factor is determined by the type of soil at the site, its condition, and
other parameters that establish a value indicative of the tendency of the soil to
absorb and hold precipitation or to allow precipitation to run off the surface. The
analyst can obtain CN values of uncontrolled hazardous waste sites from Table 2-6.
The peak runoff rate, qp, is determined as follows (Haith 1980):
aARtQr (2.24)
Tr(Rt - 0.2Sw)
where
qp » the peak runoff rate, (ft3/sec, m3/sec).
a a conversion constant, (1.01 English, 0.028 metric).
A a contaminated area, (acres, ha).
Rt a the total storm rainfall, (in, cm).
Or a the depth of runoff from the watershed area, (in, cm).
Tr a storm duration, (hr).
H-9
-------
TABLE 2-6
RUNOFF CURVE NUMBERS
Soil Group
A
B
C
D
Description
Lowest runoff potential: Includes deep
sands with very little silt and clay, also
deep, rapidly permeable loess
(infiltration rate = 8-12 mm/h).
Moderately low runoff potential : Mostly
sandy soils less deep than A, and loess less
deep or less aggregated than A, but the
group as a whole has above-average
infiltration after thorough wetting
(infiltration rate = 4-8 mm/h).
Moderately high runoff potential:
Comprises shallow soils and soils
containing considerable clay and colloids,
though less than those of group 0. The
group has below-average infiltration
after presaturation (infiltration rate = 1-
4 mm/h).
Highest runoff potential: Includes mostly
clays of high swelling percent, but the
group also includes some shallow soils
with nearly impermeable subhorizons
near the surface (infiltration rate = 0-1
mm/h).
Site Type
Overall
site*
59
74
82
86
Road/right
of way
74
84
90
92
Meadow
30
58
71
78
Woods
45
66
77
83
Source: Adapted from Schwab et al. 1966.
* Values taken from farmstead category, which is a composite including buildings, farmyard,
road, etc.
H-10
-------
Sw = water retention factor, (in, cm).
Dissolved/Sorbed Contaminant Release
As discussed in Mills et al. (1985), the analyst can predict the degree of
soil/water partitioning expected for given compounds once the storm event soil loss
has been calculated with the following equations. First, the amounts of absorbed
and dissolved substances are determined, using the equations presented below as
adapted from Haith (1980):
Ss = [1/0 + ec/Kd8)](Cj)(A) (2-25)
and
Ds = [1/0 + Kd8/ec)](Ci)(A) (2-26)
where
S« s sorbed substance quantity, (kg, Ib).
8C = available water capacity of the top cm of soil (difference between
wilting point and field capacity), (dimensionless).
K
-------
can be estimated according to procedures described in Lyman etal. (1982). Initially,
the octanol-water partition coefficient can be estimated based on the substance's
molecular structure. If necessary, this value can be used, in turn, to estimate either
solubility in water or bioconcentration factor.
After calculating the amount of sorbed and dissolved contaminant, the total
loading to the receiving waterbody is calculated as follows (adapted from Haith
1980):
PXj =: [Y(S)E/100B]S$ (2-27)
and
PQi = [Qr/RtJ Ds (2-28)
where
PXJ = sorbed substance loss per event, (kg, Ib).
Y($)E > sediment yield, (tons per event, metric tons).
6 3 soil bulk density, (g/cm3).
Ss = sorbed substance quantity, (kg, Ib).
PQi « dissolved substance loss per event, (kg, Ib).
Qr * total storm runoff depth, (in, cm).
Rt = total storm rainfall, (in, cm).
Ds ~ dissolved substance quantity, (kg, Ib).
PXJ and PQi can be converted to mass per volume terms for use in estimating
contaminant concentration in the receiving waterbody by dividing by the site
storm runoff volume (Vr, see Equation 2-21).
H-12
-------
REFERENCES
Delos C. G., Richardson, W.L, DePinto J. V., et al. 1984. Technical guidance manual
for performing wasteload allocations, book II: streams and rivers. U.S.
Environmental Protection Agency. Office of Water Regulations and Standards.
.Water Quality Analysis Branch. Washington, D.C. (Draft Final.)
Haith D. A., 1980. A mathematical model for estimating pesticide losses in runoff.
Journal of Environmental Quality. 9(3):428-433.
Lyman, W. J., Reehl W. F., Rosenblatt D. H., 1982. Handbook of chemical property
estimation methods. New York. McGraw-Hill.
Mills W. B., Dean J. D., Porcella D. B., et al. 1982. Water quality assessment: a
screening procedure for toxic and conventional pollutants: parts 1, 2, and 3.
Athens, GA: U.S. Environmental Protection Agency. Environmental Research
Laboratory. Office of Research and Development. EPA/600/6-85/002 a, b, c.
Schwab G. 0., Frevert R. K., Edminster T. W., Barnes K. K., 1966. Soil and water
conservation engineering. 2ndedn. New York: John Wiley and Sons.
USDA. 1974. Department of Agriculture. Universal soil loss equation. Agronomy
technical note no. 32. Portland, Oregon. U.S. Soil Conservation Service. West
Technical Service Center.
Williams J. R., 1975. Sediment-yield prediction with the universal equation using
runoff energy factor. In present and prospective technology for predicting
sediment yields and sources. U.S. Department of Agriculture. ARS-S-40.
Wischmeier W. H., 1972. Estimating the cover and management factor on
undisturbed areas. U.S. Department of Agriculture. Oxford, MS: Proceedings of
the USDA Sediment Yield Workshop.
H-13
-------
OSWER DIRECTIVE 9502.00-60
INTERIM FINAL
RCRA FACILITY INVESTIGATION (RFI) GUIDANCE
VOLUME IV OF IV
CASE STUDY EXAMPLES
EPA 530/SW-89-031
MAY 1989
WASTE MANAGEMENT DIVISION
OFFICE OF SOLID WASTE
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
ABSTRACT
On November 8, 1984, Congress enacted the Hazardous and Solid Waste
Amendments (HSWA) to RCRA. Among the most significant provisions of HSWA are
§3004(u), which requires corrective action for releases of hazardous waste or
constituents from solid waste management units at hazardous waste treatment,
storage and disposal facilities seeking final RCRA permits; and §3004(v), which
compels corrective action for releases that have migrated beyond the facility
property boundary. EPA will be promulgating rules to implement the corrective
action provisions of HSWA, including requirements for release investigations and
corrective measures.
This document, which is presented in four volumes, provides guidance to
regulatory agency personnel on overseeing owners or operators of hazardous waste
management facilities in the conduct of the second phase of the RCRA Corrective
Ac+ion Program, the RCRA Facility Investigation (RFI). Guidance is provided for the
development and performance of an investigation by the facility owner or operator
based on determinations made by the regulatory agency as expressed in the
schedule of a permit or in an enforcement order issued under §3008(h), §7003,
and/or §3013. The purpose of the RFI is to obtain information to fully characterize
the nature, extent and ratt of migration of releases of hazardous waste or
constituents and to interpret this information to determine whether interim
corrective measures and/or a Corrective Measures Study may be necessary.
-------
DISCLAIMER
This document is intended to assist Regional and State personnel in exercising
the discretion conferred by regulation in developing requirements for the conduct
of RCRA Facility Investigations (RFIs) pursuant to 40 CFR 264. Conformance with this
guidance is expected to result in the development of RFIs that meet the regulatory
standard of adequately detecting and characterizing the nature and extent of
releases. However, EPA will not necessarily limit acceptable RFIs to those that
comport with the guidance set forth herein. This document is not a regulation (i.e.,
it does not establish a standard of conduct which has the force of law) and should
not be used as such. Regional and State personnel must exercise their discretion in
using this guidance document as well as other relevant information in determining
whether an RFI meets the regulatory standard.
Mention of company or product names in this document should not be
considered as an endorsement by the U.S. Environmental Protection Agency.
ii
-------
RCRA FACILITY INVEST1ATION (RFI) GUIDANCE
VOLUME IV
CASE STUDY EXAMPLES
TABLE OF CONTENTS
SECTION PAGE
ABSTRACT i
DISCLAIMER ii
TABLE OF CONTENTS iii
TABLES vii
FIGURES ix
LIST OF ACRONYMS xii
in
-------
VOLUME IV CONTENTS (Continued)
SECTION
14.0
14.1
14.2
15.0
INTRODUCTION
USE OF CASE STUDIES
ORGANIZATION OF VOLUME IV
CASE STUDIES
CASESTUDY1.
CASE STUDY 2.
CASE STUDY 3.
CASE STUDY 4.
CASE STUDY 5.
CASE STUDY 6.
CASE STUDY 7.
CASE STUDY 8.
CASE STUDY 9.
CASE STUDY 10.
CASE STUDY 11.
USE OF THE 40 CFR 261 LISTING BACKGROUND
DOCUMENTS FOR SELECTING MONITORING
CONSTITUENTS
ESTIMATION OF DEGRADATION POTENTIAL OF
CONTAMINANTS IN SOIL
SELECTION AND EVALUATION OF A SOIL
SAMPLING SCHEME
SAMPLING OF LEACHATE FROM A DRUM
DISPOSAL AREA WHEN EXCAVATION AND
SAMPLING OF DRUMS IS NOT PRACTICAL
USE OF QUALITY ASSURANCE/QUALITY
CONTROL (QA/QC) AND DATA VALIDATION
PROCEDURES
PRESENTATION OF DATA COLLECTED DURING
FACILITY INVESTIGATIONS
CORRELATION OF CONTAMINANT RELEASES
WITH A SPECIFIC WASTE MANAGEMENT UNIT
USING GROUND-WATER DATA
WASTE SOURCE CHARACTERIZATION FROM
TOPOGRAPHIC INFORMATION
SELECTION OF GROUND-WATER MONITORING
CONSTITUENTS AND INDICATOR PARAMETERS
BASED ON FACILITY WASTE STREAM
INFORMATION
USING WASTE REACTION PRODUCTS TO
DETERMINE AN APPROPRIATE MONITORING
SCHEME
CORRECTIVE MEASURES STUDY AND THE
IMPLEMENTATION OF INTERIM MEASURES
PAGE
14-1
14-1
14-1
15-1
15-1
15-6
15-10
15-14
15-19
15-29
15-43
15-47
15-50
15-54
15-58
IV
-------
VOLUME IV CONTENTS (Continued)
SECTION
CASE STUDY 12.
CASE STUDY 13.
CASE STUDY 14.
CASE STUDY 15.
CASE STUDY 16.
CASE STUDY 17.
CASE STUDY 18.
CASE STUDY 19.
CASE STUDY 20.
CASE STUDY 21.
CASE STUDY 22.
CASE STUDY 23.
CASE STUDY 24.
PAGE
USE OF AERIAL PHOTOGRAPHY TO IDENTIFY 15-63
CHANGES IN TOPOGRAPHY INDICATING
WASTE MIGRATION ROUTES
IDENTIFICATION OF A GROUND-WATER 15-68
CONTAMINANT PLUME USING INFRARED
AERIAL PHOTOGRAPHY
USE OF HISTORICAL AERIAL PHOTOGRAPHS 15-74
AND FACILITY MAPS TO IDENTIFY OLD WASTE
DISPOSAL AREAS AND GROUND-WATER FLOW
PATHS
USING SOIL CHARACTERISTICS TO ESTIMATE 15-78
MOBILITY OF CONTAMINANTS
USE OF LEACHING TESTS TO PREDICT 15-87
POTENTIAL IMPACTS OF CONTAMINATED SOIL
ON GROUND WATER
USE OF SPLIT-SPOON SAMPLING AND ON-SITE 15-97
VAPOR ANALYSIS TO SELECT SOIL SAMPLES
AND SCREENED INTERVALS FOR MONITORING
WELLS
CONDUCTING A PHASED SITE INVESTIGATION 15-105
MONITORING BASEMENT SEEPAGE 15-110
USE OF PREDICTIVE MODELS TO SELECT 15-114
LOCATIONS FOR GROUND-WATER
MONITORING WELLS
MONITORING AND CHARACTERIZING 15-119
GROUND-WATER CONTAMINATION WHEN
TWO LIQUID PHASES ARE PRESENT
METHODOLOGY FOR CONSTRUCTION OF 15-124
VERTICAL FLOW NETS
PERFORMING A SUBSURFACE GAS 15-137
INVESTIGATION
USE OF A SUBSURFACE GAS MODEL IN 15-144
ESTIMATING GAS MIGRATION AND
DEVELOPING MONITORING PROGRAMS
-------
VOLUME IV CONTENTS (Continued)
SECTION
CASE STUDY 25.
CASE STUDY 26.
CASE STUDY 27.
CASE STUDY 28.
CASE STUDY 29.
CASE STUDY 30.
CASE STUDY 31.
USE OF METEOROLOGICAL/EMISSION
MONITORING DATA AND DISPERSION
MODELING TO DETERMINE CONTAMINANT
CONCENTRATIONS DOWNWIND OF A LAND
DISPOSAL FACILITY
USE OF METEOROLOGICAL DATA TO DESIGN
AN AIR MONITORING NETWORK
DESIGN OF A SURFACE WATER MONITORING
PROGRAM
USE OF BIOASSAYS AND BIOACCUMULATION
TO ASSESS POTENTIAL BIOLOGICAL EFFECTS
OF HAZARDOUS WASTE ON AQUATIC
ECOSYSTEMS
SAMPLING OF SEDIMENTS ASSOCIATED WITH
SURFACE RUNOFF
SAMPLING PROGRAM DESIGN FOR
CHARACTERIZATION OF A WASTEWATER
HOLDING IMPOUNDMENT
USE OF DISPERSION ZONE CONCEPTS IN THE
DESIGN OF A SURFACE WATER MONITORING
PROGRAM
PAGE
15-153
15-158
15-165
15-174
15-185
15-188
15-194
VI
-------
TABLES (Volume IV)
NUMBER PAGE
14-1 Summary of Points Illustrated 14-2
15-1 Uses and Limitations of the Listing Background Documents 15-2
15-2 Results of Original Surface Soil and Tap Water Analyses 15-23
15-3 Laboratory QC Results 15-25
15-4 Field QC Results 15-26
15-5 Summary of Data Collected 15-33
15-6 Typical Methods for Graphically Presenting Data Collected 15-42
During Facility Investigations
15-7 Indicator Parameters 15-52
15-8 Results of Monitoring Well Sampling 15-55
15-9 Average Values of Parameters in Ground Water and Stream 15-73
Samples
15*10 Relative Mobility of Solutes 15-82
15-11 HEA Criteria, Constituent Concentrations and Relevant 15-91
Physical/Chemical Property Data For Constituents
Observed At Site
15-12 Leaching Test Results (mg/l) 15-93
15-13 Ground-Water Elevation Summary Table Phase II 15-127
15-14 Model Results 15-149
15-15 Summary of Onsite Meteorological Survey Results 15-156
15-16 Relationship of Dissolved and Sorbed Phase Pollutant 15-169
Concentrations to Partition Coefficient and Sediment
Concentration
15-17 Parameters Selected for Surface Water Monitoring Program 15-170
15-18 Selected Surface Water Monitoring Stations and Rationale 15-171
VII
-------
TABLES (Volume IV - Continued)
NUMBER PAGE
15-19 Mean Concentrations (jig/0 of Organic Substances and 15-178
Trace Metals in Leachate and Surface Waters
15-20 Mean Sediment Concentrations (pg/kg Dry Wt) of Organic 15-179
Substances and Trace Metals
15-21 Mean Liver Tissue Concentrations (iig/kg Wet Wt) of 15-180
Organic Substances and Trace Metals
15-22 Mean LCso and ECso Values (Percent Dilution) for 15-181
Surface-Water Bioassays
15-23 Relative Toxicity of Surface-Water Samples 15-182
15-24 Arsenic and Lead Concentrations (ppm) in Runoff 15-187
Sediment Samples
15-25 Summary of Sampling and Analysis Program for 15-191
Wastewater Impoundment
15-26 Relationship of Dissolved and Sorbed Phase 15-198
Contaminant Concentrations to Partition
Coefficient and Sediment Concentration
15-27 Parameters Selected For Surface Water 15-199
Monitoring Program
15-28 Selected Surface Water Monitoring Stations and 15-200
Selection Rationale
VIII
-------
FIGURES (Volume IV)
NUMBER
15-1
15-2
15-3
15-4
15-5
15-6
15-7
15-8
15-9
15-10
15-11
15-12
15-13
15-14
15-15
15-16
15-17
15-18
15-19
15-20
Results of Laboratory Bench Tests for Pesticide
Degradation
Isoconcentration Map of the Lead Concentrations in ppm
Around the Smelter
Schematic Diagram of Gas Control System Utilized at Pit
Schematic Drawing of Wireline Drill Bit and Reaming Shoe
Map of the Smelter Site and Associated Tailings Ponds
Locations of Copper Leach Plant and Waste Storage Ponds
Schematic of Surface Water System
Ground-Water Flowlines Based on Measured Water Levels
Selected Surface Water Quality Parameters at Key Stations
Changes in Suifate Over Time at Selected Wells Located
Within the Site
Field Sketch of Tailings Trench T-3
Depth vs. Concentration Profiles for Selected Variables
for Borehole 88A
Location of Ground-Water Monitoring Wells
Site Map Showing Waste Disposal Areas
Site Map and Monitoring Well Locations
Ground-Water Elevations and Flow Directions in
Upper Limestone Aquifer
October 1983 Aerial Photograph of Land Disposal Facility
Aerial Photograph Interpretation Code
February 1984 Aerial Photograph of Land Disposal Facility
Facility Plan View
PAGE
15-8
15-13
15-16
15-17
15-30
15-31
15-35
15-36
15-37
15-38
15-40
15-41
15-44
15-48
15-56
15-60
15-65
15-66
15-67
15-69
IX
-------
FIGURES (Volume IV - Continued)
NUMBER PAGE
15-21 Generalized Geologic Cross-Section 15.71
15-22 Infrared Aerial Photograph of the Site 15-72
15-23 Site Layout: LWDA-2, SDWA-2 and Stream Channel 15-76
Identified Through Use of Aerial Photograph Analysis
15-24 Schematic Cross-Section of Waste Disposal Site 15-80
15-25 Hypothetical Adsorption Curves for a) Cations and 15-83
b) Anions Showing Effect of pH and Organic Matter
15-26 Schematic Diagram Showing Plumes of Total Dissolved 15-86
Solids (TDS>, Total Organic Halogens (TOX) and Heavy
Metals Downgradient of Waste Disposal Site
15-27 Facility Map Showing Soil Boring and Well Installation 15-88
15-28 Facility Map Showing Ground-Water Contours 15-90
15-29 Site Plan Showing Disposal Areas and Phase I Well 15-98
Locations
15-30 Geologic Cross-Section Beneath Portion of Site 15-100
15-31 Ground-Water Elevations in November 1984 15-101
15-32 Example of Borehole Data Including HNU and 15-102
OVA/GC Screening
15-33 Proposed Phase II Soil Borings 15-107
15-34 Proposed Phase II Monitoring Wells 15-108
15-35 Geologic Cross-Section Beneath Sitt 15-111
15-36 Estimated Areal Extent of Hypothetical Plumes 15-116
from Four Wells
15-37 Consideration of Solute Migration Rates in Siting 15-118
Sampling Wells
15-38 Well Locations and Plant Configuration 15-121
15-39 Behavior of Immiscible Liquids of Different Densities 15-123
in a Complex Ground-Water Flow Regime
-------
FIGURES (Volume IV • Continued)
NUMBER PAGE
15-40 Top of Lowest Till Contour Map and Location of 15-125
Vertical Flow Net
15-41 Recharge/Discharge Areas and Flow Directions 15-130
15-42 Vertical Flow Net T-T' 15-132
15-43 Site Plan 15-138
15-44 Gas Monitoring Well 15-140
15-45 Facility Map 15-145
15-46 Uncorrected Migration Distances for 5 and 1.25% 15-147
Methane Concentrations
15-47 Correction Factors for Landfill Depth Below Grade 15-148
15-48 Impervious Correction Factors (ICF) for Soil Surface 15-150
Venting Condition Around Landfill
15-49 Landfill Perimeter Gas Collection System Wells 15-152
15-50 Site Map Showing Location of Meteorological Sites A and B 15-154
15-51 Site Plan and Locations of Meteorological Stations 15-159
15-52 Sampling Station Locations for Surface Water Monitoring 15-167
15-53 Site Plan and Water Sampling Locations 15-176
15-54 Bioassay Responses to Surface Water Samples 15-183
15-55 Surface Water and Sediment Sample Locations 15-186
15-56 Schematic of Wastewater Holding Impoundment 15-190
Showing Sampling Locations
15-57 Sampling Station Locations for Surface Water 15-196
Monitoring
XI
-------
LIST OF ACRONYMS
AA
Al
ASCS
ASTM
BCF
BOO
CAG
CPF
CBI
CEC
CERCLA
CFR
OR
CM
CMI
CMS
COD
COLIWASA
DNPH
DO
DOT
ECD
EM
EP
EPA
FEMA
FID
Foe
FWS
GC
GC/MS
GPR
HEA
HEEP
HPLC
HSWA
HWM
ICP
ID
Kd
Koc
Kow
LEL
MCL
MM5
MS/MS
NFIP
NIOSH
Atomic Absorption
Soil Adsorption Isotherm Test
Agricultural Stabilization and Conservation Service
American Society for Testing and Materials
Bioconcentration Factor
Biological Oxygen Demand
EPA Carcinogen Assessment Group
Carcinogen Potency Factor
Confidential Business Information
Cation Exchange Capacity
Comprehensive Environmental Response, Compensation, and
Lability Act
Code of Federal Regulations
Color Infrared
Corrective Measures
Corrective Measures Implementation
Corrective Measures Study
Chemical Oxygen Demand
Composite Liquid Waste Sampler
Dmitrophenyl Hydrazine
Dissolved Oxygen
Department of Transportation
Electron Capture Detector
Electromagnetic
Extraction Procedure
Environmental Protection Agency
Federal Emergency Management Agency
Flame lonization Detector
Fraction organic carbon in soil
U.S. Fish and Wildlife Service
Gas Chromatography
Gas Chromatography/Mass Spectroscopy
Ground Penetrating Radar
Health and Environmental Assessment
Health and Environmental Effects Profile
High Pressure Liquid Chromatography
Hazardous and Solid Waste Amendments (to RCRA)
Hazardous Waste Management
Inductively Coupled (Argon) Plasma
Infrared Detector
Soil/Water Partition Coefficient
Organic Carbon Absorption Coefficient
Octanol/Water Partition Coefficient
Lower Explosive Limit
Maximum Contaminant Level
Modified Method 5
Mass Spectroscopy/Mass Spectroscopy
National Flood Insurance Program
National Institute for Occupational Safety and Health
XII
-------
LIST OF ACRONYMS (Continued)
NPDES
OSHA
OVA
PID
pKa
ppb
ppm
PUF
PVC
QA/QC
RCRA
RFA
RfD
RFI
RMCL
RSO
SASS
SCBA
SCS
SOP
SWMU
TCLP
TEGD
TOC
TOT
TOX
uses
USLE
UV
VOST
VSP
WQC
National Pollutant Discharge Elimination System
Occupational Safety and Health Administration
Organic Vapor Analyzer
Photo lonization Detector
Acid Dissociation Constant
parts per billion
parts per mi 11 ion
Polyurethane Foam
Polyyinyl Chloride
Quality Assurance/Quality Control
Resource Conservation and Recovery Act
RCRA Facility Assessment
Reference Dose
RCRA Facility Investigation
Recommended Maximum Contaminant Level
Risk Specific Dose
Source Assessment Sampling System
Self Contained Breathing Apparatus
Soil Conservation Service
Standard Operating Procedure
Solid Waste Management Unit
Toxicity Characteristic Leaching Procedure
Technical Enforcement Guidance Document (EPA, 1986)
Total Organic Carbon
Time of travel
Total Organic Halogen
United States Geologic Survey
Universal Soil Loss Equation
Ultraviolet
Volatile Organic Sampling Train
Verticle Seismic Profiling
Water Quality Criteria
XIII
-------
14.0 INTRODUCTION
14.1 Use of Case Studies
This document, Volume IV of the RCRA Facility Investigation (RFI) Guidance,
contains case studies selected to illustrate various concepts and procedures
presented in Volumes I, II, and III. These case studies are provided to explain,
through example, how various tasks can be conducted during RFIs. The case studies
also identify some of the potential problems that can occur if the RFI sampling and
analytical programs are not carefully designed and executed. The case studies,
however, should not be used as the primary source of guidance for RFI program
design and conduct. Instead, Volumes I, II and III should be consulted. The studies
do not necessarily address details specific to individual facilities, and omission of
certain RFI tasks should not be interpreted as an indication that such tasks are
unnecessary or of less significance. Most of the case studies are based on actual
sites. In some cases, existing data have been supplemented with hypothetical data
to illustrate a particular point.
14.2 Organization of Volume IV
The case studies are organized primarily by the order in which the subject
matter was presented in Volumes I, II and III. In some cases, individual case studies
present materials relevant to more than one topic or media. Table 14-1 lists the
points illustrated and identifies the case studies which provide information relevant
to these points.
The following general format was used as appropriate for each case study:
• Tit*
• Identification of points illustrated
• Introduction/Background
• Facility description
• Program design/Data collection
• Program results/Data analysis
• Case discussion.
14-1
-------
TABLE 14-1
SUMMARY OF POINTS ILLUSTRATED
POINTS ILLUSTRATED
CASE STUDY
NUMBER
SELECTION OF MONITORING CONSTITUENTS
• Us* of 40 CFR Part 261 Listing Background Documents in selecting
monitoring constituents
e Consideration of degradation as a factor in identifying monitoring
constituents
SAMPLING SCHEMES
e Selection of a sampling scheme that appropriately characterizes soil
contamination
e Evaluation of the effectiveness of a sampling scheme using
statistical analyses
e Us* of release monitoring/leachate collection to characterize wastes
when the actual waste stream is inaccessible, as in the case of buried
drums
3
3
4
QUAUTY ASSURANCE AND CONTROL
• Use of quality assurance and control and data validation procedures
DATA PRESENTATION
* Techniques for presenting data for facility investigations involving
multimedia contamination
WASTE CHARACTERIZATION
* Correlation of a contaminant release with a specific waste
management unit using ground-water data
* Use of site topographic information to s*l*ct test boring and
monitoring well locations at facilities where large volumes of waste
have been disposed
* Use of waste stream information to select indicator parameters and
monitoring constituents in a ground-water monitoring program to
minimize the number of constituents that must be monitored
• Us* of information on possible wast* reaction products in designing
a ground-water monitoring program
7
8
10
CORRECTIVE MEASURES INCLUDING INTERIM MEASURES
• Us* of biodegradation and removal for interim corrective measures
• Corrective action and th* implementation of interim corrective
measures
2
11
AERIAL PHOTQgU PHY
• Us« of Mhal photographs to identify actual and potential wast*
migration routes and areas requiring comtctiv* action
• identification of a ground-wat*r contaminant plum* using infrared
aerial photography
• Us* of historical a*rial photographs and facility maps to identify old
wast* disposal areas and ground-water flow paths
12
13
14
14-2
-------
TABLE 14-1
SUMMARY OF POINTS ILLUSTRATED (continued)
POINTS ILLUSTRATED
CASE STUDY
NUMBER
SOIL
• Use of soil characteristics to estimate mobility of contaminants in
soil
• Effects of degradation in determining the fate of a contaminant in
soil
• Use of leaching tests to predict potential impacts of contaminated
soils on ground water
15
2
16
GROUND WATER
• Use of split-spoon sampling and organic vapor monitoring to select
screened intervals for ground-water monitoring
• Development of a two-phase boring program to investigate
ground-water contamination
e Use of basement monitoring to estimate contaminant migration
e Use of mathematical models to determine locations of ground-
water monitoring wells
• Monitoring and characterization of ground-water contamination
when two liquid phases are present
e Methodology for construction of vertical flow nets
17
18
19
20
21
22
SUBSURFACE GAS
e Design of a phased monitoring program to adequately characterize
subsurface gas migration
e Use of predictive models to estimate extent of subsurface gas
migration
23
24
AIR
Use of dispersion modeling and meteorological/emissions
monitoring data to estimate downwind contaminant
concentrations
Design of an upwind/downwind monitoring program when
multiple sources are involved
25
26
SURFACE WATER
e Use of existing site-specific data to design a surface water
monitoring program
• Use of bioassays and bioaccumulation studies to assess potential
biologkal effects of off-site contaminant migration
e Use of sediment sampling to indicate off-site contaminant
migration via surface runoff
e Design of a sampling program to account for three-dimensional
variations in contaminant distribution
e Use of dispersion zone concepts in the design of a surface water
monitoring program
27
28
29
30
31
14-3
-------
15.0 CASE STUDIES
CASE STUDY 1: USE OF THE 40 CFR 261 LISTING BACKGROUND DOCUMENTS FOR
SELECTING MONITORING CONSTITUENTS
Point Illustrated
• The 40 CFR 261 Listing Background Documents can be of direct help in
selecting monitoring constituents.
Introduction
The RCRA Hazardous Waste Listing Background Documents developed for the
identification and listing of hazardous wastes under 40 CFR Part 261 represent one
source of potential information on waste-specific constituents and their physical
and chemical characteristics. The documents contain information on the
generation, composition, and management of listed waste streams from generic
and industry-specific sources. In addition to identifying hazardous constituents that
are present in the wastes, the documents may also provide data on potential
decomposition products. In some background documents, migratory potentials are
discussed and exposure pathways are identified.
Appendix B of the Listing Document provides more detailed information on
the fate and transport of hazardous constituents. Major physical and chemical
properties of selected constituents are listed, including molecular weights, vapor
pressures and solubilities, octanol-water partition coefficients, hydrolysis rates,
biodegradation rates, and volatilization rates. Another section of the appendix
estimates tht migratory potential and environmental persistence of selected
constituents based on a conceptual model of disposal in an unconfined landfill or
lagoon.
The appropriate uses and limitations of the Listing Documents are outlined in
Table 15-1. A case study on how the Documents may be used in investigating a
release follows.
15-1
-------
TABLE 15-1
USES AND LIMITATIONS OF THE LISTING BACKGROUND DOCUMENTS
Uses
Limitations
• identifies the hazardous constituents for
which a waste was listed
• Applicable only for listed hazardous wastes
e in some cases, provides information on
additional hazardous constituents which
may be present in a listed waste
Industry coverage may be limited in scope
(e.g., the wood preserving industry). Listing
Documents only cover organic
preservatives, not inorganics (15 percent of
the industry), such as inorganic arsenic salts
In some cases, identifies decomposition
products of hazardous constituents
Data may not be comprehensive (i.e., not all
potentially hazardous constituents may be
identified). Generally, limited to the most
toxic constituents common to the industry
as a whole
Provides overview of industry; gives
perspective on range of waste generated
(both quantity and general characteristics)
Data may not be specific. Constituents and
waste characteristic data often represent an
industry average which encompases many
different types of production processes and
waste treatment operations
May provide waste-specific characteristics
data such as density, pH, and teachability
Listing Documents were developed from
data/reports available to EPA at the time,
resulting in varying levels of detail for
different documents
May provide useful information on the
migratory potential, mobility, and
environmental persistence of certain
hazardous constituents
Hazardous waste listings are periodically
updated and revised, yet this may not be
reflected in the Listing Documents
May list physical and chemical properties of
selected constituents
Listing Documents for certain industries
(e.g., the pesticides industry) may be subject
to CBI censorship due to the presence of
confidential business information. In such
cases, constituent data may be unavailable
(i.e., expurgated from the document)
15-2
-------
Facility Description
The facility is a wood preserving plant located in the southeast. The facility
uses a steaming process to treat southern pine and timber. Contaminated vapors
from the wood treating process are condensed and transported to an oil/water
separator to reclaim free oils and preserving chemicals. The bottom sediment
sludge from this and subsequent waste water treatment units is a RCRA listed
hazardous waste: K001.
Use of Listing Background Documents
Due to the presence of small, but detectable, levels of phenolic compounds in
the ground water of an adjacent property, a RCRA Facility Assessment (RFA) was
conducted and it was determined that a release from the facility had occurred. The
owner was instructed to conduct a RCRA Facility Investigation (RFI). Before
embarking on an extensive waste sampling and analysis program, the owner
decided to explore existing sources of information in order to better focus analytical
efforts.
The owner obtained a copy of the Wood Preserving Industry Listing
Background Document from the RCRA Docket at EPA Headquarters. He also had
available a copy of 40CFR Part 261, Appendix VII, which identifies the hazardous
constituents for which his waste was listed. For K001, he found the following
hazardous constituents listed: pentachlorophenol, phenol, 2-chlorophenol, p-
chloro-m-cresol, 2,4-dimethylphenyl, 2,4-dinitrophenol, trichlorophenols,
tetrachlorophenols, 2,4-dichlorophenol, creosote, chrysene, naphthalene,
fluoranthenc, benz(b)fluoranthene, benz(a)pyrene, ideno(1,2,3-cd)pyrene,
benz(a)anthractne,dibenz(a)anthracene, and acenaphthaiene.
From the Summary of Basis for Listing section in the Listing Document, the
owner found that phenolic compounds are associated with waste generated from
the use of pentachlorophenol-based wood preservatives, and that polynuciear
aromatic hydrocarbons (PAHs) (i.e., chrysene through acenaphthaiene in Appendix
VII) are associated with wastes from the use of creosote-based preservatives.
Examining the facility records, he determined that pentachlorophenol had been the
15-3
-------
sole preservative used; moreover, it had come from a single manufacturer. Based
on a demonstrable absence of creosote use, the owner felt confident in excluding
creosote and PAHs.
To help focus on which phenolics might be present in his waste, the owner
turned to the Composition section of the Listing Document. In Table 4, he found
typical compositions of commercial grade pentachlorophenol. The sample from his
manufacturer contained 84.6 percent pentachlorophenol, 3 percent
tetrachlorophenol, and ppm levels of polychlorinated dibenzo-p-dioxins and
dibenzo-furans. The owner was surprised by the absence of the other phenolics
mentioned in Appendix VII, and he was concerned by the presence of dioxins and
furans. Reading the text carefully, he discovered that the majority of the phenolic
compounds listed as hazardous constituents of the waste are actually
decomposition products of penta- and tetrachlorophenol. He also learned that
while the Agency had ruled out the presence of tetrachlorodibenzo(p)dioxin (TCOO)
in the listed waste (except where incinerated), it had not ruled out the possibility
that other chlorinated dioxins might be present: "... chlorinated dioxins have been
found in commercial pentachlorophenol and could therefore be expected to be
present in very small amounts in some wastes." Due to their extreme toxicity and
because his facility had historically used the commercial pentachlorophenol with
the highest concentration of dioxins and furans, the owner thought it prudent to
include a scan for dioxins in his waste analysis plan.
The owner found no further data in the Composition section to help him
narrow the list of phenolics; however, Table 6 gave a breakdown of organic
compounds found in different wood preserving plants (i.e., steam process vs.
Boneton conditioning), but only two phenolics were listed. A note in the text
highlights on* of the limitations of using the Listing Document: "The absence in
this table (TaMt 6) of certain components... probably indicates that an analysis for
their presence was not performed rather than an actual absence of the
component." It should be kept in mind that the waste analyses in the Listing
Background Documents are not comprehensive and that they are based, as the
Agency acknowledges, on data available at the time. In the absence of more
detailed waste-specific data, the owner decided to include pentachlorophenol,
tetrachlorophenol, unsubstituted phenol, and the six listed decomposition-product
phenolic compounds in his waste analysis plan.
15-4
-------
In reading the Listing Documents, the owner found useful information for
other phases of the RFI. In the Migratory Potential Exposure Pathways section, he
learned that pentachlorophenol is highly bioaccumulative, with an octanol/water
partition coefficient of 102,000. Tetrachlorophenol, tri-chlorophenol, and
dichlorophenol are likewise bioaccumulative, with octanol/water coefficients of
12,589,4,169, and 1,380, respectively. He also learned that the biodegradability of
pentachlorophenol is concentration limited.
In Appendix B of the Listing Background Documents; Fate and Transport of
Hazardous Constituents, the owner found data sheets for six out of nine phenolic
compounds, also some for dioxins and furans. Information on water chemistry, soil
attenuation, environmental persistence, and bioaccumulation potential were listed
along with chemical and physical properties such as solubility and density.
Case Discussion
Although the Listing Background Document did not provide the owner with
enough specific data to fully characterize his waste, it did help him refine the list of
monitoring constituents, alert him to the potential presence of dioxins, and gave
him physical and chemical waste characteristic data which could be useful in
predicting contaminant mobility.
15-5
-------
CASE STUDY 2: ESTIMATION OF DEGRADATION POTENTIAL OF CONTAMINANTS
IN SOIL
Point Illustrated
• Degradation, either chemical or biological, can be an important factor in
determining the fate of a contaminant in soil, and can also be a factor in
identifying constituents to monitor. The degradation rate can also be
accelerated as a means of conducting interim or definitive corrective
measures.
Introduction
Degradation of contaminants in the environment can occur through several
mechanisms, and can be a factor in identifying monitoring constituents. Under
natural conditions, these processes are often very slow, but studies have shown that
chemical and biological degradation can be accelerated in the soil by modifying soil
conditions. Parameters such as soil moisture content and redox condition can be
altered to encourage contaminant degradation in soils.
Site Description
The site is situated in an arid region that was used during the 1970s by aerial
applicators of organochlorine and organophosphate pesticides. The applicators
abandoned the site in 1980 and homes were built in the vicinity. The site can be
divided into three areas based on past use. The most contaminated area, the "hot
zone", is a 125 feet by 50 feet area at the north end of the site that was used for
mixing, loading, and unloading the pesticides. Soil samples from this area
contained toaaphtnt, ethyl parathion, and methyl parathion at concentrations up
to 15,000 mg/kg. The present residential area was used as a taxiway and an area to
rinse tanks and clean planes. Soils from this zone were low in parathions but
toxaphene concentrations ranging from 20 to 700 mg/kg were found. This area is
approximately 1.7 acres in size and located immediately south and west of the hot
zone. The runway itself was approximately 10 acres in size and south of the
residential zone. Soil sample results from the runway area had low concentrations
of all three pesticides.
15-6
-------
A number of factors influence degradation of organic compounds in soils.
These include:
• chemical nature of the compound
• organic matter content of the soil
• soil pH
• oxidation/reduction environment of the soil
• concentrations of the compound.
At the subject site, the soils were low in moisture content, were oxidizing, and
exhibited soil pH values of 6.8 to 8.0. Under such conditions, parathion can be
degraded slowly by alkali catalyzed hydrolysis reactions. The rate of these reactions
increases with increasing soil pH. Parathion can also be biodegraded to 0,O-Oiethyl
phosphoric acid. At a nearby site, it was shown that toxaphene will degrade
anaerobically if reducing conditions can be achieved in the soil. It has also been
observed that the loss of toxaphene by volatilization is enhanced by high soil
moisture content. Other data indicated that toxaphene will degrade in the
presence of strong alkali, by dechlorination reactions. This information can be used
in identifying monitoring constituents and in performing interim and definitive
corrective measures.
To test the feasibility of chemically degrading the contaminated soil, m, situ.
laboratory bench-scale tests were performed. Two treatments were evaluated,
application of calcium oxide (quicklime) and sodium hydroxide (lye). Figure 15-1
shows that the pesticides were degraded by both of these strong alkalis.
Those responsible for the remedial measures felt that the hot zone was too
contaminated for in situ treatment to be effective over reasonable time periods.
The upper 2 feet of soil from this area was excavated and transported to an
approved landfill for disposal. However, the 1.7-acre residential area was treated in
situ. To promote degradation, approximately 200 g/ft* of sodium hydroxide was
applied using a tractor with a fertilizer-spreading attachment. A plow and disc
were used to mix the sodium hydroxide into the soil to a depth of 1.5 feet. At 70
days after the application, concentrations of ethyl parathion had decreased by 76
percent, methyl parathion by 98 percent, and toxaphene by 45 percent.
15-7
-------
1000
800 •
600
I
400 •
200
25
20
15
10
• NaOH
o CaO
Laboratory Banch Tatt, Ethyl ParMhion Degradation
2468
DAYS
Laboratory Banch Tact, Methyl Parathion Degradation
17,500
15,000
12.500
10,000
7,500
5,000
2,500
• NaOH
eCaO
2468
DAYS
Laboratory Bench Tatt, Toxapbana Degradation
Figure 15-1. Results of Laboratory Bench Test for Pesticide Degradation
Swim: (from Kinf et d.. IMS).
15-8
-------
Case Discussion
Knowledge of the properties of a contaminant as well as its environment are
important in assessing the potential for degradation, and this information can be
used to identify monitoring constituents and conduct interim or definitive
corrective measures. It may be possible to alter the site's physical or chemical
characterisitcs to enhance degradation of contaminants. Under appropriate
conditions, m situ treatment of contaminated soils can be an effective corrective
measures method.
Reference
King, J., T. Tinto, and M. Ridosh. 1985. In Situ Treatment of Pesticide Contaminated
Soils. Proceedings of the National Conference of Management of Uncontrolled
Hazardous Waste Sites. Washington, O.C
15-9
-------
CASE STUDY 3: SELECTION AND EVALUATION OF A SOIL SAMPLING SCHEME
Points Illustrated
• Sampling methodologies must be properly selected to most
appropriately characterize soil contamination.
• Statistical analyses can be used to evaluate the effectiveness of a chosen
sampling scheme.
Introduction
Selection of a sampling scheme appropriate for a soil contamination problem
is dependent on the objectives of the sampling program. A grab sampling scheme
may be employed; however, grab sampling can produce a biased representation of
contaminant concentrations because areas of gross contamination are most often
chosen for sampling. Random sampling can provide an estimate of average
contaminant concentrations across a site, but does not take into account differences
due to the proximity to waste sources and soil or subsurface heterogeneities. A
stratified random sampling scheme allows these factors to be considered and, thus,
can be appropriate for sampling. Depending on the site, additional sampling using
a grid system may be needed to further define the areas of contamination.
Facility Description
The example facility operated as a secondary lead smelter from World War II
until 1984. Principal operations at the smelter involved recovery of lead from scrap
batteries. Air emissions were not controlled until 1968, resulting in gross
contamination of local soils by lead particulates.
Land use around the smelter is primarily residential mixed with
commercial/industrial. A major housing development is located to the northeast
and a 400-acre complex of single family homes is located to the northwest. Elevated
blood lead levels have been documented in children living in the area.
15-10
-------
Program Design/Data Collection
Initial soil sampling was conducted at the lead smelter and in the surrounding
area to document suspected contamination. Sample locations were selected based
on suspected areas of deposition of airborne lead and in areas where waste
dumping was known to have occurred. High lead concentrations were documented
in samples collected from these sources. Because data obtained in the exploratory
sampling program (grab sampling) were not adequate to delineate the area) extent
of contamination, a stratified random sampling scheme was developed.
Based on wind rose data and the behavior of airborne participate matter, a
sampling area was selected encompassing a 2-mile radius from the smelter. Specific
sampling sites were selected using a stratified random sampling scheme. The study
area was divided into sectors each 22.5 degrees wide and aligned so that prevailing
winds bisected the sectors. Each sector was further divided into approximately one-
tenth mile sections. A random number generator was used to select first the
direction and then the section. Random numbers generated were subject to the
following restrictions: two-thirds of the sites selected had to fall in the major
downwind direction; both residential and non-residential sites had to exist in the
sector; sampling sections were eligible for repeat selection only if they were
geographically within 1/2 mile of the smelter or if the section contained both
residential and non-residential sites. Sites that were biased towards lead
contamination from other than the lead smelter were not sampled (e.g., gas
stations and next to roads). A total of 20 soil sampling locations were selected, 10 at
residences and 10 at non-residential sites such as schools, parks, playgrounds and
daycare centers.
Sample cores were collected using a 3/4-inch inner diameter stainless steel
corer. Total sample depth was 3 inches. A minimum of four and maximum of six
samples were collected at each sampling location within a 2 ft radius. Cores were
divided into 1 inch increments and the corresponding increments were composited
from each depth to make up one sample. This approach provided data on lead
stratification in the top 3 inches of soil. All samples were analyzed for total lead.
15-11
-------
The results of the stratified random sampling indicated several acres with over
2,000 ppm lead in the soil. To further define the extent of these areas, a grid
sampling plan was designed. Seven hundred and fifty foot increments were used.
The grid was oriented along the axis of the release. Both residential and non-
residential areas were sampled. At each grid point, four 3-inch cores were collected
30 m from the grid point in each major compass direction. The cores were
composited by depth as discussed above.
Program Results/Data Anavsis
Analytical results from the soil sampling program indicated significant lead
contamination within the study area. Maximum concentrations observed were
2,000 ppm lead with a background level of 300 ppm. Kriegmg of the data from the
grid sampling plan was used to develop a contour map as shown in Figure 15-2.
Lead concentrations were highest northwest and southwest of the smelter.
Case Discussion
Because of the large area potentially affected by lead emissions, development
of a sequential sampling plan was necessary to determine the maximum soil lead
concentrations surrounding the smelter and the areas having elevated
concentrations. A grab sampling scheme was first used to confirm that soil
contamination existed. A stratified random sampling scheme was developed to
provide representative data throughout the study area. This type of sampling
allowed consideration of prevailing wind directions and the need to sample both
residential and non-residential areas. To further define areas of contamination, a
grid sampling plan was developed. From these data, lead isoconcentrations maps
were prepared d*lineating areas with elevated concentrations.
15-12
-------
-i r
Lead isovaives
• SMELTER
I I 1 1 1 1 1 1 1 I 1 P
ESTIMATED LEAD CONCENTRATIONS
-------
CASE STUDY 4: SAMPLING OF LEACHATE FROM A DRUM DISPOSAL AREA WHEN
EXCAVATION AND SAMPLING OF DRUMS IS NOT PRACTICAL
Points Illustrated
• It is not always possible to perform waste characterization prior to
establishing the RFI monitoring scheme because the waste may not be
directly accessible, as in the case of buried drums.
• When direct waste characterization is not practical, release monitoring
should be performed for the constituents listed in Appendix B of Volume
I of the RFI Guidance.
Introduction
Insufficient waste characterization data existed for a former drum disposal
facility that was suspected of releasing contaminants into the subsurface
environment. Leachate within the disposal pit was sampled and analyzed for all
constituents listed in Appendix B of Volume I of the RFI Guidance. The resulting
information was used to determine the major waste constituents to be monitored
during the RFI.
Facility Description
The unit of concern was a pit containing an estimated 15,000 drums. Due to
poor recordkeeping by the facility operator, adequate information regarding the
contents of the drums was not available. It was also not known if the drums were
leaking and releasing contaminants to the environment Because insufficient data
existed regarding the drum contents, it was not known what constituents should be
monitored in nearby ground and surface waters. Due to the risk to workers and the
potential for causing a multi-media environmental release, excavation and
sampling of the drums to determine their contents was not considered practical.
Instead, it was decided that leachate around the perimeter of the drum disposal pit
would be sampled to identify constituents which may be of concern.
15-14
-------
Program Design/Data Collection
To determine the physical extent of the buried drums, a geophysical survey
was conducted using a magnetometer. Borings were located at positions having
lower magnetometer readings than surrounding areas in order to minimize the
potential for drilling into drums.
Soil borings were drilled around the perimeter of the drum disposal pit, as
defined by the magnetometer survey. Drilling was accomplished using a hydraulic
rotary drill rig with a continuous cavity pump. Water was used as the drifting fluid.
To prevent surface runoff from entering the borehole and to control gaseous
releases from the borehole, primary and secondary surface collars were installed.
These consisted of 5-foot sections of 4-inch steel pipe set in concrete. A device to
control liquid and gaseous releases from the borehole was threaded onto the collars
to form a closed system (Figure 15-3).
Drilling was performed using a wireline operated tri-ccm roller bit with a
diamond tipped casing advancer (Figure 15-4). Water was pumped down inside the
casing and out the drill bit, returning up the borehole or entering the formation.
The use of water to aid in drilling also helped reduce the escape of gases from the
borehole. Air monitoring showed no releases. Split-spoon samples were collected
at 5-foot intervals during the drilling and a leachate monitoring well was installed
at each boring location.
The soil and leachate samples were analyzed for the compounds contained in
Appendix B of Volume I of the RFI Guidance.
Program ResuKs/Data Analysis
The leachate samples were found to contain high levels of volatile organic
compounds including 2-butanone, 4-methyl-2-pentanone, and toluene.
Concentrations were higher on the downgradient side of the pit
15-15
-------
r8MJ.-VM.VC OPCRATCD
SAMSUNG POUT
KCLLY /too
•KCU.Y SWVCL
•KCLLT HOSi
CNCLOSO
TANK (200 GM.)
S7QL
Figure 15-3. Schematic Diagram of Gas Control System Utilized at Pit
15-16
-------
WIRELINE CA8L£
OVERSHOT LATCHING
DEVICE
CASING
RETRACTABLE 2 15/16'
TRI-CONE ROLLER 3IT
W/ LOCKING INNER SU3
DIAMOND TIPPED CASING
ADVANCER (REAMING SHOE)
Figure 15-4. Schematic Drawing of Wireline Drill Bit and Reaming Shoe
15-17
-------
Case Discussion
Leachate sampling can be useful in determining whether buried drums are
leaking and to identify materials that are being released. This methodology can be
safer than excavation and sampling of individual drums. It can also identify the
more soil-mobile constituents of the leachate.
The data gathered in this case study were used in designing a monitoring
program, and the contaminants found were used as indicator compounds to link
downgradient ground-water contamination to this waste disposal unit.
15-18
-------
CASE STUDY 5: USE OF QUALITY ASSURANCE/QUALITY CONTROL (QA/QC) AND
DATA VALIDATION PROCEDURES
Points Illustrated
• A comprehensive field and laboratory QA/QC program is necessary for
assessing the quality of data collected during an RFI.
• Timely validation of laboratory data can uncover problems correctable by
reanalysis or by resampling, thus preventing data gaps.
Introduction
A company in the mining and smelting industry sampled domestic wells and
surface soils in the vicinity of a tailings pile to monitor possible leaching of metals
into the aquifer and possible soil contamination due to wind-blown dust. Because
the data would be used to assess corrective measures alternatives and to conduct a
health and environmental assessment the company chose to conduct both its
sampling and analysis efforts under a formal QA/QC Project Plan and to subject all
laboratory data to rigorous data validation procedures.
Facility Description
At this facility, a tailings pond had received smelter waste for many years.
Local water supply wells were potentially at risk due to percolation of water
through the pile and possible leaching of heavy metals. Local surface soils in nearby
residential areas (e.g., yards, public playgrounds) were also subject to
contamination from wind-blown dust originating from the pile during dry windy
weather.
Sampling Program
Before sampling began, a set of documents were drafted following U.S. EPA
guidelines (U.S. EPA 1978,1980a, 1980b, 1981,1982,1985a, 1985b) that specified in
detail sampling sites and parameters to be measured, field and laboratory
15-19
-------
procedures, analytical laboratory protocols, and all field and laboratory QC checks
including frequencies, and corrective actions. The important elements of each
document are described below.
Standard Operating Procedures (SOPs)--
This document contained step-by-step procedures for the following items:
• Calibration, operation, and maintenance of all instruments used in the
field and laboratory.
• Equipmentdecontamination.
• Ground water and soil sampling, including compositing.
• Use of field notebooks and document control.
• Sample packaging, shipping, and chain-of-custody.
Field Operations Plan (FOP)--
This document included the following:
• Rationale for choice of sampling locations, sampling frequency, and
analytes to be measured
• List of sampling equipment and SOPs to be used for each sampling event.
• List of field QC checks to be used and their frequency for each sampling
event
• Health and safety issues and protective measures for field personnel.
• Sampling schedule.
15-20
-------
Laboratory Analytical Protocol (LAP)--
This document included the following:
• Sample size, preservation, and analysis protocol for each analyte.
• List of laboratory QC checks, QC statistics to be calculated and their
control limits, and corrective actions for QC checks outside control limits.
• Detailed list of deliverable documents and their formats.
• Procedures for sample custody, independent audits, and general
laboratory practices.
QA/QC Project Plan (QAPP)--
This document gathered into one place the overall data quality objectives for
the sampling and detailed QC procedures needed to attain those objectives.
Included were:
• Quality assurance objectives in terms of precision, accuracy,
completeness, comparability, and representativeness.
• Procedures for the screening of existing data.
• Data management reduction, validation, and reporting.
• Overview of both field and laboratory QC checks and their frequencies,
control limits, and corrective actions.
• Data assessment procedures.
Results
Five surface soil samples were taken in high traffic areas of two playgrounds
and three residential yards. Five tap water samples were collected at two public
15-21
-------
drinking fountains at the playgrounds and at the three private residences. The
analysis results, as received from the laboratory, are shown in Table 15-2. The data
indicated that a soil hot spot existed for cadmium, that elevated lead occurred at all
five soil stations, and that all of the domestic wells showed elevated levels of
mercury.
The laboratory data package was subjected to a thorough data validation, as
detailed in the QA Project Plan. The following information and QC results were
checked by examination of original documents or photocopies of the documents.
Sampling, Sample Shipping, Chain-of-Custody-
Copies of field and field laboratory notebook pages were examined to insure
that ail SOPs were correctly followed, that there were no notations of anomalous
circumstances (such as sample spillage) that may have affected analysis results, and
that the samples were correctly preserved, packaged, and shipped. Copies of all
chain-of-custody forms, bills-of-lading, and sample analysis request forms were
examined to insure that chain-of-custody was not broken and that samples arrived
intact at the laboratory.
Laboratory Raw Oata-
The QAPP had specified that one of the deliverables from the laboratory was
copies of all instrument readouts and laboratory notebook pages. The digestion
raw data were checked to insure that no holding time violations had occurred. This
is important for mercury because the holding time is only 28 days for aqueous
samples.
All raw calibration data were recalculated and tested against instrument-
calculated sample results. Recoveries of calibration verification standards and
continuing calibration standards were checked to insure that all instruments were
correctly calibrated, were not drifting out of calibration, and were correctly
calculating raw analysis results.
15-22
-------
TABLE 15-2
RESULTS OF ORIGINAL SURFACE SOIL AND TAP WATER ANALYSES
Sample*
SOIL-1
SOIL-2
SOIL-3
SOIL-4
SOIL-5
WATER- 1
WATER-2
WATER 3
WATER-4
WATER-5
Cd
14
7
<20e
19
1200
<50
<50
<50
<50
<50
Cu
5200
2400
720
680
1080
NA
NA
NA
NA
NA
Pb
800
400
4530
350
460
<30
<30
<30
<30
<30
Hg
NA°
NA
NA
NA
NA
1.5
1.3
1.0
1.4
1.2
Zn
1200
190
70
350
420
NA
NA
NA
NA
NA
a Soils in units of mg/kg, water in ug/L
t> Not analyzed.
c Undetected at detection limit shown.
15-23
-------
Final analysis results were recalculated from raw data using dilution and
digestion factors, as summarized in the lab notebooks, and compared to the data
summary sheets. No transcription errors were found. However, the cadmium result
for SOIL-5 contained a calculation error, and the correct final result was 12 mg/kg
instead of the 1200 mg/kg reported.
Laboratory QC Checks-
The QAPP had specified that the laboratory had to analyze p re-digestion
duplicates and spikes, U.S. EPA laboratory control samples, and reagent blanks. The
laboratory QC results are summarized in Table 15-3 and indicated accuracy and
precision well within U.S. EPA guidelines. The mercury preparation blank also
indicated that the tap water results were not due to laboratory digestion reagents
or procedures.
Field QC Checks--
As specified in the QAPP and FOP, the following field QC samples were
included with each of the soils and tap water samplings: bottle blank, field blank,
standard reference material (SRM), triplicate, and an interlaboratory split to a
"reference" lab. The results are summarized in Table 15-4.
Although no U.S. EPA control limits or corrective actions exist for field-
generated QC checks, the results of their analysis can aid in the overall assessment
of data quality. The triplicate, SRM, and interlaboratory split analyses indicated
good overall analysis and sampling precision and accuracy. The field blanks
indicated tht possibility of mercury contamination from one of the four possible
sources: thcprt-deaned bottles, the preservation reagent, the distilled water used
in the field, or an external contamination source such as dust. The high positive
mercury result in the water bottle blank eliminated all of these sources except the
first because the bottle blanks remained sealed throughout the sampling effort.
The laboratory was immediately called and, upon personal inspection, the
laboratory manager discussed the remnants of a broken thermometer bulb in the
plastic tub used to acid-soak the bottles. An unused bottle from the same lot and
15-24
-------
TABLE 15-3
LABORATORY QC RESULTS
Analyte
Cd
Cu
Pb
Hg
Zn
Duplicate RPD'(%)
SOIL-2
18
S
14
NA
7
WATER-4
NC*
NA"
NC
NC
NA
Spike Recovery "(%)
SOIL-2
100
93
110
NA
85
WATER-4
98
NA
92
103
NA
LCS<
(%)
101
97
106
NA
99
Soil
Preparation
Blank*
<509
<100
<200
NA
<150
Water
Preparation
Blank*
<50
NA
<3Q
<0.20
NA
a RPO s relative percent difference * (difference/mean) x 100. Controllimits 3 135% for
solids and ± 20% for aqueous samples.
b Spike Recovery » (spike + sample result) • (sample result) x 100. Control limit.» 75-125%.
(spike added)
c LCS = laboratory control sample. Control limit -90-110%.
d mg/kg.
ug/i.
NC = not calculated due to one or both concentrations below detection limit.
Undetected at detection limit shown.
NA = not analyzed.
15-25
-------
TABLE 15-4
FIELD QC RESULTS
Analytt
Cd
Cu
Pb
Hg
Zn
Triplicate
CV'(%)
SOIL-1
22
3
7
NA
1
WATER- 1
NC
NA'
NC
18
NA
SRM
Recovery** (%)
BCSS-T
83
94
97
NA
110
U.S. EPA*
105
NA
101
103
NA
interlab.
RPDf(%)
SOIL-1
-12
0
14
NA
24
WATER- 1
NC
NA
NC
19
NA
Field
Blanks'
SOIL-1
Recovery « (certified value/result) X100.
c National Research Council of Canada marine sediment
-------
still at the laboratory as well as two bottles washed in previous lots were analyzed.
The bottles previously washed contained no detectable mercury, and the bottle
from the same lot as used in the sampling effort contained 0.75 ug. The water
mercury data were rejected, and a second sampling effort using new bottles was
conducted. All of the new samples contained no detectable mercury.
Discussion
This case study demonstrates the need for the establishment of a formal
QA/QC program that not only specifies field QC protocols but also incorporates
thorough data package validation. In this instance, a potential hot spot was found
to be due to a calculation error, and potential mercury contamination of domestic
well water was found to be a result of using contaminated sample containers. In
the latter case, timely QA/QC review allowed for a speedy resampling effort which
could be done at this site. In situations where resampling is not possible, adequate
QA is crucial.
References
U.S. EPA. 1978 (revised 1983). NEIC Policies and Procedures. EPA-33079-78-001-R.
U.S. EPA, National Enforcement Investigations Center, Denver, CO.
1978 (revised 1983). NEIC Policies and Procedures. EPA-330/9-78-001-R. U.S. EPA,
National Enforcement Investigations Center, Denver, CO.
U.S. EPA. 1980a. Interim guidelines and specifications for preparing quality
assurance project plans. QAMS-005/80. U.S. EPA. Office of Monitoring Systems and
Quality Assurance, Washington, DC. 18pp.
U.S. EPA. 1980b. Samplers and sampling procedures for hazardous waste streams.
EPA-600/2-80-018"U.S. EPA, Municipal Environmental Research Laboratory,
Cincinnati, OH.
U.S. EPA. t9i1. Manual of qrqundwater quality sampling procedures.
EPA-600/2-8t-T«0. Roberts. Kerr Environmental Research Laboratory, Ada, OK. 105
PP-
U.S. EPA. 1986. Test methods for evaluating solid waste. SW-846. 3rded. U.S. EPA,
Office of Solid Waste and Emergency Response, Washington, DC.
15-27
-------
U.S. EPA. 1985a. Contract laboratory prot
analysis, multi-media, multi-concentration.
Qf
U.J
,-..,, ........ ..,^w,q.. muiu-iuriteniranon. iUW No 785 Ju u idi
Environmental Monitoring Support Laboratory, Las Vegas, NV
:UJ:.E?A:'. ly.Sb.^LatioratorY data validation Funrtional guidelines for evaluating
jmedial
inorganic analysis' October, 1985.
Response, Washington, DC.
U-S-
.ee o
aneme
15-28
-------
CASESTUDY6: PRESENTATION OF DATA COLLECTED DURING FACILITY
INVESTIGATIONS
Point Illustrated
• Techniques for presentation of data for facility investigations involving
multimedia contamination.
Introduction
Data acquisition and interpretation are integral parts of facility investigations.
Depending on the size, complexity, and hazards posed at a particular site,
significant quantities of meteorologic, hydrologic, and chemical data can be
collected. To make the best use of these data, they should be presented in an easily
understood and meaningful fashion. This case study focuses on widely used and
easily implemented graphical techniques for data presentation.
Site Description
The site is a former copper smelter that ceased operation in the early 1980s.
During the operation of the smelter, large quantities of mine tailings were slurried
to tailings ponds that remain today (Figure 15-5). The tailings contain high solid
phase concentrations of inorganic contaminants such as copper, zinc, lead,
cadmium, and arsenic. In the Smelter Hill area, flue dust and stack emission
deposition have contaminated surficial soils. Numerous other units were operated
at the complex including an experimental plant designed to leach copper using
ammonia. The copper leach plant is shown in Figure 15-6. Three disposal ponds (I,
II, and III) received wastes slurried from the plant.
As a result of smelting and waste disposal practices, multimedia contamination
of ground water, surface water, and soils has occurred. Also, episodes of air
contamination have been documented due to entrainment of tailings during windy
periods.
15-29
-------
l/t
•o
o
"8
««
2
'3
o
(9
0)
o
a
tn
i
ui
15-30
-------
•a
o
Ou
o
V
I
c
c
u
HJ
0)
a
a
o
u
o
u
O
0;
3
O»
15-31
-------
Field Sampling and Data Collection
Data collection activities at this site were comprehensive. Over 100,000 pieces
of data were collected in the categories shown in Table 15-5.
Oata Presentation
This section illustrates a number of graphical techniques that can be used to
present data from facility investigations. Graphical presentations are useful for the
following general purposes:
• Site feature identification, source identification, and mapping;
• Hydroiogic characterization; and
• Water quality characterization.
For large sites, aerial photography is often very useful for defining the locations and
boundaries of waste deposits, and for establishing time variability of site
characteristics. Figure 15-6, for example, was developed from aerial photographs at
a 1:7800 scale. Types of information obtained by comparing this photograph to one
taken 10 years earlier include:
• Pond III was originally constructed earlier than Ponds I and II, and was not
lined. Ponds I and II wtrt lined.
• Tht red sands (a slag deposit) shown in Figure 15-6 are present only north
of tht railroad tracks. Earlier photographs showed that the red sands
exfcwrfed to the highway, but were leveled and covered with alluvium
during construction of the copper leach plant.
This type of photographic* information is valuable for locating waste deposits,
estimating quantities of wastes, and determining waste proximity to sensitive areas.
15-32
-------
TABLE 15-5
SUMMARY OF DATA COLLECTED
Category
Ground Water
Surface Water and
Sediment
Alluvium'
Soil*
Tailings*
Slag and Flue Oust*
Miscellaneous
Parameters
Water level elevations, potentiometric heads
Concentration of Al, Sb, As, Ba, Be, Bo, Cd, Ca, Cr,
Co, Cu, Fe, Pb, Mg, Mn, Hg, Mo, Ni, K, Se, Ag, Na,
Sn, V, Zn, P, Cl, F, S04, pH, 0?, EC, Eh, Alkalinity,
TDS
Flow rates, bed particle size distributions,
suspended solids concentrations, dissolved
concentrations of same inorganic parameters as
ground water
Moisture content, soil, pH, EC, Sb, As, Cd, Cu, Fe,
Pb, Mn, Se, Ag, Zn, particle-size distribution
Cd, Cu, Fe, Pb, Mn, Ni, Zn, Sb, As, Cd, Cr, Hg, Se,
Ag, Zn, particle-size distribution, Eh, S, TOC
Sb, Ar, Be, Cd, Cu, Fe, Pb, Mn, Ag, Se, Zn, particle
size, moisture, pH, EC, sulfur, carbonate
Sb, As, Cd, Cu, Fe, Pb, Mn, Se, Ag, Zn, S04, EC, pH,
alkalinity
Meteorology, aerial photographs and other
photographic documentation, well log data,
surface topography, volumetric surveys of waste
piles
Element data are solid phase.
15-33
-------
For sites with complex hydrologic interaction, it is often helpful to graphically
represent the flow system. Figure 15-7 illustrates the surface water system at the
site. The diagram is useful because it shows the hydrologic interconnections of the
drainage system.
For the ground-water system, flow direction and velocities provide
information needed for contaminant transport predictions. This information is
generated by plotting water levels on a site map, and then drawing contours
through points of equal elevation. An example is shown in Figure 15-8. Because the
contours form a relatively simple pattern in this case, they were drawn by hand.
However, computer-based contour packages exist that could be used to plot more
complicated contour patterns.
Inferred flow directions are also shown in Figure 15-8. From a knowledge of
the hydraulic gradient, hydraulic conductivity and effective porosity, the average
linear velocity can be calculated, as shown in the upper left hand corner of the
figure. A velocity of 79 m/yr is calculated, for example, which means that
approximately 126 years would be required for conservative solutes to move across
the site (approximately 10,000 meters).
Water quality data can be presented as shown in Figure 15-9. This figure
shows the spatial distribution of calcium, sulfate, and IDS at key surface water
stations. This data presentation method provides a broad area! view of these
parameters.
Time series plots art useful for showing temporal variations in water quality.
For example, time trends of SO4 at three ground-water monitoring locations are
shown in Figure 15-10. Well 19 is slightly downgradient from the source, and the
high S04I eve* reflect that the well is receiving solutes generated within the source.
Wells 26 and 24 are further upgradient, and reflect better water quality conditions.
The plot indicates that variability between stations generally is more significant
than time variability at a given location. One exception is at well 24 where a
temporary increase in sulfate levels was noted in 1975-76.
15-34
-------
UNCAGED
DIVERSIONS
NCW i
LIMI ;
DITCH!
RANGUS
CREEK
HILL
CREEK
SOUTH
DITCH
OLD
UM€
DITCH
SEWAGE
DITCHES
BYPASS
r' /"
GOLDEN
CREEK
PONOS
DRAIN
DITCH
COLD
CREEK
GARDINER
OITCM
UNCAGED
DIVERSIONS
GREEN
RIVER
Figure 15-7. Schematic of Surface Water System
15-35
-------
0)
01
*-
i
c
o
1
-------
X n/9WI NOUTX1N33M03 ' '
15-37
-------
2000-
= 1600-
S
cT
W 1200-
800-
400-
• Well 19
a Well 24
• Wen 26
1974 1975 1976 1977 1978 1979 I960 1981 1982 1983 1984 1985
Y«ar
Figure 15-10. Changes inSulfate Over Time at Selected Wells
Located Within the Site
15-38
-------
To identify leachate and soil interactions beneath a waste site, trenches may
be dug. The trench walls are then logged and photographed. Detailed sampling
may be done at closely spaced intervals to confirm that reactions such as
precipitation have occurred. Figure 15-11 shows a cross-section of a tailings deposit
that was developed based on a trench excavated through the tailings into the
underlying alluvium. The plot shows the demarcation between wastes and natural
alluvium.
Figure 15-12 shows the details of the chemical composition of one borehole
through the tailings and into the underlying alluvium. The chemical composition is
shown to varv significantly with depth. These types of plots contain a wealth of
chemical information that can help to explain the geochemical processes operative
in the tailings. Figure 15-12 also shows the marked contrast between the
composition of the tailings (in the top 16 feet) and the underlying alluvium.
Summary
The graphical presentations illustrated in this case study are a few of the many
techniques available. With the proliferation of graphical packages available on
microcomputers, scientists and engineers have a wide range of tools available for
data presentation. Some of these tools are summarized in Table 15-6.
15-39
-------
o—*
a
5
w
.J
5
M
LU
5
kU
c
at
f
is
i
s
^^
z
2<
K 5
32
Q M
3 C
55 **
w !*t
2 ^
IU U<
5 £
z z
si
9 Ul
s *
o $
* 5
Ul ^
H!
ii '
a!
*1
< S
| y
32
« i
< it
I
"o
2
a»
in
15-40
-------
!
i
1 i
t
»«
1 ••*;sss»
i
i
Si
z a.
I
'» s j
kit
••«••» 9: *•«
JS
s : * * «
I
I 3
•I
I «
I
>t 3 :
c
o
c <
0) 00
g «
o ^
u o
>t : :
•j
:i
f
t
! *
I
^: s
• *
I^B^^^I^^M^B • •HH^B^Hi^^
OJ
3
il
••••fSSt««
15-41
-------
TABLE 15-6
TYPICAL METHODS FOR GRAPHICALLY PRESENTING DATA COLLECTED
DURING FACILITY INVESTIGATIONS
Data
METEOROLOGIC DATA
Wind speed and direction
Air temperature
Precipitation
Evaporation
SURFACE WATER DATA
Flow rates
Water quality
GEOHYDROLOGIC DATA
GROUND-WATER DATA
MISCELLANEOUS
Graphical Presentation Methods
e Wind rose showing speed, direction and percent of
observations for each 10° increment
e Bar chart, by month
e Bar chart, by month
e Bar chart, by month
e Hydrographs; distance profiles, cumulative frequency
distributions, flood frequency plots
Hydrologic network depiction and water budgets
Tri linear diagram
Stiff diagrams
Contour showing vertical concentration or temperature
variability in two deep water bodies
Time history plots showing daily/annual variability
Bar charts of major cations/anions or contaminants at
multiple locations shown on a single map
Geologic map of site and vicinity
Stratigraphic cross-sections of site in direction of and
perpendicular to ground water flow
Well logs
Cross-sections near waste deposits
Solid phase chemical analyses by depth at borings near
waste deposits and into alluvium
Water level contours
Flow directions and velocities
Time history of water table at important locations
Stiff diagrams
Tri linear diagrams
Contaminant plumes, showing isopleths
• Figures with important site features, including waste
sources, storage ponds, disposal areas, buildings,
sampling locations, well locations
• Operational aspects for special sampling equi pment
(e.g., lysimeters)
15-42
-------
CASE STUDY 7: CORRELATION OF CONTAMINANT RELEASES WITH A SPECIFIC
WASTE MANAGEMENT UNIT USING GROUND-WATER DATA
Point Illustrated
i
• Development of an effective ground-water monitoring program can tie
releases of contaminants to specific waste mangement units.
Introduction
Documentation of a release from a specific waste management unit may
require the development of a comprehensive ground-water monitoring program
coupled with an extensive hydrogeologic investigation. Determination of ground-
water flow direction and horizontal and vertical gradients are necessary to assess
the direction of potential contaminant migration. Historical data on wastes
disposed in specific units can provide information on contaminants likely to be
detected downgradient.
Facility Description
Chemicals were manufactured at a 1000-acre facility for over 30 years. The
facility produced plastics including cellulose nitrate, polyvinyl acetate, poly vinyl
chloride and polystyrenes, and other chemicals such as phenols and formaldehyde.
Wastes produced in the manufacturing processes were disposed on site in an
unlined liquid waste impoundment and in two solid waste disposal areas. Readily
combustible materials were incinerated in four burning pits. Ground-water
contamination has been documented at the site. Figure 15-13 shows the facility
plan and locations of ground-water monitoring wells.
The site is located in a glacial valley and is adjacent to a major river. A minor
tributary runs through the southwestern portion of the facility and drains into the
river. Approximately 200 dwellings are located downgradient of the site.
15-43
-------
I
rf
1.8
CM
cu
O)
c
c
o
u.
01
15
i
•o
c
3
O
o
c
o
Oi
3
O>
15-44
-------
Data Collection
Initial studies to assess the extent of ground-water contamination began m
1981. Studies focused on ground water in the vicinity of various waste disposal
units. A limited number of monitoring wells were installed in 1983. These wells
provided general data on the direction of ground-water flow and chemical
constituents that had entered the ground water. In 1984, a two-phased approach
was developed to define the areal and vertical extent of contamination and to
identify contaminant releases from specific waste management units. The first
phase involved the characterization of facility geologic and hydrogeologic
conditions using historical data, determination of the chemical nature of
contaminants in the ground water using existing monitoring wells, and
development of a contaminant contour map delineating the horizontal boundaries
of contamination. Based on this data, 33 soil borings were drilled in Phase 2. The
goals of the second phase were to: 1) detail subsurface geologic characteristics,
vertical and horizontal water flow patterns, contaminant migration, and site-
specific chemical contaminants; and 2) install wells that would be used to monitor
contaminants being released from all units of concern at the facility.
Continuous split spoon samples were collected in each boring and headspace
analyses for volatile organic compounds (VOC) were conducted on each sample.
Chemical constituents were identified using a field gas chromatograph.
Confirmational analysis by GC/MS were conducted on selected samples.
Geotechnical analyses were also conducted on the split spoon samples.
Chemical and hydrogeologic data (direction of flow, gradients) obtained from
the borings were used to select appropriate ground-water monitoring well
locations and screen depths. Fifty-two (52) nested monitoring wells were installed
at 25 locations upgradient and downgradient of each waste management unit, and
near the river and its tributary. Screen depths were determined by the depth of
maximum VOC contamination observed in the borings and the permeability of soil
layers.
15-45
-------
Data Analysis
Ground-water contamination data from new wells coupled with historical
waste disposal data allowed releases from three specific waste management areas
to be defined. Sample analyses showed organic solvents in nearly all locations.
However, more unusual constituents associated with specific manufacturing
processes were detected in some samples, allowing them to be correlated with
releases from specific waste management units. The two situations below illustrate
how these correlations were accomplished:
1) PCBs detected in some samples were correlated with Solid Waste Disposal Area
#1. This area received construction debris, resins, plastics, metals, drums, and
PCB containing transformers. Records indicated that this unit was the only
location where transformers were disposed onsite. PCBs could not be
associated with any of the other waste management units.
2) The solvent dimethylformamide (DMF) detected in some samples was
correlated with Burning Pit B. It was discovered that the building that housed
this unit had been used to tint windshields and that DMF is a component of
the dye used in this process. DMF could not be tied to any of the other waste
management units. A leachfieid in which waste dyes had been disposed was
discovered under the building and the contamination was traced back to that
source.
Case Discussion
An exttmivt hydrogeologic investigation of the facility was completed and, in
conjunction wWt historical data, was used to develop a comprehensive ground-
water monitoring program. Placement of the monitoring wells and screens was
essential in providing data that unequivocally linked contaminant releases to
specific waste management units and manufacturing processes.
15-46
-------
CASESTUDY8: WASTE SOURCE CHARACTERIZATION FROM TOPOGRAPHIC
INFORMATION
Points Illustrated
• Mapping of changes in site topography can support the selection of
locations for test borings and monitoring wells.
• This technique is especially useful at sites where large volumes of waste
have been disposed of over several years.
Introduction
Topographic surveys conducted prior to and at different times during the
operation of a waste management facility can be used to help characterize the
vertical and horizontal extent of waste disposal areas. Because the resolution of
this technique is limited, it is most useful when large volumes of waste are involved.
Facility Description
This facility is the same as discussed in Case Study 7 above.
Topographic Survey
In 1984, a topographic survey measuring elevations in feet relative to mean
sea level was conducted for the areas shown in Figure 15-14. These elevations were
plotted on a map of appropriate horizontal scale and contoured in 2-foot intervals.
This topography was transferred to an existing site plan (horizontal scale 1" to
200'). Topographic maps from 1935 (showing the natural topography before waste
deposition) to 1960 (showing the topography in the earlier .stages of the facility
operation) were compared to the 1984 map. By examining the changes in
elevations which occurred over time, contours were developed showing the
estimated changes in vertical and horizontal units of the liquid waste and solid
waste disposal areas.
i
15-47
-------
-------
Results
From the analysis, it was apparent that the deepest portion of Solid Waste
Disposal Area (SWDA) No. 1 (Figure 15-14) was approximately 48 feet, and the
Liquid Waste Disposal Area (LWDA) was approximately 30 feet deep. The
horizontal limits of the disposal areas were also defined in part by this review, but
other field surveys provided more accurate information on the horizontal
boundaries of the waste disposal areas.
Case Discussion
Topographic surveys can provide useful information for characterizing
disposal areas. The results of these studies can facilitate the selection of
appropriate test boring locations, and may reduce the number of borings necessary
to describe the subsurface extent of contamination. It should be noted that
techniques such as infrared aerial photography and topographic surveying are
approximate in their findings. They are useful methods in the early phases of an
investigation, but do not replace the comprehensive characterization of the
environmental setting needed for the full investigation.
15-49
-------
CASE STUDY 9: SELECTION OF GROUND-WATER MONITORING CONSTITUENTS AND
INDICATOR PARAMETERS BASED ON FACILITY WASTE STREAM
INFORMATION
Points Illustrated
• Waste stream information can be used to identify potential
contaminants, and thus to select appropriate ground-water monitoring
constituents and indicator parameters.
• The number of initial monitoring constituents analyzed may be
significantly reduced from the 40 CFR Part 261 Appendix VIII list when
detailed waste stream information is available.
Introduction
Hazardous waste treatment, storage, and disposal facilities subject to RCRA
are required to identify all waste streams managed the facility, waste volumes,
concentrations of waste constituents, and the waste management unit in which
each waste type is disposed. Ground-water monitoring programs should be
developed to adequately monitor contaminant migration from each unit.
Constituents to be analyzed in the ground-water monitoring program should be
established prior to sample collection. When waste stream data are not available,
the full set of Appendix VIII monitoring constituents may be required to
characterize ground-water contamination. Knowledge of the waste streams
managed by a facility simplifies the selection of monitoring constituents and
indicator parameters because potential contaminants and their likely reaction and
degradation products can be more easily identified.
Facility Description
The 600-acre facility is a permitted waste disposal site operated since 1980.
Solid waste management units occupy 20 acres of the site and include four surface
impoundments and one container storage area subject to RCRA. Until 1985, three
units (two surface impoundments and one solids disposal unit) not subject to RCRA
were used for geothermal waste disposal. However, the two surface impoundments
15-50
-------
were replaced by a RCRA regulated landfill. RCRA wastes managed by the facility
include metals, petroleum refining wastes, spent non-halogenated solvents,
electroplating wastewater treatment sludge, spent pickle liquor from steel finishing
operations, and ignitable, corrosive, and reactive wastes. Ground-water monitoring
wells have been installed downgradient of each waste mangement unit.
Program Design
Prior to disposal, each load of waste received is analyzed in an on-site
laboratory to provide a complete characterization of waste constituents. Periodic
sampling of the waste management units is also conducted to identify waste
reaction products and hazardous mixtures. Even though the incoming wastes have
been characterized, the facility owner also analyzed initial ground-water samples
from each monitoring well for all Appendix VIII constituents. The resulting data
were used to establish existing concentrations for each constituent and to select a
set of monitoring constituents and indicator parameters to identify migration of
waste to the ground-water system. Table 15-7 includes a list of the indicator
parameters analyzed at the facility. Rationale for indicator parameter selection are
included in this table. A separate list of hazardous constituents to be monitored
was also developed based on the waste analysis.
Because the facility accepts only a limited number of 40 CFR Part 261 Appendix
VIII constituents and initial monitoring verified the absence of many constituents,
the facility owner or operator was able to minimize the total number of
constituents monitored in ground water. The process of constituent elimination is
dependent on the actual wastes received by the facility and the physical and
chemical properties of these constituents that influence their migration potential
(e.g., octanol/water partition coefficients, solubility, adsorptivity, susceptiblity to
biodegradation).
Non-halogenated solvents have relatively low partition coefficients
(Kow: benzene « 100; toluene = 500) and are not readily retained by soils.
Conversely, polycyclic aromatic hydrocarbons, constituents of petrochemical wastes,
have very high partition coefficients (e.g., chrysene a 4x10s) and are generally
immobile in soils. Migration rates of metals are also influenced by the exchange
15-51
-------
TABLE 15-7
INDICATOR PARAMETERS
Parameter
Total Organic Carbon (TOO
Total Petroleum Hydrocarbons
Total Organic Halogen (TOX)
Nitrates
Chloride
Sulfides
PH
Total phenols
Criteria for Selection
Collective measure of organic substances present
Indication of petroleum waste products
Halogenated organic compounds are generally
toxic, refractory, and mobile
Mobile contaminant, degradation product of
nitrogen compounds
Plating solution constituent, highly mobile in
ground water. Early indicator of plume arrival
Toxic, biodegradation by product, strong
reducing agent, may immobilize heavy metals
Good indicator of strongly acidic or alkaline waste
leachates close to sources
Collective measure of compounds likely to be in
waste. Even small concentrations can cause
olfactory problems following water treatment by
chlorination
15-52
-------
capacity of the soil. Different metal species are sorbed to different extents.
Following an assessment of the migration potential of each waste constituent, the
need for analysis of that constituent can be prioritized.
Case Discussion
Waste stream information was used to determine appropriate monitoring
constituents and indicator parameters. The use of the existing initial ground-water
quality data and the incoming waste analyses allowed for prediction of
contaminants of concern in ground water and reduced the number of constituents
requiring analysis.
15-53
-------
CASE STUDY 10: USING WASTE REACTION PRODUCTS TO DETERMINE AN
APPROPRIATE MONITORING SCHEME
Point Illustrated
• It is important to consider possible waste reaction products when
developing monitoring procedures.
Introduction
Volatile organic priority pollutants have been detected in ground water at
various areas across the country. These compounds, widely used as solvents, are
generally considered environmentally mobile and persistent. Increasing evidence,
however, indicates that chlorinated solvents can be degraded under anaerobic
conditions by reductive dehydrochlorination. The sequential removal of chlorine
atoms from halogenated 1 and 2 carbon aliphatic compounds results in formation
of other volatile priority pollutants which can be detected during investigations of
solvent contamination.
Facility Description
The facility is a small municipal landfill sited on a former sand and gravel
quarry. In addition to municipal wastes, the landfill accepted trichloroethane and
tetrachloroethene contaminated sludge from a local fabrication plant until 1975. In
1983, a municipal well located downgradient of the landfill tested positive for
dichloroethane, dichloroethene isomers, and vinyl chloride. This prompted the city
to investigatt the cause and extent of the problem.
Site Investigation
According to records kept at the facility, some of the compounds found in the
municipal well were not managed at the facility. This prompted the city to request
that a monitoring program be developed to determine whether another source was
causing well contamination. A careful search of the city records, however, failed to
indicate a credible alternative source of the compounds. Suspecting that the
landfill was the source of the well contaminants, five monitoring wells were
15-54
-------
installed (Figure 15-15) and water samples were analyzed for halogenated
compounds using EPA Method 601. The results, given in Table 15-8, show an
increase in degradation products of trichloroethane and tetrchloroethene with
increasing distance from the landfill. Using these data, supported by hydrogeologic
data from the monitoring wells, the municipal landfill was shown to be the source
of the observed contamination.
TABLE 15-8
RESULTS OF MONITORING WELL SAMPLING
Chlorinated Ethanes
(l)Trichloroethanes
(2) 1,1-Dichloroethane
1,2-Dichloroethane
Chloroethane
Chlorinated Ethenes
(l)Tetrachloroethene
Trichloroethene
(2) 1,2-Dichloroethenes
1,1-Dichloroethene
Vinyl Chloride
WELL NUMBER (See Figure 15-15 for
well locations)
1
10(3)
71
ND
NO
80
12
ND
ND
ND
2
68
240
12
21
13
100
990
ND
120
3
ND(4)
130
?1
18
ND
62
950
ND
59
4
ND
11 '
ND
160
ND
ND
150
ND
100
5
ND
13
ND
ND
ND
ND
ND
ND
ND
(1) Parent Compounds
(2) Degradation Products
(3) All Concentrations In Micrograms/L
(4) ND means < 10 Micrograms/L
Case Discussion
Based on the compounds found in the municipal well, the city believed that
the municipal landfill could not be the source of the contamination. If this
reasoning had been followed, then a system of monitoring wells might have been
needlessly installed elsewhere in the attempt to find an alternate source of the
15-55
-------
.„._?T3perty Line
Approxim*c« Scale L"»!00
Direction
Of
Ground
•low
MOTE: Locations of ntarby industrial
facilities noc shown.
Figure 15-15. Site Map and Monitoring Well Locations
15-56
-------
contamination. Instead, after carefully researching local industries, it was
determined that the landfill was the most reasonable source of the contamination
and that the observed well contaminants were probably degradation products of
the landfilled solvents. The progressive dehalogenation of chlorinated ethanes and
ethenes, as shown in Table 15-8, is commonly encountered in situations where
chlorinated solvents are subjected to anaerobic conditions (Wood, 1981). Different
degradation reactions may occur when pesticides are subjected to acidic or alkaline
conditions or biological degradation. Therefore, it is important to keep reaction
products in mind when designing any monitoring scheme or interpreting
contaminauon data.
Reference
Wood, P.R., R.F. Lang, I.L Payan.and J. DeMarco. 1981. Anaerobic Transformation.
Transport and Removal of Volatile Chlorinated Orqanics in Ground Water First
International Conference on Ground Water Quality Research, October 7-10, 1981,
Houston, Texas.
15-57
-------
CASE STUDY 11: CORRECTIVE MEASURES STUDY AND THE IMPLEMENTATION
OF INTERIM MEASURES
Points Illustrated
• Interim corrective measures may be necessary to protect human health or
the environment.
• The evaluation of the need for definitive corrective measures.
Introduction
The development and implementation of a comprehensive Corrective
Measures Study can be a time-consuming process. Between the time of the
identification of a contaminant release and the completion of definitive corrective
measures, existing conditions or contaminant migration can endanger human
health or the environment. Under such conditions interim -measures may be
necessary. The case study presented below illustrates the implementation of
interim measures to reduce contaminant migration and to remove the imminent
threat to the nearby population from exposure to contaminants in drinking water,
and also illustrates the decision- making process as to whether definitive corrective
measures may be necessary.
Facility Description
The facility in this case study is an underground tank farm located at a
pharmaceutical manufacturing plant. The tank farm encompasses an area
approximately 140 feet by 260 feet and contains 30 tanks ranging in size from
12,000 to 20,OtO gallons. The tanks are used to store both wastes and raw materials
for the various batch manufacturing processes performed at the plant. Typical
wastes include carbon tetrachloride, acetonitrile and chloroform. At the time of the
release, the tank farm had no cap to prevent the infiltration of rainfall or runoff. It
also did not have berms to provide containment for surface spills. No leak detection
or leachate collection systems were present.
15-58
-------
Geological and Hydroloqical Setting
The site is underlain by silty soil overlying limestone. The weathered limestone
beneath the site is very permeable (up to 210 ft/day) due to the solution of rock
along joints and bedding planes in the limestone. Depth to the limestone varies
from 3 to 80 feet beneath the tanks and from 15 to 190 feet downgradient of the
site.
The ground-water system beneath the site consists of two aquifers. The upper
one, an unconfined limestone aquifer, is about 300 feet below the surface. The
deep aquifer is an artesian aquifer in another limestone formation about 1200 feet
below the land surface. Ground-water flow in the upper aquifer is controlled by
both the regional flow system and local channelized flow through solution
conduits. The upper aquifer discharges to a canal 3 miles north of the site. Figure
15-16 shows the ground-water elevation contours in the vicinity of the site.
Regional average ground-water flow velocity was estimated at 4 ft/day, but ground-
water velocities on the order of 50 ft/day have been measured in some channelized
areas. Channelized flow is also responsible for local deviations in flow direction.
Release Characterization
A contaminant release from the tank farm was discovered when one of the
tanks used for waste storage was found to be empty. The waste stored in the tank
was predominately carbon tetrachloride (CCU) (a carcinogen with an MCL of 0.005
mg/l, with some acetonitrile (a systemic toxicant with a water-based health criterion
of 200 ug/l) and chloroform (a systemic toxicant with a water-based health criterion
of 400 ug/0 reference dose (RfD) is 0.4 mg/l). Approximately 15,000 gallons of waste
liquids had been routed to the tank before the leak was discovered. Excavation of
the tank revealed ruptures in at least three locations. Initial ground-water
monitoring after the tank rupture was" discovered identified CCU in a well 2500 feet
downgradient of the site, at concentrations above the MCL for CCU of 0.005 mg/l.
Contaminants from the leaking tank were found to have dispersed laterally
within a two-foot-thick sand bed which underlies the tanks. The contaminated area
15-59
-------
®MW22
SMW20
313
Elevation aoove
Mean Sea Level
' Grounowattr
Flow Lines
Groundwattf
Monitoring Well
(ana El«v m Ft )
Contour* BiMd
taken on 9/2/84
(Contour interval 0 2 Ft )
-co:
<§MW1
Water ueveta
Figure 15-16. Ground-Water Elevations and Flow Directions in Upper Limestone
Aquifer
15-60
-------
was approximately 5600 ft2. High levels of CCU were found throughout the sand
layer. Concentrations of CCU in the natural soil ranged between undetected and
2200 mg/kg. Observed concentrations were well above the soil RSD for CCU (2.7
mg/kg). Concentrations generally decreased with depth due to adsorption onto the
clay particles in the soil. Carbon tetrachloride apparently moved downward with
Httle lateral dispersion until reaching the soil-limestone interface. Upon reaching
the unsaturated limestone, the contaminants then appeared to have rapidly
dispersed over an area of about 12 acres before entering the aquifer.
Interim Corrective Measures
Immediate action to contain the release in the aquifer was taken. This
involved pumping the well where CCU had been found continuously at its full
capacity of 450 gpm.
All drinking water in the vicinity of the release was obtained from wells
installed in either the shallow or artesian aquifer. Immediately after the detection
of the release, all domestic and industrial wells north of the facility were tested for
CCU contamination. Test results showed contamination of several shallow water
supply wells. Based on this information and the inferred ground-water flow
direction to the north-northeast, wells serving two small communities and a nearby
motel were closed. The facility operator hired all mobile water tanks available and
supplied water for immediate needs until a temporary water supply could be
implemented. Water from an unaffected artesian well was then used to supply
water to these communities.
The design and operation of the tank farm was altered in an attempt to avoid
similar problems in the future. A fiber-reinforced concrete cap was installed over
the tank farm to prevent the infiltration of rainfall and runoff, thus minimizing
further contaminant migration in the soil. The ruptures were repaired, and a tank
monitoring system was also developed and implemented at the site.
Definitive Corrective Measures: Saturated and Unsaturated Zones
A comparison of CCU concentrations within the ground water to the MCL for
CCU (O.OOSmg/l) indicated that definitive corrective measures may be necessary.
15-61
-------
Due to the high mobility of CCU within the unsaturated zone, and the potential for
continued inter-media transfer from this zone to the ground water, definitive
corrective measures for both the saturated (ground water) and unsaturated zones
should be evaluated in a Corrective Measures Study (CMS).
Case Discussion
The development and implementation of definitive corrective measures at a
site may take a substantial length of time. Depending on the nature and severity of
the release and the proximity of receptors, interim measures, such as alternative
water supplies, were required to minimize the effects on human health and the
environment. Comparison of constituent concentrations with health and
environmental criteria indicated that definitive corrective measures may be
necessary and that a Corrective Measures Study (CMS) should be initiated.
15-62
-------
CASE STUDY 12: USE OF AERIAL PHOTOGRAPHY TO IDENTIFY CHANGES IN
TOPOGRAPHY INDICATING WASTE MIGRATION ROUTES
Points Illustrated
• Aerial photographs can be used to obtain valuable data on facility-
related topographic features, including type of waste management
activity, distance to residences and surface waters, adjacent land use, and
drainage characteristics.
• Detailed interpretation of aerial photographs can identify actual and
potential waste migration routes and areas requiring corrective action.
Introduction
Stereoscopic pairs of historical and current aerial photographs were used to
assist in the analysis of waste management practices at a land disposal facility.
Stereo viewing enhances the interpretation of aerial rvhotographs because vertical
as well as horizontal spatial relationships can be observed, and because the
increased vertical resolution aids in distinguishing various shapes, tones, textures,
and colors within the study area. Typical items that should be noted include pools
of unknown liquid that may have been released from buried materials which could
migrate off site through drainage channels. Soil discoloration, vegetation damage,
or enhanced vegetation growth can also be indicative of contaminant migration.
Facility Description
The site contains an active land disposal facility which receives bulk hazardous
waste, including sludges and contaminated soil for burial, and liquid wastes for
disposal into solar evaporation surface impoundments. Operations at the facility
began in 1969. Historical and current aerial photographs were reviewed to assess
waste management practices and to identify potential contaminant migration
pathways requiring further investigation and corrective action.
15-63
-------
Data Collection and Anavsis
Low altitude color aerial photographs of the facility (scale = 1:8400) were
obtained in October 1983 and Feburary 1984. The photos were interpreted by an
aerial photo analyst at the U.S. EPA Environmental Monitoring and Support
Laboratory at Las Vegas, Nevada. Figure 15-17showstheanalyzed photograph. The
interpretation code is given in Figure 15-18. Analysis of the photograph indicates
several areas of seepage at the base of the surface impoundments. This seepage
indicates that either the impoundments are not lined or the liners have failed.
Drainage from the western portion of the facility which contains most of the
impoundments flows into a drainage reservoir formed by a dam across the main
drainage. Drainage from the northeast portion of the facility where seepage was
also observed appears to bypass this reservoir and enter the main drainage which
flows offsite. Besides possible surface contamination, this seepage also indicates
potential subsurface contamination.
The aerial photograph obtained in February 1984 (Figure 15-19) indicates the
continued existence of seepage from the surface impoundments. There is evidence
of possible discharge from the drainage reservoir to a stream channel, as a pump
and piping were observed. Additional material in the solid waste disposal area has
altered the drainage pattern. At the south end of this area, seepage is evident in
association with damaged vegetation. Drainage from this area enters a drainage
system and appears to be diverted offsite.
Case Discussion
Analysis of aerial photographs of the land disposal facility enabled
investigators to identify potential contaminant sources and migration pathways.
This information was used by investigators to identify areas for surface water,
sediment, soil, and subsurface sampling.
15-64
-------
'
-------
INTERPRETATION CODE
BOUNOAHI6S AND LIMITS
!_•_!.. 'iNcn an IOUNOAHV
—~ UNMNCID art MUNOAMV
I I I I I »f NC(
_-_ MOPtMTV UNI
^ /— GAri/Accra MINT
-|— SICT10M COHNfH
DRAINAGE
••— — - OMAINACI
« • 'LOW OlNICTION
•••-.••• INO(TfH«lMNAT|
TBAMSPOHTATIQN/UTlLlTV
SXS33 vlHICLf ACCIM
art
IMHIMttlH OIKK
ITANOIMO LIQUID
«, STANDING UOUlO
iSMAkU
OIXCAVATION. *«T
KXTINIIVII
MouMon
MOUNOIO MATIMIAk
(SMAkU
Cfi
OM OHUMB
HT MOMnNTAk rA«M
»r MUSUIIf TAM«
VT VINTICAk TADM
CA CUAHfO AIIIA
OG OISTUMIO aHOUNO
IM
to
00 OHM OUM»
SO SLUOGI
ST STAIN
SW 1OLIO MAtTt
T« TMINCM
«0 NAfTt OIVOSAk A«|A
Figure 15-18. Aerial Photograph Interpretation Code
15-66
-------
EROSION!
I OF ROAD
I AERIAL
SURVEY
MARKER!
TANK TRUCK]
UNLOADING gDG]
r\
I TANK
"<.Z^*
I AERIAL
I SURVEY
I MARKER!
tite
*"***' '
I DAMAGED
(VEGETATION!
.UIH l-
[RESERVOIR"
AERIAL
SURVEY
MARKER!
[DISCHARGE
INTO STREAM|
CHANNEL
Figure 15-19. February 1984 Aerial Photograph of Land Disposal Facility
15-67
-------
CASE STUDY 13: IDENTIFICATION OF A GROUND-WATER CONTAMINANT
PLUME USING INFRARED AERIAL PHOTOGRAPHY
Point Illustrated
• Infrared photography can assist in identifying contaminant plumes and
in locating monitoring wells by showing areas of stressed vegetation and
contaminated surface water.
Introduction
Infrared aerial photography can assist .1 identifying contaminant plumes at
sites where little or no monitoring has been conducted. By identifying areas of
stressed vegetation or contaminated surface water, it may be possible to focus on
contaminant discharge points and roughly define the extent of a release.
Hydrogeologic investigations and surface water sampling can then be performed to
further characterize the release. Infrared photography offers the potential to
increase the efficiency of a sampling program.
Facility Description
The facility is a municipal solid waste landfill which has served a population of
22,000 for 30 years. The facility covers an area of 11 acres, holding an estimated
300,000 tons of refuse. The majority of waste in the landfill was generated by the
textile industry. Until July 1978, the facility was operated as an open dump with
sporadic management. City officials indicated that original disposal occurred in
open trenches with little soil cover. After July 1978, the facility was converted to a
well-operated sanitary landfill. Figure 15-20 shows the facility.
Geologic Setting-
The landfill is located on a sandy to silty till varying in thickness from 23 feet at
the hill crest to 10 feet on the side slope. A swamp is present at the base of the hill
at about 255 feet above sea level. There is a dam at the southern drainage outlet
15-68
-------
o BACKGROUND WELL a
METAL
BACKGROUND WELL
N
TREE
KILL
AND >
STRESS (
SCALE (APPROXIMATE)
0 332' 564'
SWAMP
TREE LINE
• WATER
O WELL LOCATION
Q VEGETATION SAMPLING
STREAM
« STREAM SAMPLING POINT
• HOUSING
Figure 15-20. Facility Plan View
15-69
-------
of the swamp, a distance of 2,500 feet from the landfill. Ground water is
approximately 20 feet below the surface at the crest of the hill, while on the slope it
is approximately 6 feet below the surface. The swamp at the foot of the hill is the
surface expression of the ground water (Figure 15-21).
Aerial Photography and Sampling Program
Figure 15-22 shows the infrared aerial image of the site. The landfill
corresponds to the light area in the northwest portion of the photograph (Figure
15-21). The dark area to the south of the site is stressed vegetation, and the light
area within it is contaminated swamp water. The 33-acre area of tree kill and stress
is clearly visible in the original photograph. Plants under stress may be detected by
infrared photography because of changes in infrared reflectance.
Ground-water monitoring wells and vegetation sampling points are shown in
Figure 15-20. Data collected from the wells indicated elevated levels of chromium,
manganese, iron, and total organic carbon (TOC). Table 15-9 lists the average
concentrations of the parameters tested. The vegetation study indicated an
accumulation of heavy metals.
Case Discussion
The vegetative stress apparent in the infrared photography was confirmed by
the data from the ground water and vegetation sampling. However, the site
requires further characterization to determine the vertical extent of contamination
and to assess the potential for impact beyond the present area of stressed
vegetation.
It should be emphasized that infrared photography is not a substitute for
hydrogeologic characterization. However, it is a useful tool for identifying areas of
stressed vegetation that may be associated with releases from waste disposal sites.
15-70
-------
c
o
1/1
o
k_
u
u
CTi
73
0)
N
"2
oJ
c
in
a;
3
o
LU
i
o
X
o
ui
o
V)
2
z
o
N
E
O
u
15-71
-------
15-72
-------
UJ
_J
a.
UJ
OC
(/>
Q
Z
QC
Ul
I
S
QC
UJ
r-
ui
2
oe
u.
O
UJ
O
ce
UJ
I
a
u
wo
a
1—
u
a
00
u
Q
I/I
c
•a
ITJ
C
2
a»
0.
O
*~
II
o
z
X
z
z
*
H>
u
o
c?
o
00
^
m
o
o
00
-.
IN
0
IN
in
CM
00
V
o
rs
(N
no
Ps
«-
-^
^
•s.
O
V
3
0
w
^
U
00
W
IN
—
in
00
00
no
O
o
o
o
0
no
PS
u>
to
in
(N
in
rst
*
00
00
0
^"
•—
o
00
0
s
0
-
*
•"
0
(N
V
CM
=%
^^
1
in
\f\
,
j
o
no
(N
(N
O
o
o
in
T
in
no
o
IN
i
no
>O
to
Cft
m
in
^
o
o
o
c*
no
in
in
VO
in
0
(N
nr>
:fe
^^
1
V
in
-.
-s
o
o
o
00
_
rt
a»
N
nn
^
O
O
o
V
o
o
V
in
IN
§
m
^
f-f
%
1
00
in
"
o
no
VO
00
no
O
o
o
IN
(S.
•^
***
in
no
00
m
VO
_
^
00
o
in
•—
o
CM
CM
*—
VO
o
in
in
*
0
CM
in
:fe
I
rs.
in
-«
%
PS
o
o
o
no
O
IN
cn
in
IO
in
to
PS
PS
in
—
*-
o
V
PS
(N
OO
in
in
*
2
10
_
1
00
iO
-s
£
00
rs
0
o
8
no
(S,
to
^
0
no
IN
no
PS
^
^
00
*""
o
o
in
o
o
PS,
9^
»r
Ps
no
in
CM
PS
O
CM
V
b
10
*
c
•a
a
S
E
^ (0
•§<
1 =
8|
•M 7
15-73
-------
CASE STUDY 14: USE OF HISTORICAL AERIAL PHOTOGRAPHS AND FACILITY
.MAPS TO IDENTIFY OLD WASTE DISPOSAL AREAS AND
GROUND-WATER FLOW PATHS
Points Illustrated
• Aerial photographs taken over many years in the life of a facility can be
used to locate old solid waste management units (SWMUs).
• Historical aerial photographs can be used to identify
geologic/topographic features that may affect ground-water flow paths.
Introduction
In gathering information pertaining to investigation of a release, historical
aerial photographs and facility maps can be examined and compared to current
aerial photographs and facility maps. Aerial photographs can be viewed as stereo
pairs or individually. Stereo viewing, however, enhances the interpretation because
vertical as well as horizontal spatial relationships can be observed. The vertical
perspective aids in distinguishing various shapes, tones, textures, and colors within
the study area.
Aerial photographs and facility maps can be used for the following:
• Providing evidence of possible buried drums. Historical photographs can
show drums disposed of in certain areas where later photographs show
no indications of such drums, but may show that the ground has been
covered with fill material.
• Showing previous areal extent of landfill or waste management area.
Earlier photographs might show a much larger waste management area
than later photographs.
• Showing areas that were dry but now are wet, or vice versa, indicating a
possible release from an old waste management area.
15-74
-------
• Showing changes in land use patterns (e.g., a landfill m an early
photograph could now be a park or be covered by buildings).
• Soil discoloration, vegetation damage, or enhanced vegetative growth
can sometimes be detected, indicating possible contaminant migration.
• Geologic/hydrologic information, such as faults, fracture or joint systems,
old stream courses (channels), and the contact between moraines and
outwash plains.
Facility description
This facility is the same as previously described in Case Studies 7 and 8.
Data collection and analysis
Over the past 50 years aerial photographs were taken of the facility area.
Interpretation of the photographs produced important information that is shown
diagramatically in Figure 15-23. Solid Waste Disposal Area 2 (SWDA-2) was lower in
elevation in 1940 than it is now. In fact, the area appears to have been leveled and
is now covered by vegetation, making it difficult to identify as a SWMU at ground
level. Another area was identified as a possible waste disposal area from a historical
review of photos. Further study of photographs, facility maps and facility files
revealed this to be a former Liquid Waste Disposal Area (LWDA), designated as
LWDA-2 on Figure 15-23.
The use of these historical photographs also revealed geologic features that
could affect the ground-water flow system under the facility. In this case,
monitoring well data indicated a general northwesterly ground-water flow
direction, in addition to a complex flow pattern near LWDA-1 and SWDA-1 (Figure
15-23). Recent photographs were analyzed, but because of construction and other
nearby activities (e.g., cut and fill, sand and gravel mining), conclusions could not be
drawn. A review and analysis of old photographs revealed the existence of a buried
stream channel of the river (Figure 15-23). This buried stream channel was
identified as a preferential path for ground water and consequently contaminant
15-75
-------
•o
O)
c
Ol
Ol
c
c
E
TJ
o; ,/»
C
O
< -M
s 5
^ ±
< .2
Q 5
(fl CT)
—I D
« 2
15-76
-------
migration. Additional monitoring data and further analysis of subsurface geologic
data is needed to determine the full impact of the buried stream channel on the
ground-water flow regime.
Case Discussion
Analysis and interpretation of a series of historical aerial photographs and
facility maps spanning a period of over 50 years enabled facility investigators to
identify the following:
(1) Location of waste disposal areas (e.g., old SWMUs);
(2) Changes in topography (related to earlier disposal activities); and
(3) Possible preferential pathways (e.g., old stream channel) for migration of
ground water and contaminants.
This information was used to identify areas for more detailed sampling and
analysis.
Analysis of historical facility maps and historical aerial photographic
interpretation can be a very powerful tool in a RCRA Facility Investigation, but
should be used in combination with other investigative techniques to result in a
thorough characterization of the nature, extent, and rate of contaminant
migration.
15-77
-------
CASE STUDY 15: USING SOIL CHARACTERISTICS TO ESTIMATE MOBILITY OF
CONTAMINANTS
Point Illustrated
• Information on soil characteristics can be used to estimate the relative
mobility of contaminants in the subsurface environment.
Introduction
The relative mobility of contaminants can be estimated using soil
characteristics and aquifer hydraulic characteristics. Although metals do precipitate
at higher concentrations, at the levels encountered in most subsurface
environments, sorption is the dominant attenuation process. The degree to which a
metal sorbs onto soil particles depends on the soil pH, the percent clay, the percent
soil organic matter, the presence of particular coatings (e.g., iron, manganese, and
aluminum oxide/hydroxides) and, to a lesser extent, the type of clay present. For
organic contaminants, there are several processes which may be important in
predicting their fate in soils. These include sorption, biodegradation, hydrolysis
and, to a lesser extent, volatilization. The sorption of a given organic compound
can be predicted based on its octanol-water partition coefficient, the percent
organic carbon in the soil, and the grain-size distribution of the soil.
Determining the relative mobility of contaminants can be helpful in selecting
appropriate sampling locations. For example, if wastes containing metals were
present in an impoundment, samples to determine the extent of any downgradient
metal contamination would normally be collected within a short distance of the
impoundment. On the other hand, for fairly mobile waste constituents such as
trichloroethylene (TCE), samples could be taken over a much larger downgradient
distance. The case study presented below illustrates how contaminant mobility can
be estimated.
Facility Description
A 17-acre toxic waste dump was operated in a mountain canyon for 16 years.
The facility received over 32 million gallons of spent acids and caustics in liquid
15-78
-------
form. These wastes were placed in evaporation ponds. Other wastes sent to the
facility included solvents and wastes from electroplating operations containing
chromium, lead, mercury and zinc. Pesticides including DDT had been disposed of in
one corner of the site.
Site Description
The site was underlain by alluvium and granitic bedrock (Figure 15-24). The
bedrock, as it was later discovered, was fractured to depths of between 50 and 100
feet Ground water occurred in the alluvial deposits at depths of 10 to 30 feet.
Several springs existed in the upgradient portion of the site. A barrier dam was
built across part of the canyon at the downgradient edge of the site in an effort to
control leakage. Because of the extensive fracture system, this barrier was not
effective. Instead, it appears to have brought the ground-water table up into the
wastes and, at the same time, pressurized the underlying fracture system, thereby
creating seepage of contaminated water under the dam.
Estimation of Contaminant Mobility
Because of the variety of constituents accepted at this site, an estimate of their
relative mobility was needed prior to designing the remedial investigation. The first
step was to estimate the average linear velocity using the following equation :
Ki
V s
He
where
v = horizontal seepage velocity, ft/day
K = hydraulic conductivity, ft/day
i = ground-water gradient
He = effective porosity, decimal fraction.
The hydrogeologic data needed were obtained from existing site assessment
reports. The alluvium underlying the site had an average hydraulic conductivity of
0.8 ft/day and an estimated effective porosity of 11 percent. The average ground-
15-79
-------
K
|
Q
M
a
a
i
u
<
x
*.
VI
O
a.
Q
a;
t«
i
*4*»
O
c
0
-------
water gradient below the site was 0.06. Using the above equation, the average
linear velocity was estimated to be 160 ft/yr. This represents the average velocity at
which a conservative constituent would migrate downgradient along the centerline
of the plume. Examples of such constituents include chloride and bromide. As
shown in Table 15-10, nitrate and sulfate also behave conservatively in many cases.
Due to the absence of highly weathered, sesquioxide soils, sulfate behaved
conservatively at this site. Using the above average linear velocity, an estimate was
made of the distance a conservative solute would travel in a given time (T) using
d = vT. Limited water quality data were available for 1980. Wastes were first
disposed at this site in 1956. The average extent of plume migration along the
centerline was thus estimated to be 3800 feet.
With respect to metals, additional data were needed to estimate their fate
including soil pH, presence of carbonates, organic ligands, and percent soil organic
matter and clay. At this site, the soil pH varied from less than 3.0 within 400 feet of
the acid ponds to 7.2 at a distance 2000 feet downgradient. As shown in Figure
15-25, the partition coefficients for metals are dependent on pH and organic matter
content. For example, below a pH of 5.6, for the types of soil encountered at the
site, the partition coefficient (Kp) for cadmium is about 10 ml/g. At a pH of 7.2, Kp is
about 6500 ml/g (Rai and Zachara, 1985). The relative mobility of attentuated
constituents can be estimated as follows (Mills eta[., 1985):
VA = v/Rd
where
VA = average velocity of attentuated consitutent along centerline
of plume, ft/day
v = average linear velocity as defined above, ft/day
Rd = retardation factor (unitless)
and
Rd = 1 +
15-81
-------
TABLE 15-10
RELATIVE MOBILITY OF SOLUTES
Group
Conservative
Slightly Attenuated
Moderately Attenuated
More Strongly
Attenuated
Examples
TDS
CL"
BR-
NO;
SO42-
B
TCE
Se
As
Benzene
Pb
Hg
Penta-
chlorophenol
Exceptions
Reducing conditions
Reducing conditions
and in highly
weathered soils coated
with sesquioxides
Strongly acidic systems
Anaerobic conditions
Master Variables*
V
v, pH, organic matter
v, organic matter
v ,pH,.Fe hydroxides
V , pH, Fe hydroxides
v, organic matter
v , pH, SCV
v.pH.CI
v , organic matter
Variables which strongly influence the fate of the indicated solute groups.
Based on data from Mills etal., 1985 and Roi and Zachara, 1984.
15-82
-------
100
Percent
Adsorbtion
by Soil
50
Shift due
to presence
of soil organic
matter
Typical
adsorbtion
curve for
heavy metal
x, on a clean
silica or
aluminum
silicate
surface
Typical adsorbtion
curve for heavy
metal x, on silica
or aluminum silicate
surface coated with
soil organic matter
pH of the Soil Solution
a) Generalized Heavy Metal Adsorbtion Curve for Cationic Species (e.g., CuOH*)
100
Percent
Adsorbtion
by Soil
50
Typical adsorbtion \
curve for heavy \
metal species, x, \
on iron hydroxide \
\
\
Shift \
due to \
\ pretence t
\ of soil \
\ organic \
\ matter
\
V . v
pH of the Soil Solution
2-,
b) Generalized Heavy Metal Adsorbtion Curve for Aniotic Species (e.g., CrOj")
Source: (Milh M •!.. 19881.
Figure 15-25. Hypothetical Adsorption Curves for a) Cations and b) Anions
Showing Effect of pH and Organic Matter
15-83
-------
where
Kp = soil-water partition coefficient for solute of concern, ml/g
PB = soil bulk density, g/ml
ne = effective soil porosity (decimal fraction).
For example, the relative mobility of cadmium at a pH of 7.2 was estimated for this
site as shown below:
Rd = 1 + 6500(1.7) = 100,456
0.11
VA = 160/100,000 * 0.002 ft/yr.
This estimate was consistent with the field data which indicated that the metals
migrated only until the pH of the contaminant plume was neutralized, a distance of
about 2000 feet. Cadmium concentrations decreased from 1.3 mg/l at a distance of
1400 feet from the ponds to below detection (<0.1 ug/l) at a distance of 2000 feet.
Estimates of mobility for organic contaminants which sorb onto soil particles
can be made in an analogous manner. The partition coefficient for organic
constitutents can be calculated using the following equation (MillsetaL, 1985):
Kp = K0c[0.2(1-f)X'oc + fXfoc]
where
Kp = soil-water partition coefficient, ml/g
KOC = organic carbon partition coefficient, ml/g
and
Koc s 0.63 Kow
KQW s octanol-water partition coefficient
f = mass of silt and clav (0< f< 1)
mass of silt, day and sand
X'oc a organic fraction of sand (X$oc<. 0.01)
Xfoc s organic fraction of silt-clay (0 <. Xfoc.< 0.1).
15-84
-------
For example, the solvent trichloroethylene (TCE) has a Kow value of 200. Using the
above equation and site data (f = 0.1, X50c = 0.001, Xfoc = 0.01), the partition
coefficient Kp was estimated to be 0.2 ml/g. The relative mobility of TCE at the site
was then estimated to be approximately 40 ft/yr (Rd = 4 and VA = 40 ft/yr).
Methods for considering additional processes influencing the fate of organics (e.g.,
hydrolysis and biodegradation) are presented in the manual entitled Water Quality
Assessment: A Screening Procedure for Toxic and Conventional Pollutants in
Surface and Ground Water (Mills et al.. 1985).
Case Discussion
As shown in Figure 15-26, contaminants downgradient of a waste disposal site
may migrate at different speeds. Using the methods illustrated above, estimates of
the relative mobility of constituents can be made. Such estimates can then be used
to locate downgradient monitoring wells and to assist in the interpretation of field
data.
References
Mills, W.B., D.I. Porcella, M.J. Ungs, S.A. Gherini, K.V. Summers, L Mok, G.L. Rupp,
and G.L Bowie. 1985. Water Quality Assessment: A Screening Procedure for Toxic
and Conventional Pollutants in Surface and Ground Water. EPA/600/6-85/002a. Vol.
I, II and III.
Rai, D. and J.M. Zachara. 1984. Chemical Attenuation Studies: Data Development
and Use. Presented at Second Technology Transfer Seminar: Solute Migration in
Ground Water at Utility Waste disposal Sites. Held in Denver, Colorado, October 24-
25,1985. 63pp.
15-85
-------
\
TDS
FEET
I
0
I
800
Figure 15-26.
Schematic Diagram Showing Plumes of Total Dissolved Solids (TDS),
Total Organic Halogens (TOX) and Heavy Metals Downgradient of
Waste Disposal Site
15-86
-------
CASE STUDY 16: USE OF LEACHING TESTS TO PREDICT POTENTIAL IMPACTS OF
CONTAMINATED SOIL ON GROUND WATER
Point Illustrated
o Soil leaching tests can be used in conjunction with waste and site-specific
factors to predict potential impacts on ground water.
Introduction
Contaminated soil, whether deep, or surficial in nature, has the potential to
impact ground water, primarily through leaching. In many cases, soil
contamination has already lead to contamination of the ground water and
decisions can be made regarding clean-up of the contaminated soil and ground
water based on the constituent concentrations observed in these media. However,
in cases where contaminated soil has not yet resulted in contaminated ground
water, but has some potential to do so, decisions need to be made regarding the
contaminated soil and whether it should be removed or some other action should
be taken because of the soil's potential to contaninate ground water above levels
of concern. This evaluation may be especially critical in those cases where only deep
soils are contaminated, or where constituent concentrations within surficial soils do
not exceed soil ingestion criteria. Both theoretical (mathematical) and physical
(leaching test) models can be used in this evaluation, as well as or in conjunction
with a qualitative evaluation of release and site-specific factors. This case illustrates
the use of leaching tests and consideration of release and site-specific factors to
determine whether contaminated soil has the potential to contaminate ground
water above levels of concern.
Facility Description
The facility is an industrial chemical and solvent facility located on a leased 2.5
acre site within the corporate limits of a major city in the north-central United
States (see Figure 15-27). Periodic overtopping of the surface impoundment, which
is now empty, and suspected contamination of the soil with organic solvents
from the surface impoundment, resulted in an RFI in which the facility was directed
to characterize the nature, extent and rate of release migration.
15-87
-------
i i 5
111 2 M
(9
•M • O
c
o
0)
•o
<
en
O
03
'5
Wl
01
_c
o
.c
v/>
a
-------
Release characterization revealed that the soil surrounding the surface
impoundment, which was mostly fine sand and silt with some clay, was
contaminated with tetrachloroethylene and 1,1,1-trichloroethane at concentrations
ranging from 0.05 to 0.10 and 2 to 20 mg/kg, respectively. This contamination was
observed at depths of up to 5 feet, which was approximately 20 feet above the
water table (i.e., the water table was approximately 25 feet below the land surface).
The soil beneath the site was relatively permeable, with a hydraulic conductivity of
approximately 9x10-4 cm/sec.
Ground-water monitoring conducted during the RFI showed no current
contamination of the ground water, which flows in a northerly direction and
eventually intersects the river (Figure 15-28). The river is used for irrigation and
drinking at downstream locations. Grab samples taken from the river and river
sediments showed no contamination.
The soil in the immediate vicinity of the railroad spur also showed isolated
pockets of mercury contamination, ranging in concentration from 1 to 2 mg/kg, and
to a depth of 1 foot below the land surface. The source of the mercury
contamination could not be determined.
Contamination Evaluation
The relevant health and environmental (HEA) criteria, the constituent
concentrations observed at the site, and selected physical/chemical properties for
the three constituents are shown in Table 15-11. Although comparison of the HEA
criteria for ingestion with the consituent concentrations observed at the site
showed no exceedances, the regulatory agency overseeing the RFI was concerned
that leaching of the contaminated soil could lead to eventual contamination of the
underlying ground water. This concern was based on the relatively high
permeability of the soils beneath the site and the relatively high mobility of the two
organic constituents detected. The facility obtained the regulatory agency's
approval to conduct a leaching evaluation using EPA's Method 1312 (Synthetic Acid
Precipitation Leach Test for Soils).
15-89
-------
s,
Iw
u<
u.
.IS
3
O
•»-•
c
O
u
L.
0)
TJ
C
3
O
(5
en
c
O
>s
oo
(N
IT)
15-90
-------
TABLE 15-11
HEA CRITERIA, CONSTITUENT CONCENTRATIONS AND RELEVANT
PHYSICAL/CHEMICAL PROPERTY DATA FOR CONSTITUENTS OBSERVED AT SITE
Chemical
Tetrachloroethylene
1,1,1-Trichloroethane
Mercury
CAS No.
127-18-4
71-55-6
7439-97-6
HEA
Criteria
(Ingestion)
(mg/kg)
69 '
2,000
-
H20
Sol
(mg/l)
150
1500
~
HEA
Criteria
(Water)
(mg/l)
0.0069
0.2
0.002
Constit.
Cone.
(mg/kg)
0.10
20
2
Koc
(mg/l)
364
152
Low
Log
Kow
2.6
25
-
Det.
Limit*
(mg/l)
0.01
0.01
0.0004
* Detection limits presented are those for water. Detection limits for soil vary greatly, but may
be assumed to be approximately 1 mg/kg.
15-91
-------
Leaching Evaluation.
Prior to collecting samples and applying the leaching test, the facility first
decided to determine if the contaminated soils could possibly result in leaching test
(extract) concentrations that exceed the relevant HEA criteria (See Table 15-11). To
do this, the facility calculated the maximum theoretical extract concentration by
assuming that 100 percent of the constituents would leach from the soil. The
following equation was used:
Maximum Theoretical Concentration of Toxicant
s
Extract Concentration (mg/l) in Soil (mq/kq)
20
where 20 refers to the liquid to solid ratio applied in EPA Method 1312.
Using this simple equation, the facility determined that the maximum leachate
concentration for tetrachloroethylene was, in fact, below the HEA criteria for water
(see Table 15-11), and that the level could not possibly be exceeded even if 100
percent of the contaminant leached from the waste. For 1,1,1-trichloroethane and
mercury, however, it was determined that the HEA criteria level could be reached if
only a portion of the contaminant present leached from the soil, and that
application of the leaching test would be necessary. Using this screening-type
evaluation, the facility was able to reduce the number of constituents that would
need to be analyzed when applying the leaching test, from three to two.
Samples of the contaminated soil were then collected at selected locations
(i.e., those expected to produce the more heavily contaminated samples) and
Method 1312 applied. Total constituent analyses were also conducted in order to
ensure that the samples represented the more heavily contaminated areas of the
site. Analyses of the soils and leaching test extract were conducted for 1,1,1-
trichloroethane and mercury. The results are shown in Table 15-12.
The leaching test results for 1,1,1-trichloroethane and mercury showed extract
concentrations above the respective HEA criteria (action levels) for these
15-92
-------
TABLE 15-12
LEACHING TEST RESULTS (mg/l)*
Constituent
1,1,1-Trichloroethane
Mercury
D-C
0.3
0.002
C-C
0.2
0.002
C-B'
0.5
0.003
* Resampled at locations close to original sampling point. Samples analyzed
are result of composite of three grab-samples. All samples were taken from
the top 0-1 ft of the soil surface.
15-93
-------
constituents, indicating that there might be a basis to require some sort of
corrective action. The facility, however, presented arguments to show that mercury
would be attenuated in the soil column as the leachate percolates towards the
water table, and that 1,1,1-trichloroethane would be degraded to below the level
of concern in the ground water. Below is a synopsis of the two arguments.
Mercury: The facility first examined theoretical Eh-pH fields of stability for the
various aqueous mercury species; determined that the predominant mercury species
would be elemental mercury, and further predicted (using Eh-pH diagrams) that the
maximum equilibrium concentration of elemental mercury in water would be 0.025
mg/l. The facility interpreted the substantially lower leaching test concentration to
indicate that attenuation processes such as sorption play a major role in restricting
the mobility of elemental mercury. The facility cited high soil/water partition
coefficients (i.e., Kd values), and several scientific studies to further support their
contention that mercury would strongly sorb to both organic and inorganic
components of the soil before any leachate reached the ground water.
1.1.1-Trichloroethane: The facility recognized that due to its high solubility
(1500 mg/l) and low Kd (0.011 ml/g), 1,1,1-trichloroethane would not be attenuated
appreciably as the leachate percolates towards the water table. The facility argued,
however, that abiotic hydrolysis would significantly degrade 1,1,1-trichloroethane
during leaching. Several environmental half-life studies were cited which indicated
/
that the half life for 1,1,1-trichloroethane ranged between 0.5 and 2.5 years. Based
on these studies, the facility predicted that 1,1,1-trichloroethane would be
degraded to below levels of concern within one to three years. Usin'g additional site
information and simple time of travel calculations, the facility predicted that
concentration levels for 1,1,1-trichloroethane would be decreased to below the
level of concern well before reaching any potential receptors.
The regulatory agency's evaluation of the facility's arguments is presented
below:
Mercury: The facility's argument with respect to mercury is essentially correct
if it can be assumed or proven that the mercury originally present at the site is
inorganic in nature. If, however, the mercury present is organic in nature (e.g.,
15-94
-------
methyl mercury), the potential for migration of the mercury is substantially
increased. Organic mercury compounds generally have higher volatility, higher
solubility, and much lower Kd values than inorganic mercury compounds. It should
also be noted that even if the original release was of inorganic mercury,
biotransformation (i.e., biomethylation) of elemental mercury may occur. The
facility should be required to determine the actual form(s) of mercury present at the
site.
1.1,1-Trichloroethane: The facility's argument with respect to 1,1,1-
trichloroethane raises many technical questions. For example, the facility uses
published data on the half life of 1,1,1-trichloroethane, which may not be
applicable to the facility's soil and ground-water environment. As another example,
the half-life degradation rate argument may only be applicable for ground-water
transport. The facility does not address degradation in soil or effects on surface
water (assuming that contaminated ground-water will eventually migrate to the
river). Most important, however, is the fact that the facility did not address the
degradation products of 1,1,1-trichloroethane, one of which is 1,1-
dichloroethylene, which is also a hazardous constituent. 1,1,1-trichloroethane
should be assumed to pose a threat to ground water.
Conclusions
The next step in the RFI process would be to determine if interim corrective
measures or a Corrective Measure Study was warranted for the release. Although
none of the soil ingestion HEA criteria were exceeded at the site, application of the
leaching evaluation indicated that 1,1,1-trichloroethane could leach to ground
water and result in exceedance of the HEA criterion for water. On this basis, the
facility should be directed to perform a Corrective Measures Study.
To prevent further contaminant migration, the application of interim
corrective measures may also be considered. Construction of a temporary cap over
the contaminated area is one option. Perhaps a more appropriate measure would
be to remove the contaminated soil. Such an action, taken as an interim corrective
measure, may negate the need for a formal Corrective Measures Study.
15-95
-------
Case Discussion
Leaching tests and similar evaluations (e.g., application of validated
mathematical leaching models) can be used to identify potential problems due to
leaching of contaminated soils. In this case, application of a leaching evaluation
was instrumental in identifying a potential threat to ground water as a result of
leaching of contaminated soil. This finding was particularly significant as HEA
ingestion criteria were not exceeded.
It should be noted, however, that in some cases leaching tests may provide
results that are difficult to interpret. For example, consider what would have
happened if the soil underlying the facility was predominantly clay with a
permeability on the order of 10-8 cm/sec. In this case, demonstrating that leaching
will most likely occur within the forseeable future may be difficult. As another
example, if the soil underlying the facility were predominantly sand, leaching would
be probable. In both these cases, application of a leaching test may not provide any
more useful information than is already available. Careful consideration of release
and site-specific information is always warranted prior to application of leaching
tests.
15-96
-------
CASESTUDY17: USE OF SPLIT-SPOON SAMPLING AND ON-SITE VAPOR
ANALYSIS TO SELECT SOIL SAMPLES AND SCREENED
INTERVALS FOR MONITORING WELLS
Point Illustrated
• HNU and OVA/GC screening can provide a relative measure of
contamination by volatile organics. It can be used to select soil sample
locations and can assist in the selection of screened intervals for
monitoring wells.
Introduction
On-site vapor screening of soil samples during drilling can provide indications
of organic contamination. This information can then be used to identify apparent
hot spots and to select soil samples for detailed chemical analyses. In this manner,
the use of higher powered laboratory methods can be focused in an effective way
on the analysis of samples from critical locations and depths. The vapor analyses o,.
site can also be helpful in selecting screened intervals for monitoring wells.
Facility Description and History
Manufacturing of plastics and numerous other chemicals has occurred at the
site over the past 30 years. Some of the major products included cellulose nitrate,
polyvinyl acetate, phenol, formaldehyde, and polyvinyl chloride. The entire site
covers 1,000 acres. The location of the buildings and waste disposal areas are shown
in Figure 15-29. This is the same facility as used in Case Studies 7, Sand 14.
Three disposal methods are known to have been employed at the site. Readily
combustible materials were incinerated in four burning pits, while non-
combustibles were either disposed of in landfills or in a liquid disposal area. All on-
site disposal operations were terminated in 1970, and monitoring programs have
been implemented to identify contaminants, define and monitor ground-water
contaminant plumes, and assess the resulting environmental impacts.
15-97
-------
8
en
-I
— o
Crt
LU
a ?
CO
51 co
r^
~Si
p—
O
<
**•
u.
LU
LU £
Cfl 2
< CC
CL CD
= o
CO
I
c
o
Oi
0)
1/1
fl
-C
a.
•o
c
TJ
1/1
(TJ
0|
<
v/i
O
a
C
i
a.
0)
CN
u->
O>
3
15-98
-------
Geologic and Hydroloqic Setting
The site is located in a well-defined glacial valley, adjacent to a river. Three
major units underlie the site, consisting primarily of sand and gravel outwash
deposits; fine-grained lacustrine sands; and till. The middle sand unit contains
lenses of silt, clay and till. Only the deep till formation appears to be continuous
across the site. A geologic cross-section beneath two of the disposal areas is shown
in Figure 15-30.
The ground-water flow direction at the site is to the northwest. However,
there appears to be a buried stream channel running across the site which strongly
influences the local ground-water flow regime (see Figure 15-31). Ground water
from the site is thought to discharge to the river. The depth to ground water varies
from 10 to 40 feet.
Sampling Program
As part of the remedial investigation at this site, 33 borings were drilled using
a hollow-stem auger rig. Continuous soil samples were collected using split-spoon
samplers. Samples for laboratory chemical analysis were selected based on the
volatile organic concentrations detected by initial vapor screening of the soil
samples in the field.
This field screening was achieved by placing a portion of each sample core in a
40 ml glass headspace vial. An aliquot of gas was extracted from the vial and
injected directly into a portable OVA gas chromatograph (OVA/GC). The
chromatograph was equipped with a flame ionization detector to identify
hydrocarbons. Each sample was also screened using an HNU photoionization
detector because of its sensitivity to aromatic hydrocarbons, particularly benzene,
toluene and the xylenes. Following completion of drilling, gamma logs were run on
all boreholes.
An-example of the vapor screening results (HNU and OVA/GC) and geological
and gamma logs for one of the boreholes are shown in Figure 15-32. The data
shown demonstrate the differential sensitivity of the HNU and OVA/GC detectors.
Because the OVA/GC is more sensitive to the organics of interest (aliphatics),
15-99
-------
<
VI
g
i S
LLJ UJ tO
O H< ,
UJ COUJ -^
-J
UJ
UJ
i-o
LU
_J
<
8.8
O
N
. ." ' • • •'-'••".' •
-?r:'••''.•:'.'<•• •.'';;'
,v:,•.•.•.;;-.•'.
s
Is f:@^l
13A31 V3S 3AO8V 133d Nl NOI1VA313
15-100
-------
I
O
1 O
•v
— o
LU
(—
z
cr
o
O
00
-Q
E
-------
CTI
C
'E
-------
c
d)
O)
U
O
Z
I
O)
C
o
-C
w
o
03
o f
9> 3
a c
E c
x O
,71 u
CM
m
CT
15-103
-------
these results were used to select samples for detailed chemical analysis in the
laboratory. As shown in Figure 15-32, samples in zones with OVA/GC readings of
365 ppm (45 feet deep), 407 ppm (65 feet deep), and 96 ppm (85 feet deep) were
selected. In the laboratory, samples were first analyzed for total organic carbon
(TOC). The ten samples with the highest TOC levels were then analyzed for
purgeable organics using EPA Method 50-30 and extractable organics using EPA
Method 82-50 (U.S. EPA, 1982 - Test Methods for Evaluating Solid Waste, SW 846).
The OVA/GC results were also used to select well screen intervals. Examination
of the data in Figure 15-32 shows that the highest levels of volatile organics (by
OVA/GC) were found at a depth of 65 feet. In addition, the gamma and geologic
logs indicated that the permeable medium at that depth was coarse sand which
would be a suitable location for the placement of a well screen. Thus, a 5-foot
stainless steel screen was set over the depth interval of 62 to 67 feet.
Case Discussion
This sampling program incorporated field techniques that detect the presence
of volatile organics and allow on-site, rapid identification of likely contaminant
"hot spots" for detailed laboratory anaysis and to select depths for monitoring well
screens.
15-104
-------
CASE STUDY 18: CONDUCTING A PHASED SITE INVESTIGATION
;
Points Illustrated
• When ground-water contamination is known or suspected at a site, a set
of initial borings is typically made to determine site hydrogeologic
characteristics and to identify areas of soil and ground-water
contamination (Phase I).
• These findings are then used to select well locations to fully delineate the
extent of contamination (Phase II).
Introduction
To identify the extent of ground-water contamination in an efficient manner,
information is needed on the ground-water flow regime. Phase I investigations
typically focus on determining site geologic characteristics and ground-water flow
directions and velocities. Waste sources are also identified. The Phase I results are
then used in planning the Phase II investigation to determine the extent of
contamination and to refine estimated rates of contaminant migration.
Facility and Site Description
Descriptions of the facility and site geologic characteristics were included in
Case Studies 7,8,14 and 17.
Sampling Program
The Phase I sampling program included geophysical surveys, water level
monitoring, soil sampling, and ground-water quality sampling. Three seismic
refraction lines were run to estimate the depth to the top of the deep till. The top
of the till was found to occur at a depth of 70 to 120 feet over most of the site.
Available historical data indicated that the general ground-water flow
direction was to the northwest across the site. The ground water was thought to
discharge to the river. This information and historical drawings and maps of known
15-105
-------
disposal areas were used to locate the Phase I borings (see Figure 15-29 in Case
Study 17). One well (MW4) was located on the suspected upgradient side of the
site. The other wells were located near waste sources to determine which sources
appeared to be contributing contaminants to the ground water. For example, two
wells (MW6 and 7) were immediately downgradient of solid waste disposal area #2.
To determine the presence of vertical gradients, three two-well clusters were
drilled-each with one well screened just below the water table and a second well
screened considerably below that at the base of the till.
The results of the Phase I investigation indicated that all the wells contained
solvents. Thus, investigations of the waste sources and contaminant plumes were
continued in Phase II. The highest solvent concentrations were found in wells
located near the liquid waste disposal area where downward vertical gradients
were present. The contaminants had migrated down to depths of 75 feet in this
portion of the site. The Phase I data confirmed the general northwest ground-
water flow direction but showed a complex flow pattern near the buried stream
channel. A second concern was whether observed lenses of fine-grained till under
the site were producing localized saturated zones which could be contaminated.
Based on the Phase I results, a Phase II monitoring program was designed to
determine the extent of contamination around the major disposal sites. Typically,
two soil borings were made - one up- and one downgradient of the waste source.
Because of the high solvent concentrations observed in the wells downgradient of
the liquid disposal area, a more intensive field investigation of this area was
included in Phase II. Instead of two borings per waste source at the liquid disposal
area, 11 soil borings and five new monitoring wells were drilled. This represented
one-third of the total effort for the entire 1,000 acre site. The total number of
Phase II soil borings was 33 (Figure 15-33) and the total number of Phase II wells was
15 (Figure 15-34). The Phase II data indicated that most of the solvent
contamination originated from the liquid disposal area and not from solid waste
disposal area #1 which is upgradient of the liquid disposal area. The Phase II data
did identify PCBs from solid waste disposal area #1 but not from any of the other
sources. This was consistent with site records indicating that transformers had been
disposed at this area.
15-106
-------
-------
en
o
o
g
- O
— O
O
LU
cr
O
Q
LU
1 £ 5 9
cr
a.
O)
c
c
o
-C
Q.
-D
CU
1^1
O
Q.
O
tn
0)
3
15-108
-------
Case Discussion
Investigation of a large complex site is commonly conducted sequentially.
Basic information is needed on site geologic characteristics and ground-water
velocities and directions to appropriately locate wells for determining the extent of
contamination. Thus, the initial installation of a limited number of exploratory
borings and wells can provide the data needed to design a complete and effective
investigation. Results from the latter investigation can then be used to determine
the need for remedial action and to evaluate alternative remediation methods.
15-109
-------
CASE STUDY 19: MONITORING BASEMENT SEEPAGE
Point Illustrated
• Basement monitoring can be used to estimate the extent of contaminant
migration.
Introduction
Leachate produced in a landfill can be transported downgradient in ground
water by advection and dispersion. Shallow ground water may surface and seep
into basements.
Site Description
A channel, originally constructed as part of a hydroelectric power generation
system, was used as a disposal site for a variety of chemical wastes from the 1920s
through the 1950s. More than 21,000 tons of waste were dumped in and around
the site before its closure in 1952. After closure, homes and a school were
constructed on and around the site. In the 1960s, residents began complaining of
odors and residues. During the 1970s, the local water table rose, and contaminated
ground water seeped into nearby basements.
Geologic and Hydroloqic Setting
Figure 15-35 shows a cross-section of the site. The site has both a shallow and
a deep aquifer. The shallow aquifer consists of approximately 5 feet of interbedded
layers of silt and fine sands overlying beds of clay and glacial till. The deeper aquifer
is a fractured dolomite bedrock overlying a relatively impermeable shale. Travel
times from the shallow to the deeper aquifer are relatively long. Contamination has
occurred in the shallow aquifer because of the "bathtub effect". The impermeable
channel filled because of infiltration, and leachate spilled over the channel sides.
The leachate contaminated the shallow ground water and was transported laterally
in this system.
15-110
-------
LU
_J
CD
<
QCT^
^ifl'rt
0?^= =
§2ss§
1 1 1 1 1
o
n
CM
1
f
o
CO
n
E
u
CO
JZ
J-l
(fl
CO
c
o
l/t
vrt
o
_0
o
Lft
PO
I
LT>
0)
15-111
-------
Sampling Program
The houses surrounding the channel were grouped into three sets
(upgradient, downgradient, and on-site) based on preliminary data on the
underlying strata and ground-water flow directions. Four houses from each group
were selected for sampling for a total of 12 houses. Samples of water and
sediments were collected from the sump pump wells in each basement. Water
samples were collected when the sump pumps were running and 24 hours after
pumping had ceased. Water and sediment samples were analyzed for purgeable
and extractable organics. Benzene, carbon tetrachloride, chloroform, and
trichloroethylene (TCE) were found in the water samples. Water samples taken
while the sump pumps were running had higher concentrations of volatile organics.
Sediment samples contained PCBs and dioxin, possibly due to cosolvation.
Relatively immobile organics can become dissolved in another more mobile solvent.
The mobile solvent containing traces of other organics can be adverted along with
the water. This process (cosolvation) is one facet of enhanced transport which has
recently been proposed as a possible mechanism for the observed mobility of
otherwise immobile organics. Samples of water and sediments from storm drains
were also collected and analyzed to determine if discharges from the sumps to the
storm drains were a significant source of organics in the storm runoff.
In addition to determining water quality, indoor and outdoor air quality was
measured in the basements at each house. Tenax and polyurethane foam tubes
were placed in air monitoring systems in each basement to measure 12-hour
average concentrations of volatile organics (e.g., carbon tetrachloride, benzene,
and TCE) and semi-volatile organics (e.g., pesticides). Volatile organics were present
in the indoor air samples but semi-volatile organics were not detected. The highest
volatile organic concentrations were observed when the sump pumps were
operating.
Case Discussion
At sites where hydrogeologic factors favor shallow lateral ground-water flow,
initial site characterization may involve sampling of basements. Results from such
an initial site characterization can provide information on contaminant migration
15-112
-------
which can be used in the design and implementation of detailed soil and ground
water monitoring programs.
The results of the sampling program described above led to the evacuation
and destruction of a number of homes. A system of monitoring wells has been
installed to replace the basement sump sampling sites. The shallow aquifer is being
pumped and treated to arrest contaminant migration.
15-113
-------
CASE STUDY 20: USE OF PREDICTIVE MODELS TO SELECT LOCATIONS FOR
GROUND-WATER MONITORING WELLS
Point Illustrated
• Simple mathematical models can be used to estimate the longitudinal
and transverse spread of a contaminant plume. Wells can then be
located in areas expected to have elevated contaminant concentrations
and in areas thought to be both up-anddowngradientofthe plume.
Introduction
The use of mathematical models to estimate the migration of contaminants
can be helpful for several reasons, including: 1) fewer wells may be needed to
delineate a contaminant plume, and 2) wells can be rationally located in an attempt
to determine the maximum concentrations in a plume, its furthest extent, and
locations where concentrations should be at background levels.
Facility Description
The site was an electronics manufacturing plant that had been in operation for
20 years. Four large diameter, rock-filled "dry wells" had been used to dispose of
solvents and process wastes. These disposal units were between 35 and 60 feet
deep. Depth to ground water was over 460 feet. Disposal Units 1 and 2 had
received paint wastes and solvents, including trichloroethylene (TCE) and
tetrachloroethylene, between 1964 and 1979. Disposal Units 3 and 4 had been used
to dispose of plating solutions and spent acids between 1971 and 1977. These
solutions contained copper, chromium, nickel, lead and tin. All the disposal units
were dosed in 1982. Exact quantities of wastes disposed are not known.
Geologic and Hvdroloqic Setting
The site is located in a large alluvial basin in an arid region. The basin alluvium
is over 1,000 feet thick and consists of an upper sand and gravel unit, a middle silty-
clay unit, and a lower sand and gravel unit. Granitic bedrock underlies the
unconsolidated formations. Prior to large withdrawals of ground water, the upper
15-114
-------
unit had been saturated. At present, the silty-clay unit acts as an aquitard so that
water beneath it is under confined conditions. The potentiometric surface is now
350 feet below the land surface. In addition to a drop in water level elevations, the
ground-water flow direction has changed over the years from east to north in
response to changing pumping regimes. Estimated horizontal flow velocities have
varied from 10 to 40 feet/year.
Site Investigation
In 1982, city water officials discovered TCE in water samples from wells within
3 miles of the site. On its own initiative, the site owner began a pre-remedial
investigation, and then later a remedial investigation, to determine whether his site
could be a source of the TCE. The pre-remedial investigation provides an example
of how simple models can be used to determine well locations. The pre-remedial
investigation included sampling nearby wells and drilling a single deep sampling
well (over 500 feet deep).
Original plans called for locating the deep monitoring well between the waste
disposal units in an attempt to determine whether solutes had contaminated the
underlying ground water. However, site constraints, including an overhead power
transmission line, underground power lines and major manufacturing buildings,
necessitated that the monitoring well site be moved. The next step was to
determine an appropriate location for this well. Because of the changing ground-
water flow direction at this site, it was decided to use a simple mathematical model
to predict the areal extent of contamination from the disposal units. The results
would then be used in selecting a new location for the deep monitoring well. Data
were collected to determine historical hydraulic gradients, pumping histories, and
aquifer hydraulic characteristics (e.g., conductivity, porosity). Following data
collection, a vector analysis model "the method of Mido" (1981) was used to predict
plume evolution. The results showed that the major plume migration was to the
north (Figure 15-36). Thus, the well was located north of the disposal units at a
distance of 60 feet from Unit 4.
15-115
-------
I
0
Final Site of Deep
Monitoring Well
DISPOSAL UNIT #4
DISPOSAL UNIT #3.
Original Planned
Deep Monitoring
Well Location
DISPOSAL UNIT #2-
DISPOSALUNITtV
Scale
100
Feet
BUILDING 2000
BUILDING 1000
Figure 15-36. Estimated Areal Extent of Hypothetical Plumes from Four Wells
15-116
-------
Case Discussion
Use of a model to predict potential plume migration at this site provided a
means of evaluating the long-term consequences of changing ground-water flow
directions and velocities. Thus, the pre-remedial investigation deep monitoring
well could be sited in the direction of net plume displacement, rather than at a
location which might have had a low probability of intercepting contaminated
ground water. A concentration below the detection limits from a well located
beyond the expected plume boundaries would have been inconclusive (for example,
see Figure 15-37). However, the deep monitoring well was located close to the
disposal units and in the direction of plume migration. Additional wells are now
being planned forthe full-scale remedial investigation.
Reference
Mido, K.W. 1981. An economical approach to determining extent of ground water
contamination and formulating a contaminant removal plan. Ground Water,
Vol. 19, No. 1, pp. 41-47.
15-117
-------
WASTE SOURCE
SAMPLING WELL
YEARLY INCREMENTS OF WATER AND
CONSERVATIVE SOLUTE MOVEMENT
Figure 15-37. Consideration of Solute Migration Rates in Siting Sampling Wells.
If a monitoring well is sited farther downgradient than solutes could
have traveled in the time since disposal, low concentrations in the well
would certainly not prove that ground-water contamination had not or
was not occurring. Prior to locating a well, average linear velocities
should be estimated (v = Ki/rie where v = average linear velocity for
conservative solutes, K = hydraulic conductivity, i = ground-water
gradient, and Tie 3 effective porosity). Using these estimates, and the
age of the disposal unit,!, an approximate migration distance, D, can be
computed (D « T/v) for conservative solutes associated with the waste.
For soil interactive solutes, migration distances will be less. Methods for
estimating these distances are given by Mills et at. (1985).
15-118
-------
CASE STUDY 21: MONITORING AND CHARACTERIZING GROUND-WATER
CONTAMINATION WHEN TWO LIQUID PHASES ARE PRESENT
Point Illustrated
• Monitoring and characterizing ground-water contamination when two
or more liquid phases are present requires knowledge of the physical and
chemical properties of each phase.
Introduction
Ground-water supplies are susceptible to contamination by immiscible organic
liquids. Organic liquids such as PCB-contaminated transformer oils, petrochemical
solvents, and motor fuels, because of their nature, often form a second liquid phase.
This separate liquid, in either the vadose or saturated zone, represents a problem in
multiphase flow. It is necessary to understand how these separate phases behave
when designing monitoring and sampling programs for sites contaminated with
such liquids. Techniques commonly used for single-phase flow systems may not be
appropriate.
»
Site Description
The facility is a transformer manufacturing plant which experienced a major
discharge of polychlorinated biphenyls (PCBs) and trichlorinated benzenes (TCBs).
The discharge resulted from a break in a buried pipeline, but surface spillage may
have also occurred during production. The volume and duration of the subsurface
discharge is not known; neither is the quantity released by above ground spillage.
Geological and Hvdroloqic Setting
The site is comprised of 10 feet of fill over lacustrine clay which varies in
thickness from 20 to 30 feet. Fractures with openings of approximately 0.1 cm have
been observed in the clay. Below the clay lies a thin silt layer. Below that is a 40- to
60-foot-thick layer of glacial till composed of fine sand near the top, and gravel,
sand, and silt below.
15-119
-------
Perched water about 3 feet deep flows laterally in the fill. The permanent
water table, located in the till, is partially confined. Potentiometric levels in this
latter system are between 25 and 30 feet below the land surface.
Sampling Program
Over 1000 soil samples were taken as part of the site investigation. A mobile
atmospheric pressure chemical ionization mass spectrophotometer (APCI/MS) was
employed for rapid, on-site characterization of soil samples. This instrument can
detect PCBs down to a minimum concentration of 100 mg/kg. About 20 percent of
the PCB analyses were replicated by conventional gas chromatography.
Granular dry materials were sampled from an auger with care taken in
cleaning sampling equipment to avoid cross-contamination. In taking samples from
the clay, special effort was made to sample the surfaces of obvious fractures. This
was done to maximize the changes of detection of PCBs in largely uncontaminated
soil. Due to dilution, large bulk samples can prevent the detection of contaminant
migration through fractures in low permeability soils.
Vertically, the soil sampling program showed PCBs to be distributed in a non-
homogeneous pattern within the clay zone. Concentrations of PCBs greater than
500 mg/kg PCBs were detected. The lateral spreading of PCBs throughout the fill
was much more extensive than the vertical movement. This could be due to the
nature of the discharge/spillage, pressure from the broken pipe, or the fact that the
fill is more permeable than the clay. The PCBs appear to have formed a layer along
the fill/clay interface. Movement of PCBs more than 300 feet laterally from the
original spill site has been confirmed.
Based on the soil sampling results, 12 well locations (Figure 15-38) were chosen
to further characterize the site. Four boreholes were drilled into the till aquifer.
One well, 686-B, was placed upgradient of the spill site with a screened interval
between depths of 45 and 50 feet. The three downgradient wells in the till aquifer
were screened over different intervals to increase the possibility of detecting a
separate organic liquid layer. The screened intervals used were at depths 45 to 50
feet (well 686-A), 50 to 55 feet (well 686-C), and 55 to 60 feet (well 686-D). Eight
15-120
-------
03
06
• • 6S6.D
686-A Oj
PARKING LOT
pipeline
31
MANUFACTURING
• 686-B
direction of
ground water
flow
t
N
• deep well locations
o shallow well locations
Figure 15-38. Well Locations and Plant Configuration
15-121
-------
shallow wells were also placed in the fill to monitor the perched water. The fill is
approximately 10 feet deep and a layer of PCBs was suspected at the fill/clay
interface. The depth of the perched water fluctuates between 7 and 8 feet. Six of
the eight wells in the fill, 1, 3,4, 6, 7, and 8, are screened from 7 to 10 feet. Samples
from wells 1, 6, 4, and 7 showed PCB levels much higher than the solubility limits.
The sampling results suggest that two separate liquid layers exist at these locations
and that the liquids are being mixed during sampling. Wells 2 and 5 were screened
from 5 to 8 feet to determine if a floating liquid layer was present. Again, samples
having concentrations far in excess of solubility limits indicated the existence of a
layer of organic liquid.
Case Discussion
Ground-water systems contaminated with immiscible liquids require special
attention. Well screen intervals should be placed to intercept flow along
boundaries between soil layers of differing hydraulic conductivities and at water
table surfaces. Sampling results must also be interpreted properly. Samples
showing contaminant concentrations far in excess of solubility limits may indicate
that two layers of different liquids are being pumped and mixed.
Finally, Figure 15-39 is offered as an illustration of the types of complexity
which can be encountered with immiscible liquids having densities both greater
than and less than water.
15-122
-------
' 'i GflOUNO WATER FLOW
Figure 15-39. Behavior of Immiscible Liquids of Different Densities in a Complex
Ground-Water Flow Regime
15-123
-------
CASE STUDY 22: METHODOLOGY FOR CONSTRUCTION OF VERTICAL FLOW
. NETS
Point Illustrated
• Construction of a vertical ground-water flow net can be a valuable tool
for evaluating ground-water (and contaminant) pathways and for
determining additional actions that may be necessary to accurately
delineate the ground-water flow regime at a facility.
Introduction
Constructing a vertical flow net at a facility provides a systematic process for
analyzing the accuracy of ground-water elevation and flow data, and can therefore
foster a better understanding of the ground-water flow regime at the site.
Facility Description and History
The site contains a large chemical manufacturing facility of approximatley 300
acres located beside a major river in the northeastern United States. The site has
been used for chemical manufacturing by different companies since 1904 and has a
long history of on-site waste management. Several solid waste management units
have been identified at the facility. This is the same facility as discussed in Case
Studies 7, 8,14,17 and 18.
Geologic and Hydrologic Setting: At depths of 150 to 200 feet the site is
underlain by bedrock identified as arkosic sandstone. Above this bedrock are glacial
deposits consisting of a thick bed of hard till, overlain by lacustrine sediments and
deltaic and outwash deposits. Discontinuous lenses of till were identified within the
deltaic deposits. A trough cut into the thick-bedded till and trending approximately
southeast to northwest has been identified. See Figure 15-40.
The river beside the facility flows westward and discharges into the main stem
of a larger river approximately 4 miles west of the facility. A small tributary (brook)
borders the facility to the southwest and west. Swamp-like areas are present near
15-124
-------
15-125
-------
the tributary. It is suspected that the arkosic sandstone outcrops in the river
adjacent to the facility. Whether this visible rock is a large glacial erratic or an
outcrop of the arkosic sandstone bedrock is an issue identified during previous
investigations and may be important in characterizing the ground-water flow
regime at the facility.
Program Design
The site was investigated in two phases. Phase I (1981-1984) included the
installation and monitoring of wells MW-1 through MW-12, while Phase II (1984-
1985) consisted of 34 soil borings, installation of wells MW-13 through MW-57, and
monitoring and sampling of all wells. This two-phased approach allowed the use of
the initial monitoring well data and soil boring data to determine the placement of
the Phase II monitoring wells. Further discussion of this two-phased approach is
provided in Case Studies 7 and 18.
Data Analysis
Evaluation of the data was conducted based on information provided by the
owner or operator, including the water-level elevation data presented in Table
15-13. Well locations and water-level elevations in the wells were mapped and
compared to elevations of the midpoint of the well screens to show relative
hydraulic head differences from well to well. Vertical gradients are a reflection of
different head values at different elevations. For each well, the head can be
determined at the elevation of the midpoint of the well screen by measuring the
water-level elevation in the well. Different head values corresponding to different
screen elevations were used to evaluate vertical gradients. During the plotting of
this map, anomalous data were identified and marked for further investigation.
The geology of the site and the depositional processes forming the aquifer
were studied to determine what sorts of hydrogeologic phenomena might be
expected. Glacial outwash deposits exhibit trends in sediment size and sorting.
Sediment size decreases and sorting increases from the marginal to the distal
portions of the deltaic/lacustrine deposits. 1 It is expected that this tendency will be
'Mary P Anderson, "Geologic Faces Models: What Can They Tell Us About Heterogeneity," presented to the American
Geophysical Union. Baltimore. May 18.1987
15-126
-------
TABLE 15-13
GROUND-WATER ELEVATION SUMMARY TABLE PHASE II
, Well
Number
MW-1
MW-2
, MW-3
MW-4
MW-5
MW-6
MW-7
MW-8
|MW-9
MW-10
MW-11
MW-12
MW-1 3
MW-1 4
MW-1 5
MW-1 6
MW-1 7
MW-1 8
MW-1 9
MW-20
MW-21
MW-22
MW-23
MW-24
MW-25
MW-26
MW-27*
MW-28
MW-29
MW-30
MW-31
MW-32
MW-33
Ground
Elevation
(ft)
162.80
162.50
174.20
201.90
186.30
144.30
144.60
155.10
160.50
160.40
154.70
159.50
162.20
162.10
162.00
162.00
162.00
161.90
137.10
137.20
141.40
141.60
204.30
143.90
143.80
143.80
142.70
142.80
172.00
172.20
203.10
174.20
Well
Depth
(ft)'
76.50
22.50
31.00
5400
47.50
39.50
19.50
24.00
61.00
30.00
27.00
26.50
29.00
29.00
29.00
29.00
71.00
72.00
24.00
17.00
26.50
15.10
225.50
70.00
39.00
24.00
46.00
23.00
85.50
24.85
61.00
94.00
Midpoint of
Well Screen
Elevation1
145.7
150.4
141.3
107.3
127.6
133.6
135.0
132.9
130.2
135.5
139.2
139.1
139.1
135.5
104.5
103.4
116.6
123.7
118.4
13.0
-10.2
76.4
107.3
123.2
100.2
123.3
90.0
150.8
145.6
83.7
Screen
Length
(ft)
3
3
3
3
3
3
3
3
3
3
3
3
10
10
10
3
25
25
5
5
5
5
20
5
5
5
5
5
5
5
5
5 .
Water
Level Elevation
9/1/82
150.54
156.85
149.95
135.78
135.94
149.04
141.53
144.62
140.57
141.05
141.22
140.66
140.67
140.87
140.52
140.53
127.83
127.82
135.39
135.35
184.98
136.47
130.20
130.17
12786
127.88
152.70
151.68
154.78
150.49
•Not installed.
1 Assumes screens are installed one foot above the bottom of the well.
15-127
-------
TABLE 15-13 (continued)
Well
Number
MW-34
MW-35
MW-36
MW-37
MW-38
MW-39
MW-40
MW-41
MW-42
MW-43
MW-44
MW-45
MW-46
MW-47
MW-48
MW-49
MW-50
MW-51
MW-52
MW-53
MW-54
MW-55
MW-56
MW-S7
Screen
Reference
Points
SRP-1
SRP-2
SRP-3
SRP-4
SRP-5
SRP-6
SRP-7
SRP-8
Ground
Elevation
(ft)
186.20
203.20
189.40
189.50
189.30
154.90
173.80
173.70
134.20
139.50
139.50
144.32
144.15
141.50
141.60
143.00
143.00
157.00
157.00
159.30
145.80
145.90
133.60
141.90
Well
Depth
(ft)
75.80
106.25
101.20
48.00
135.30
68.00
47.50
75.30
64.00
32.10
28.00
35.00
25.00
34.00
17.00
72.20
30.20
70.30
34.00
77.90
52.00
35.00
20.30
Midpoint of
Well Screen
Elevation1
113.9
100.4
91.7
145.0
57.5
90.5
129.8
101.9
73.7
80.9
115.0
112.8
122.6
111.0
128.1
74.3
116.3
90.2
126.5
84.9
97.3
114.4
116.8
Screen
Length
(ft)
5
S
5
S
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
114.41
114.92
116.05
1 1 5.86
NA
128.81
137.28
134.11
Water
Level Elevation
9/1/82
U9.72
144.31
143.22
150.51
145.04
142.45
146.59
141.95
117.62
117.24
119.62
128.97
126.48
131.91
131.74
123.22
123.85
149.58
139.48
141.09
120.18
121.63
119.84
*Not installed.
1 Assume screens are installed one foot above the bottom of the well.
15-128
-------
reflected in hydraulic conductivities throughout the outwash deposits at the facility.
There is some suggestion of such a trend in the head data from the site.
The map of hydraulic head values and screen midpoint elevations were
evaluated considering both the possible hydrogeologic phenomena expected for
the geology of the area and the depositional processes creating the aquifer. Several
working hypotheses were developed to explain the apparent ground-water flow
patterns and the identified vertical gradients.
• Hypothesis 1: Vertical gradients can be explained by classifying areas
where the vertical gradients were reflective of discharge and recharge
areas. (See Figure 15-41.)
• Hypothesis 2: The top surface of the till forms a trough with a saddle.
(See Figure 15-40.) The vertical gradients showing higher head with
depth reflect the movement of water as it flows upward over the saddle.
• Hypothesis 3: The vertical gradient may correlate with locations of
buildings and parking lots at the site. Recharge occurs primarily where
the ground is not paved. The downward gradient near the river may be
caused by runoff flowing downhill and recharging the ground water at
the edge of the pavement.
• Hypothesis 4: Most of the ground-water flow is horizontal. The vertical
gradients reflect phenomena whose scale is smaller than the resolution
of available data, and an accurate interpretation cannot be made.
Geologic systems exhibit heterogeneity on different scales, causing
fluctuations in head on different scales. The small-scale fluctuations
detected at the site are due to undefined causes and may represent:
1. details of stratigraphy (such as till beds in parts of the outwash
deposit),
2. artificial recharge and discharge (such as leaky sewer pipes), or
3. errors in the data.
15-129
-------
15-130
-------
To characterize flow at the site and to support the design of corrective
measures (if needed), a working (conceptual) model of flow at the site should be
developed. This model, in this case a vertical flow net, can be used to identify data
gaps and to prioritize gathering, of the necessary additional information.
Considering the hypotheses developed, an area for characterizing the vertical flow
regime was selected. Determination of this area, where a geologic cross-section
and flow net will be constructed, was based on:
• Assumptions and requirements necessary to construct flow nets, as
identified in the Criteria for Identifying Areas of Vulnerable
Hydroqeoloqy. Appendix B: Ground-Water Flow Net/Flow Line
Construction and Analysis (Vulnerable Hydrogeology, Appendix B). For
example, ground-water flow should be roughly parallel to the direction
of the cross-section and vertical flow net.
• Flow being representative of the hydrogeology of the facility.
• Flow representing the major paths of ground-water movement. For
example, the aquifer is shaped like a trough and a major portion of the
ground-water flow occurs in the middle of this trough; therefore, a cross-
section and flow net should be constructed along the axis of the trough.
A geologic cross-section was constructed for the area of interest and is
identified as T-T in Figure 15-40. A flow net was then constructed following the
methodology described in Vulnerable Hvdroqeoloqy. Appendix B; see Figure 15-42.
Construction of a vertical flow net requires a graphical solution of Darcy's Law.
Data that do not fit the solution become evident in Figure 15-42 as shown, for
example, by the head value for MW 52.
Construction of a vertical flow net allowed for a systematic evaluation of the
various hypotheses. Hypothesis 1, where vertical gradients are labeled recharge and
discharge, is rejected because the gradients vary significantly in a very irregular
pattern (compare well clusters MW 14-18 and MW 12 and 53); there is no apparent
reason that natural recharge would vary so irregularly. Hypothesis 2 seemed
reasonable initially but, after closer inspection, is rejected because upward
15-131
-------
rsi
91
15-132
-------
gradients are not consistently found near the saddle. Hypothesis 3 is feasible and
deserves further study. Aerial photographs were examined to identify paved and
unpaved areas, but the available ground-water data are insufficient for detailed
correlation to these distinct areas. Additional data are needed to construct a more-
detailed flow net to further evaluate this hypothesis. Hypothesis 4, which asserts
that most of the flow is horizontal, addresses the area of the site where the major
portion of ground-water flow occurs. Although it relies on undefined causes to
explain fluctuations, it reflects the most logical explanation of the data.
Results
During construction of the flow net and testing of the hypotheses several
issues were identified. One of the most important gaps in the study to date is how
localized flow at the site fits into the regional ground-water flow regime. Regional
flow issues would need to be resolved prior to determining the extent and type of
corrective measures, if necessary. The following regional flow issues were
identified:
• Geologic information beyond the facility property boundary is necessary
to explain the suspected bedrock in the middle of the river directly beside
the site to characterize the regional ground-water flow (i.e., to
determine the possibility for contamination of regional ground water).
The difference in elevation of the top of the bedrock in the river and the
top of the bedrock throughout the facility is approximately 120 feet.
How can this be explained? Is the bedrock surface irregular or is this rock
a glacially-transported boulder exposed in the river? How does this
affect regional ground-water flow?
• Data consistently show a downward gradient (i.e., recharge conditions)
near the river. This is difficult to explain because rivers in this region are
not expected to be losing streams (Heath, 1984). The expected flow
direction near a ground-water discharge area, in this case a gaining
stream, is upward. Data points showing downward flow near the river
are not included in flow net T-T. (Further investigation of vertical
gradients near the river is recommended). If this downward gradient
near the river is confirmed, near-water-table contamination could move
15-133
-------
downward and contaminate deeper ground water. If deeper, regional
contamination must be addressed, corrective measures may be
significantly more difficult and extensive.
Other issues deal with localized flow patterns that may affect design of corrective
measures. Resolution of these issues will probably not change the overall scope of
corrective measures, but would need to be considered in the detailed design.
These localized flow pattern issues are as follows:
• The hydraulic head in the brook is higher than the head in the closest
wells in the aquifer, but the water slopes toward the stream. This is
inconsistent. If ground water from the site is not discharging into this
stream, fewer interceptor wells may be needed.
• Anisotropy must be taken into account in determining the region of flow
captured by interceptor wells, drains, etc.
• Till identified as lenses in outwash deposits may actually be continuous
with upgradient till, causing the aquifer to flow under confined
conditions. Are the till beds isolated lenses or are they continuous? If the
till beds in the outwash aquifer are continuous and isolate adjacent
zones within the aquifer, they will have the potential of blocking flow to
interceptor wells that may be included in the corrective measures plan.
• Vertical gradients of 0.25 and 0.002 in the same geologic unit are
presented. Are these gradients accurate and how can they be explained?
There could be artificial discharge (pumping) or recharge (possibly from a
leaking sewer) near the wells showing a high vertical gradient. The areas
labeled discharge areas show no signs of surface water or other surficial
evidence of discharge. Artificial recharge and discharge may create areas
of relatively constant head, such as where ground water contacts leaky
sewers; these areas could limit the growth of cones of influence of any
interceptor wells or drains. Also, any contaminated water that may be
discharging from pipes should be identified and corrected.
15-134
-------
Case Discussion
Further investigation is necessary to resolve the above issues. Regional flow
issues should be resolved first. This information would be used to better
understand localized flow patterns which would affect the design of corrective
measures. The following options for further investigation are suggested:
1. Study the regional geology and hydrogeology. Techniques that could be
employed using existing data include review of geologic maps, analysis
of well logs, and interpretation of existing surface geophysical data (e.g.,
gravity and magnetic surveys). Measurement of water level elevations in
wells outside the site would also be useful.
2. Conduct a detailed study of the depositional environment of the glacial
deposits on the site. This should provide a better understanding of flow
patterns.
3. Collect a full-year series of head data at existing wells to differentiate
transient from steady-state (e.g., artificial from natural) effects in the
measured heads.
4. Conduct multiple-well pumping tests to determine the degree of
connectivity of geologic formations using wells at different depths and
locations. [Note: this should be done with careful attention to details of
well construction so that it is understood exactly what is being
measured.]
5. Collect detailed chemical data (including major ions and contaminants)
at the existing wells and interpret them to aid in characterizing the flow
regime.
6. Drill one or more wells into the bedrock near the river to determine the
vertical component of ground-water flow at this location.
Options 1 through 5 above are recommended prior to drilling additional wells
in the outwash deposits, unless more wells are needed to delineate the release.
15-135
-------
Further single-well hydraulic conductivity tests in the glacial deposits are not
recommended at this time. The large-scale flow in the outwash aquifer should be
determined by the location and relative degree of continuity of the till versus the
sand because the permeability contrasts between the till and sand is so much
greater than the variability among the different sands. (See paper by Graham Fogg
.in Water Resources Research, 22, 679.) Single-well tests would be useful for
determing localized hydraulic conductivities of the sand bodies, not their
connectivity.
Gathering existing data and constructing an initial vertical flow net proved
useful in identifying data gaps in defining ground-water flow, and identified
problems due to differing interpretations of the existing data. Determining options
for gathering additional data necessary to resolve these issues was based on a
qualitative understanding of the ground-water flow regime gleaned from
construction of the vertical flow net.
References
Fogg, Graham. Water Resources Research. 22, 679.
Heath. 1984. Ground Water Regions of the U.S. USGS Water Supply Paper No.
2242.
U.S. EPA. 1986. Criteria for Identifying Areas of Vulnerable Hydroqeoloqy,
Vppendix B: Ground-Water Flow Net/FIc
of Solid Waste. Washington, D.C. 20460.
Appendix B: Ground-Water Flow Net/Flow Line Construction and Analysis. Office
15-136
-------
CASE STUDY 23: PERFORMING A SUBSURFACE GAS INVESTIGATION
Points Illustrated
• Design of a phased monitoring program to adequately characterize the
extent and nature of a subsurface gas release.
• The use of ambient air and basement monitoring to supplement
monitoring well data.
• The importance of subsurface characterization prior to design of a
monitoring network.
Introduction
Gases produced in a landfill will migrate via the path of least resistance.
Subsurface, lateral migration of landfill gas can occur due to natural and man-made
barriers to vertical gas migration, such as impermeable overlying soil layers, frozen
soil, or surf ace water. Installation of a gas-monitoring well network, in con,jnction
with sampling in buildings in the area, can be used to determine the need for
corrective measures.
Facility Description
The unit in question is a landfill covering approximately 140 acres and
bordered by a river on one side and a flood wall on the other. Beyond the floodwall
lies a residential area (Figure 15-43). Several factors contribute to the subsurface
gas migration problem at this landfill. The site reportedly received large quantities
of organic wastes which, when decomposed in the absence of air, produce methane
and carbon dioxide gases. The presence of "tight", low permeability soils at the
ground surface (12 feet of clayey silt at the surface grading to coarse sand and
gravel at a depth of 55 feet) in the residential area, combined with a rapidly rising
water table below the landfill due to increased infiltration, restrict the vertical area
available for gas migration and encourage lateral movement.
15-137
-------
L
c
_2
Q.
LO
3
CT>
15-138
-------
Investigation pf the gas migration began when foul odors and explosive levels
of methane (5 to 15 percent by volume in air) were discovered in the basement of a
home approximately 200 feet from the landfill. Residents in the area were
evacuated, a sampling network was installed, and monitoring was conducted.
Sampling Program
The sampling was conducted in four phases, an initial screening phase and a
more detailed three-phase sampling program. The monitoring network for the
initial screening phase consisted of four wells (W1 through W4) aligned
perpendicular to the long axis of the landfill, in the direction of (and extending
beyond) the house where the gas was initially detected (Figure 15-43). The wells
were drilled to an approximate depth of 30 feet below the land surface with the
farthest well located about 1000 feet from the landfill boundary. These wells were
sampled twice a day for a month. Samples were analyzed for methane and
combustible hydrocarbons. The results of this initial monitoring showed average
methane levels to be highest at the monitoring well closest to the landfill (30
percent by volume), and roughly grading to below the detection limit at the well
farthest from the landfill.
Grab and composite ambient air samples were also taken at the landfill and
around houses in the neighborhood where gas was detected during the initial
monitoring phase. These samples were analyzed for methane and other
combustible hydrocarbons. No gases were detected above normal background
levels in any of these above ground samples.
The next phase of monitoring (Phase I of the detailed sampling) involved the
installation of 14 new gas monitoring wells (1-1 through 1-14 in Figure 15-43). Most
of these were placed in a line 250 feet from and parallel to the longitudinal axis of
the landfill. Seven of these wells were drilled to an average depth of 55 feet, at
least 5 feet below the water table so that ground-water levels could be monitored.
The other seven wells averaged 30 feet and did not intercept ground water. As
shown in Figure 15-44, each well consists of three separate gas monitoring probes at
evenly spaced depth intervals. Each probe was packed in gravel to allow gas to
collect in its vicinity. Clay plugs were installed between each probe interval and
15-139
-------
CAST IRON COVER SET IN CONCRETE
GROUND SURFACE
VALVE
PROBE A
1/4" DIAMETER
POLYETHYLENE
TUBING
PROBE B
LEGEND
NATIVE SOIL
BACKFILL
BENTONITE PLUG
PEA GRAVEL
VALVE
PROBE A
PROBE B
2" OIAMETER
PVC PIPE
PROBE C
2" DIAMETER
WELL SCREEN
PROBE C
Figure 15-44. Gas Monitoring Well
15-140
-------
between the top probe and the surface to minimize vertical movement of gas in the
well. After two months of monitoring the well headspace twice monthly, concern
over the high levels of methane that were being measured prompted an expansion
of the monitoring well system.
The Phase II monitoring network involved another 14 wells (11-1 through 11-14)
installed to a depth of 6 feet along three radial lines from the landfill. These wells
were monitored twice monthly with the Phase I wells. Methane was not detected at
these wells because they were not deep enough to penetrate the clayey silt layer
which in this area extended to a depth of 12 feet. Had adequate boring logs been
compiled prior to the placement of these wells, the time and money involved in
their installation and sampling could have been saved.
Detailed soil boring logs were compiled during the installation of the Phase III
wells (111-1 through III-8 in Figure 15-43). These wells were drilled to ground water,
averaging 55 feet in depth, were located in the vicinity of the Phase II wells, and
were constructed in the same manner as the Phase I wells, with three gas probes
placed in each well. The Phase III wells were located from 510 to 900 feet from the
lane.'ill. These wells were monitored twice a month for two months concurrently
with the Phase I wells. Methane levels at all but two Phase III wells (which are
located along the same radial line) exhibited explosive concentrations, ranging up
to 67 percent by volume in air. These high concentrations of gas prompted another
round of sampling of homes in the vicinity of wells exhibiting high methane
concentrations.
Methane and combustible hydrocarbons were measured in basements, crawl
spaces, and living areas of 28 homes adjacent to the landfill. All proved to be well
below the lower explosive limit of methane.
Wells were then selected based upon proximity to houses exhibiting the
highest levels of combustible gases, and sampled to determine gas composition and
concentration. The proportion of constituents in the collected gas was similar in all
samples analyzed, and concentrations decreased with increasing distance from the
landfill.
15-141
-------
Ambient air sampling for organic gases at the landfill and in the residential
area was also performed at this time and showed low levels of several organic
compounds. Air samples collected in houses near the landfill showed the presence
of two of the gas components measured in the test wells (methane and ethane).
The gas migration hazard had been sufficiently characterized so that a plan for
corrective measures could be developed. This involved the installation of 31 gas
extraction wells which were located along a line between the landfill and the
residential areas, and a blower system to "pump" the gas out of these extraction
wells.
Results
The monitoring program implemented for this case was, for the most part,
effective in characterizing the extent and concentrations of subsurface gas
contamination. The four initial monitoring wells verified that the landfill was the
source of contamination. Phase I monitoring confirmed that the high levels of
methane were present at all depths monitored and along the entire length of the
landfill. The horizontal location of the Phase II wells, in lines radiating from the
landfill, was appropriate, although the lack of subsurface characterization rendered
them useless. Phase III sampling established the vertical and lateral extent of
subsurface contamination into the residential area.
Throughout the study, ambient air sampling as well as monitoring of homes in
the area of concern provided adequate safety control, as well as an additional
indication of potential migration of landfill-generated gases.
Case Discussion
Subsurface gas migration can occur when atmospheric ventilation of gases
generated in a landfill is insufficient. The gas produced migrates along the paths of
least resistance. Conditions restricting release to the atmosphere, such as saturated
or tight surficial soils, may force the gas to move laterally over considerable
distances.
15-142
-------
This case was selected as an illustration of a phased approach to monitoring a
subsurface gas release. The results of one phase of monitoring were incorporated
into the design of the next phase throughout the study. Monitoring was performed
at discrete vertical levels below the surface and at distances from the landfill that
were adequate to confirm the extent of the contaminant plume.
The study also illustrates the importance of characterizing subsurface
conditions prior to installing monitoring wells. Fourteen unusable wells were
installed and then monitored for two months because of insufficient preliminary
soil (stratigraphic) characterization.
The use of ambient and basement monitoring for gas to supplement
monitoring well data is also noted in this case study. The location of new wells can
be based in part on readings from these sources.
15-143
-------
CASE STUDY 24: USE OF A SUBSURFACE GAS MODEL IN ESTIMATING GAS
. MIGRATION AND DEVELOPING MONITORING PROGRAMS
Point Illustrated
• Predictive models can be used to estimate the extent of gas migration
from a suspected subsurface source. This information can be used to
estimate human exposure and to determine appropriate locations for
monitoring wells and gas collection systems.
Introduction
Methane is a common landfill gas and is often used as an indicator of landfill
gas migration. The subsurface methane predictive model, described in Volume II,
Appendix D of this document, will yield a methane concentration contour map and
predict the distance that methane will migrate. The model consists of a series of
charts developed by imposing a set of simplifying assumptions on a general
methane migration computer model.
A methane migration distance prediction chart is used to find a preliminary
migration distance based on the age of the site and the soil type. The remaining
charts are used to find correction factors which are in turn used to adjust the
migration distance. These factors are based upon site characteristics (e.g., depth of
the waste).
Facility Description
The unit is located on a 583-acre site in a suburb of a major metropolitan area.
Figure 15-45 shows the site layout. The landfill itself occupies 290 acres. 140 acres of
the landfill were used for the disposal of hazardous wastes. Both hazardous and
nonhazardous wastes were disposed at the site from 1968 to 1984. Hazardous waste
disposal ended in 1984. The disposal of sewage treatment sludges and municipal
refuse continues. As seen in Figure 15-45, residential development has taken place
with houses now bordering the facility to the south. A population of 30,000 to
40,000 people reside within a mile radius of the landfill center.
15-144
-------
Scale Houi«
Laboratory, aa4
Truck Scales
/\ c{V, Vi
OJ^-.SS..
UMOF1U PUN
EU. uanoi NAP
Central
Olreetioo
e(
Cround V«ttr
flow
Seal* 1*:U30'
Figure 15-45. Facility Map
15-145
-------
The unit is a V-shaped fill overlying sediment and bedrock. The rock type is a
poorly consolidated, fractured sandy silt offering no lithologic barrier to gas
migration. The shape of the water table has not been established. Also unknown
are the possible effects of local, permeable formations such as sand lenses, faults,
etc.
The warm climate at the site encourages rapid degradation of organic wastes
and therefore rapid gas production. Site characteristics suggest that vertical gas
migration is not hindered. However, the compaction of the fill cover by truck traffic
combined with the rapid production of gas has forced lateral migration through the
fractured sandy silt.
Applying the Subsurface Methane Predictive Model
The subsurface methane predictive model allows the development of a
subsurface methane concentration contour map. The model predicts the distance
methane will migrate from a unit based on its age, depth, soil type, and
environmental factors. A contour map for two different methane concentrations, 5
and 1.25 percent, is predicted. The likelihood of human exposure can be estimated
from the location of the contours with respect to on-site and off-site structures.
Application of the model involves three steps. The first step is the prediction
of gas migration distances, based on the age of the landfill and the local soil type.
The unit of interest is 18 years old and has sandy soils. Figure 15-46 shows the
unconnected methane migration distances for various soils over time. From
Figure 15-46, the unconnected migration distances for the subject site are 165 feet
and 255 feet for 5 and 1.25 percent methane concentrations, respectively.
The second step in applying the model involves the application of a correction
factor to the migration distances based on waste depth. The deeper the waste, the
greater the opportunity for subsurface migration. Figure 15-47 is used to find the
correction factors for depth. For the subject waste unit the depth is 25 feet, which
corresponds to a correction factor of 1.0 for both concentrations.
15-146
-------
1
1
^
!
I
1
Q
V)
|
1
\
\
s
*»
CLAY
X
13;
i
\
\
\
\
\
X
I
c
*
<
u
^
-
!
•)
j
)
I
\
\
\
•^
\
.
<
U
U
A I
\\
\>
k
*
"
1 « !
id-iONVlSIONOUVa?
»
— -
2 4 6 8 10 12 14 16 18 21
SITE AGE - YEARS
Figure 15-46. Uncorrected Migration Distances for 5 and 1 25% Methane
Concentrations
a o
)IW
15-147
-------
1
1
1
1
0.
UJ
UJ
O
|
1
I
\
\
\
\
\
\
\
\
\
\
1
1
1
t
1
a.
lit
UJ
O
S
j
1
\
\
\
\
>
1
\
\
\
\
V
\\
\
\
\
1
<
c
i
t
t
*
H
(
\
^x
\v
\\
%i
i
i
i
i
s
ABOVE Gf
•
]
J
5
\
V
\
i
V
\
\
(M
00
to
*•
v
^
M
0 1
. YEARS
«— **•
O
<
UJ
M
^
OJ
_o
ai
ca
a
(U
a
o
t*»
(SI
O
c
o
ai
i_
O
u
3
0>
fN
Nouoaaaoo
15-148
-------
The final step in applying the model is the correction of migration distances
based on surface venting conditions. The following equation is used to calculate
the adjusted correction factor, ACF:
ACF a [(ICF-1)(fraction of site which is impermeable)] + 1
The impervious correction factor, ICF, is obtained from Figure 15-48. In the above
equation, ICF is adjusted to account for the fraction of time the solid is saturated or
frozen and the fraction of the land area that is impermeable due to natural or man-
made barriers. If corrections for both time and area are required, the fractions are
additive. From Figure 15-48, the ICF for a unit 18 years old and 25 feet deep is 2.4.
Site charcteristics together with weather conditions indicate a value of 0.4 for the
fraction of impermeable area. Substituting these values into the above equation
yields an adjusted correction factor of:
ACF = [(2.4-1X0.4)] + 1 = 1.56.
Results
Table 15-14 summarizes the results from steps one through three of the model
application. The predicted migration distances for methane are found by
multiplying the unconnected distance from step one by the correction factors from
steps two and three. The predicted distances of travel for methane are 255 feet and
395 feet for 5 and 1.25 percent concentrations, respectively.
TABLE 15-14
MODEL RESULTS
Methane
Concentration
(percent)
5
1.25
Uncorrected
Distance
(ft)
165
255
Correction
for Depth
1.0
1.0
Correction
for Venting
1.56
1.56
Corrected
Distance
(ft)
255
395
15-149
-------
\
\
\
\
&
V*
u
C
i
j
3
ft
>
1
S
Ut
%
1
\
O
v»
\
\
\
i
a
«
\
\
\
\
1
a
u
u
i c
i i
8 '
O
I
1
"l
\
\
t
\
1
1
\
\
\
\
\
\
\
i C
J- . 2
5 P
i i
Ul
ft u,
3 oe
,.|MPERVIC
c
1
1
1
1
1
1
1
|
li
H.
1 »l
\UI
\^
\]
'
•
1
|
I i
UJ '
0 !
DVIOUS - 10
if'
1
1
1
1
1
I
1
{/
1 /
I r
f/
I/
*
'
I
I
I
a.
UJ
a
2
S
>
e
u
m
|
/
/
/
/
/
/
/
/
f
I
1
1
•
a.
Ul
a
»
3
>
e
I
f
>
^
u
C
**
C
a>
u
o
w_
o
V)
X
UJ
LU
U
V)
«e
O
•M
U
I?
« JS
2! -o
o §
u o
3 <
I §
41 -O
a c
c o
— VJ
00
*T
I
LO
aj
3
cn
aoiovj
15-150
-------
Case Discussion
Figure 15-49 is a methane concentration contour map developed from the
predicted travel distances. The map indicates that the possibility of human
exposure to landfill gas is high. Landfill gas is known to be present and well drilling
operations at the landfill have caused minor explosions. The monitoring wells along
the facility perimeter and testing in nearby homes indicate that gas has migrated
off site. Both the 5 percent and 1.25 percent methane contours enclose homes
evacuated because of gas accumulation. Measures have been taken to mitigate the
immediate problems and the landfill operators have installed additional gas
collection wells and extended the monitoring system.
15-151
-------
Q
LU
o
111
OJ
E
0)
c
o
o
u
o»
0.
"O
c
TJ
3
Ol
15-152
-------
CASE STUDY 25: USE OF METEOROLOGICAL/EMISSION MONITORING DATA
. AND DISPERSION MODELING TO DETERMINE CONTAMINANT
CONCENTRATIONS DOWNWIND OF A LAND DISPOSAL
FACILITY
Point Illustrated
• How to use meteorological/emission monitoring data and dispersion
modeling to estimate contaminant concentrations.
Introduction
Concern over possible vinyl chloride transport into residential areas adjacent
to a land disposal facility prompted initiation of this study. As a followup to a
screening assessment (involving emission modeling) a survey and emission
monitoring program with the application of an air dispersion model were used to
assess potential health hazards.
Facility Description
The facility is a landfill which has been in operation since 1963. The facility
occupies an area of 583 acres, of which 228 acres contain hazardous and municipal
waste. The facility and surrounding terrain is hilly with elevations ranging from 600
to 1150 feet above mean sea level. Residential areas are located immediately
*
adjacent to the south and southeast facility boundaries, as shown in Figure 15-50.
The facility previously received waste solutions from the synthesis of polyvinyl
chloride which included the vinyl chloride monomer. Gas is generated by municipal
waste decomposition and chemical waste volatilization. The primary air release
from the particular unit is vinyl chloride. A gas collection system has not been
installed for this unit.
15-153
-------
15-154
-------
Program Design/Data Collection
A screening assessment (based on emission/dispersion modeling) was
conducted to evaluate vinyl chloride emissions from the landfill. Evaluation of
these results indicated that emission monitoring should be conducted to more
accurately quantify the release. An isolation flux chamber was used to measure
vinyl chloride emissions during a three-day period in August. This sampling period
was selected based on the screening assessment results to represent worst case
emission and dispersion conditions.
An on-site meteorological survey program was also conducted to characterize
wind flows at this complex terrain site. Two meteorological stations were deployed
to evaluate wind flows, as influenced by complex terrain, which may impact the two
adjacent residential areas (see Figure 15-50.) A one-month data collection period
during August was conducted to characterize on-site wind and stability patterns
during worst-case, long-term emission/dispersion conditions. Although the facilty is
located in complex terrain, the diurnal wind pattern during the meteorological
survey was very consistent from day to day. Therefore, the one-month
meteorological monitoring period was adequate for this RFI application.
Program Results/Data Analysis
The emission monitoring and meteorological monitoring data were used as
input for dispersion modeling. The wind patterns were different for each of the on-
site meteorological stations (see Table 15-15). Therefore, two sets of modeling runs
were conducted (meteorological station A data were used to estimate
concentrations at residential area A and meteorological station B data were used to
estimate concentrations at residential area B).
The dispersion modeling results indicated that estimated concentrations at
both residential areas were significantly below the RFI health criteria. Therefore,
followup air release characterizations were not necessary and information was
sufficient for RFI decision making.
15-155
-------
TABLE 15-15
SUMMARY OF ON-SITE METEOROLOGICAL SURVEY RESULTS
I
i
j Prevailing daytime
; wind direction
I Prevailing nighttime
i wind direction
Station A
S
NNE
Station B
SW
ENE
15-156
-------
Case Discussion
Emission sampling was appropriate for this application because of the
uncertainties associated with emission rate modeling for landfills (including
uncertainties in emission modeling inputs such as the waste composition and spatial
distribution). The isolation flux chamber technique provided a basis for direct
measurement of vinyl chloride emission rates for dispersion modeling input.
The conduct of an on-site meteorological monitoring survey provided the
required wind and stability input for dispersion modeling. The use of multiple
meteorological towers for this application was necessary to characterize wind flow
patterns in complex terrain and to account for off-site exposure at two residential
areas subject to different wind conditions. The combination of emission
monitoring, meteorological monitoring and dispersion modeling provided an
effective air release characterization strategy for this RFI application.
References
B 'ker, L.W. and K.P. MacKay. 1985. Screening Models for Estimating Toxic Air
Pollution Near a Hazardous Waste Landfill. Journal of Air Pollution Control
Assocation, 35:11.
15-157
-------
CASESTUDY26: USE OF METEOROLOGICAL DATA TO DESIGN AN AIR
MONITORING NETWORK
Points Illustrated
• How to design an air monitoring program
• How to conduct an upwind/downwind monitoring program when
multiple sources are involved.
Introduction
A screening assessment (based on emission/dispersion modeling)
commensurate with RFI guidance was conducted to characterize hazardous air
constituents being released from a wood treatment facility. Evaluation of these
screening results indicated that it was necessary to conduct a monitoring program
to more accurately quantify air emissions from units at the facility. Meteorological
data were first collected to determine the wind patterns in the area. The wind
direction data with the locations of the potential emission sources were then used
to select upwind/downwind air sampling locations.
Facility Description
The site is a 12-acre wood treatment facility located in a flat inland area of the
southeast. Creosote and pentachlorophenol are used as wood preservatives; heavy
metal salts have been used in the past. Creosote and pentachlorophenol are
currently disposed in an aerated surface impoundment. Past waste disposal
practices included treatment and disposal of the metal salts in a surface
impoundment, and disposal of contaminated wood shavings in waste piles. The
constituents of concern in the facility's waste stream include phenols, cresols, and
poiycyclic aromatic hydrocarbons (PAH) in the creosote; dibenzodioxins and
dibenzofurans as contaminants in pentchlorophenol; and particulate heavy metals.
The potential emission sources (Figure 15-51) include the container storage facility
for creosote and pentachlorophenol, the wood treatment and product storage
areas, the aerated surface impoundment for the creosote and pentachlorophenol
wastes, and the contaminated soil area which previously contained both the surface
15-158
-------
PREVAILING
WIND
DIRECTION
INACTIVE SURFACE
IMPOUNDMENT AND
CONTAMINATED
WOOD SHAVINGS
STORAGE AREA
AERATED
SURFACE
IMPOUNDMENT
STATION 2 (V)
OFFICE
STATION 4 (V)«
TREATMENT
AND PRODUCT
STORAGE AREAS
I
STATION 1 (PVM)
CONTAINER
STORAGE
FACILITY
STATION 3 (PV)
GATE
KEY
* AIR MONITORING STATIONS
P PARTICULATE MONITORING
V VOLATILE CONSTITUENT MONfTORl
M METEOROLOGICAL MONITORING
N
Figure 15-51. Site Plan and Locations of Air Monitoring Stations
15-159
-------
impoundment for treating the metal salts and the wood shavings storage area.
Seepage from these .waste management units has resulted in documented ground-
water and surface water contamination.
The area surrounding the facility has experienced substantial development
over the years. A shopping center is now adjacent to the eastern site perimeter. This
development has significantly increased the number of potential receptors of air
releases of hazardous constituents.
Program Design/Data Collection
Preliminary Screening Survey-
A limited-on-site air screening survey was first conducted to document air
releases of potentially hazardous consituents, to prioritize air emission sources, and
to verify screening assessment modeling results and the need to conduct a
monitoring program. Total hydrocarbon (THC) levels were measured with a
portable THC analyzer downwind of the aerated surface impoundment, wood
treatment area, and product storage area. Measurements were also made upwind
of all units to provide background concentrations. Because THC levels detected
downwind were significantly higher than background levels, a comprehensive
monitoring program to characterize releases to the air was designed and
implemented.
Waste Characterization-
To develop an adequate monitoring program, the composition of wastes
handled in each waste management unit was first determined to identify which
constituents were likely to be present in the air releases. Existing water quality data
indicated contamination of ground water with cresols, phenol, and PAHs and of
surface water with phenols, benzene, chlorobenzene, and ethylbenzene. A field
sampling program was developed to characterize further the facility's waste stream.
Wastewater samples were collected from the aerated surface impoundment and
soil samples were collected from the heavy metal salt waste treatment/disposal
area. Analytical data from this sampling effort confirmed the presence of the
constituents previously identified. Additional constituents detected included
toluene and xylenes in surface impoundment wastes, and arsenic, copper,
chromium, and zinc in the treatment/disposal area.
15-160
-------
Based on their individual emission potentials and potentials for presenting
health and environmental hazards, the following constituents were selected for
measurement in the air monitoring program:
Volatile/semi-volatile constituents: toluene, benzene, total phenols, penta-
chlorophenol, PAHs, cresols
Particulate constituents: aresenic, copper, chromium, zinc.
Meteorological Data Collection--
Meteorological information is critical for designing an air monitoring program
because stations must be located both upwind and downwind of the contaminant
sources. Therefore, a one-month meteorological monitoring survey was conducted
at this flat terrain site. The survey was conducted under conditions considered to be
representative of the summer months during which air samples would be collected.
Summer represented worst-case conditions of light steady winds and warm
temperatures. The collected meteorological data showed that the local wind
direction was from the southeast. No well-defined secondary wind flows were
identified.
Initial Monitoring--
Alternative methods were considered for monitoring emissions from the
aerated surface impoundment and contaminated storage area. Direct emission
measurements (such as use of isolation flux chambers) would not be practical for
aerated ponds or for monitoring particulate emissions from area sources.
Therefore, an air monitoring program with samplers located in proximity to the
other units of concern was selected for this application.
The on-site meteorological survey data were used with the EPA atmospheric
dispersion model, ISC (Industrial Source Complex Model), to estimate worst-case air
emission concentrations and to help determine the locations for the air sampling
stations. The ISC model was used because it is capable of simulating conditions of
point and non-point source air emissions. Using the established southeast wind
direction, maximum downwind concentrations were predicted for different
15-161
-------
meteorological conditions (e.g., wind speed). Upwind background stations and
downwind monitoring stations were selected based on the predicted dispersion
pathways. Because the releases from the individual waste management areas
overlapped, the model also provided a means for separating the incremental
contamination due to each source.
Figure 15-51 shows the locations of the selected sampling stations. Station 1 is
the upwind background station. Here background volatile concentrations,
particulate concentrations, and meteorological conditions were monitored.
Stations 2 and 4 were located to identify volatile emissions from the aerated surface
impoundment and wood treatment/product storage areas, respectively. Station 3
was located downwind of the inactive surface impoundment/wood shavings
disposal area. This station was sited to document releases from these waste
management units and to document worst-case concentrations of volatiles and
particulates at the facility property boundary. For this application the locations of
Stations 2,3 and 4 were adequate to characterize air concentrations at both the unit
boundary as well as the facility property boundary (due to the proximity of these
two boundaries in the area downwind, based on the prevailing wind direction, of
the units of concern). A trailer-mounted air monitoring station was used to
supplement the permanent stations and to account for any variability in wind
direction.
Sample Collection-
The air quality monitoring was conducted over a three-month period during
the summer. Meteorological variables were measured continuously on site
throughout the study. Air samples were taken over a 24-hour period approximately
every six days. The sampling dates were flexible to insure that worst-case conditions
were documented.
Volatile and semi-volatile constituents were sampled by drawing ambient air
through a sampling cartridge containing sorbent media. A modified high volume
sampler consisting of a glass fiber filter with a polyurethane foam backup sorbent
(EPA Method TO4) was used to sample for total phenols, pentachlorophenol, and
PAHs. Benzene and toluene were collected on Tenax sampling cartridges (EPA
15-162
-------
Method T01) and cresol was collected on silica'gel cartridges (NIOSH Method Z001).
Participates were collected on filter cassettes using high-volume samplers.
In addition to the constituents previously discussed, Appendix VIII metals were
analyzed on the first few sets of samples. These analyses were conducted to identify
air releases of constituents other than those known to be present. The results
indicated that no additional constituents were present in significant concentrations,
so the additional analyses were dropped for the remainder of the study.
Program Results/Data Analysis
Standard sampling/analytical methods were available for all the target
monitoring constitutents. Analytical detection limits were below specific health
and environmental criteria for all constituents except cresol. The high analytical
detection limit for cresol which exceeded reference health criteria complicated data
analysis. This difficulty was handled by the routine collection and analysis of waste
water samples during the air monitoring program. These data were used to
estimate cresol levels in the air by comparing its emission potential to the other air
monitoring constituents which have relativeJy low detection levels.
Analytical results obtained during this sampling program established that
fugitive air emissions significantly exceeded reference health criteria. Source
control measures were implemented to reduce emission concentrations below
health criteria levels. Subsequent air monitoring was conducted at the same stations
•
used previously on a weekly basis immediately after implementation of the
remedial measures, and on a quarterly basis thereafter.
Case Discussion
This case illustrates a sequence of tasks which were taken to design an air
monitoring program at a site with multiple air emission sources. An initial field
survey was conducted to identify 'local prevailing wind patterns and to identify
potential downwind receptors of fugitive air emissions. The meteorological survey
results were used to design an effective monitoring network. Monitoring station
locations were selected to obtain background conditions and to document air
releases downwind of each emission source. Also, the monitoring strategy included
15-163
-------
use of a portable sampling station to provide flexibility in sampling locations to
account for variation in wind direction. Spatial variability in air concentration was
assessed with the aid of an air dispersion model to assist in data interpretation.
Air emissions data showed an air release of hazardous constituents
significantly above health crtiteria levels. Remedial measures were implemented,
and periodic subsequent monitoring was conducted to insure compliance with the
health criteria.
References
Methods T01 and T04, Compendium of Methods for Determination of Toxic Organic
Compounds in Ambient Air. 1984, EPA-600/4-84-041.
Method 2001, NISOH Manual of Analytical Methods. 1984, National Institute of
Occupational Safety and Health.
15-164
-------
CASE STUDY 27: DESIGN OF A SURFACE WATER MONITORING PROGRAM
Point Illustrated
• When designing a surface water monitoring program, site-specific
sediment and suspended solids information should be considered.
Introduction
Designing a surface water monitoring program to determine the extent of
contamination involves identifying the potential waste sources, the contaminants
likely to be present in each waste stream, and the flow paths by which the
contaminants could reach surface waters. The fate of the contaminants once they
reach the surface water must also be considered when selecting sampling stations
and parameters to be measured. The example described here illustrates the design
of a monitoring program for a river system.
Facility Description
A facility which processed zinc, copper and precious metals from ores operated
along a river for five years. The plant was closed after being cited for repeated fish
kills which were reportedly due to failures of a tailings pond dike. At present, the
site is covered with tailings containing high concentrations of copper, zinc,
cadmium, arsenic and lead. There is no longer a tailings pond.
Site Setting
The site is located on coarse colluvium (hill-slope deposits of weathered
bedrock) and fine-grained alluvium. These deposits are typically 50 feet thick.
Metamorphic rock (phyllite) underlies the unconsolidated materials. Ground water
moves laterally in the gravel formations from the steep valley walls towards the
river.
The site is about 400 feet from the river. Two drainage ditches cross the lower
portion of the site and merge prior to leaving the site. The ditch carries the
15-165
-------
combined flow and discharges directly into the river (Figure 15-52). No other
tributaries enter the river within two miles of this location.
Sampling Program
A surface water monitoring program was designed as part of the Phase I
remedial investigation to determine the extent of contamination in the river.
Existing data from a reconnaissance visit had shown high concentrations of metals
in the drainage ditch sediments (e.g., 5,170 mg/kg Cu and 11,500 mg/kg Zn). Ground
water data from the plant's well showed concentrations of Cu (7 ug/l) and Zn (54
ug/l). The contribution of metals to the river by ground-water discharge at the site
was considered to be relatively small.
Based on a review of the plant history and the available water quality and
sediment data, a monitoring program was designed. The potential pathways by
which metals could reach the river appeared to be direct discharge from the
drainage ditch, seepage of contaminated ground water, and storm water runoff.
Plant records indicated that typical flows in the drainage ditch at its confluence with
the river varied from 1 to 3 cubic feet per second (cfs) in the spring. During extreme
flood conditions, the flow in the ditch exceeded 20 cfs. In the summer, flows in the
drainage ditches at all locations were less than 0.5 cfs. Resuspension of
contaminated sediments in the ditches during storm runoff appeared to be the
most likely pathway for metals to reach the river. The specific metals of concern
were identified as As, Cd, Cu, Pb and Zn based on the processes used at the plant
and the composition of the ores which contained some arsenopyrites (As, Cu),
galena (Pb), and sphalerite (Zn, Cd).
The available soil and water quality data from the reconnaissance visit were
reviewed to determine the likely fate of the metals. Soils in the area were
circumneutral (pH s 6.5) and contained about 0.5 percent organic matter by
weight. Thus the metals, particularly Pb, would be expected to adsorb onto the soil
particles. In the on-site tailings piles, the pH of core samples ranged between 3.3
and 4.9. Low soil pH values had been measured in sediments in the drainage ditch
just downgradient of the tailings pile. The pH of the river during the
reconnaissance was 6.9. The suspended solids concentration was 10 mg/l.
15-166
-------
• S6
To S8
Area of
Former
Tailings
Pond
Site Operation*
Drainage Ditch
Sampling Station
N
t
Scale
I I
0 160 feet
Figure 15-52. Sampling Station Locations for Surface Water Monitoring
15-167
-------
Estimates of the distribution of metals between the dissolved and adsorbed
phases for a range of partition coefficients (Kp) are shown in Table 15-16. For
example, if Kp = 104 and the suspended solids concentration was 10 mg/l, 90
percent of the metal present would be in the dissolved phase. This information
indicated that even though a metal (e.g., lead), was known to sorb strongly, a
significant amount could be transported in the dissolved phase. Thus, both water
and suspended solids should be analyzed for metals. The complete list of
parameters selected for measurement in the Phase I investigation and the rationale
for their selection are outlined in Table 15-17.
The sampling stations were selected to determine river quality up- and
downstream of the site and to determine whether particulates with sorbed metals
were deposited on the river banks or streambed. The sampling stations and the
rationale for their selection are listed in Table 15-18. The station locations are
shown in Figure 15-52. Because floods were considered to be one cause of
contamination incidents, samples were to be collected under both high and low
flow conditions.
Selected results of the surface water quality sampling program for spring
conditions are given below:
Station
S5 (mouth of ditch)
S7 (upstream)
S8 (downstream)
Dissolved Copper
Concentration, ug/1
1110
2.7
4.0
15-168
-------
TABLE 15-16
RELATIONSHIP OF DISSOLVED AND SORBED PHASE POLLUTANT
CONCENTRATIONS TO PARTITION COEFFICIENT AND SEDIMENT
CONCENTRATION
Kp
100
101
102
103
104
SS
(ppm)
1
10
100
1000
10,000
1
10
100
1000
10,000
1
10
100
1000
10000
1
10
100
1000
10000
1
10
100
1000
10,000
Cw/
-------
TABLE 15-17
PARAMETERS SELECTED FOR SURFACE WATER MONITORING
PROGRAM
Parameters
Metals - As, Cd, Cu,Pb, Zn
pH
Dissolved Oxygen, Sulfide,
Fe(ll), Fe(lll)
Alkalinity
Total Dissolved Solids
Major Cations (Ca2 * , Mg2 * ,
Na*,KMMH%)
Major Anions (C1-,SO42-,NO3")
Suspended Solids
Streamflow
Rationale
Determine extent of contamination
Predict sorption behavior, metal
solubility, and speciation
Determine redox conditions which
influence behavior of metals,
particularly the leaching of tailings
A measure of how well buffered a
water is; allows consideration of the
likelihood of pH change
Used as a water quality indicator and
for QA/QC checks
May identify other waste sources;
can influence fate of trace metals
Predict the fraction of metal in water
which is sorbed
Compute mass balances and assist in
identifying sources of observed
contamination
15-170
-------
TABLE 15-18
SELECTED SURFACE WATER MONITORING STATIONS AND RATIONALE
Station
Drainage ditch west of site
(SI)
Drainage ditches on site (S2
and S3)
Downstream of confluence of
2 ditches (54)
Mouth of drainage ditch (55)
River (56, S7, and 59)
River (S8)
Media
Water and sediments
Water and sediments
Water and sediments
Water, suspended
sediment, bedload
Water, suspended
sediment, bedload
Water, suspended
sediment, bedload
Rationale
Determine whether off-site drainage is
significant source of contamination
Identify on-site sources
Provide information for checking mass
balances from the two drainage ditches
Determine upstream water quality
Determine upstream water quality
Determine quality downstream of site
and provide data for mass balance
15-171
-------
A mass balance was computed to determine how much of the apparent decrease
from the ditch (S5) to the downstream river sampling point (58) was due to dilution
and how much could be attributed to other processes (e.g., sorption, precipitation).
The concentration in the river considering dilution alone was predicted using the
following mass balance equation:
CUQU + CWQW
CR =
Qu + Qw
where
CR « downstream concentration of pollutant in river following mixing with
ditch waters (58), ug/l
Cw = concentration in ditch water (55), ug/l
Cu = concentration in river above ditch (57), ug/l
Qw = discharge rate of ditch, ft3/sec
Qu s flow rate of river above ditch, ft3/sec.
At the time of sampling, the flow in the ditch at station 55 was 1 cfs and the river
flow at station 57 was 155 cfs. Using the above equation, the predicted river
concentration for Cu was approximately 10 ug/l. (The observed concentration was 4
ug/l.) The observed decrease in concentration was primarily due to dilution,
although other attenuation processes (e.g., sorption) were probably occurring. The
expected sorbed concentration was estimated as follows:
X = KPC
where
X a sorbed concentration, ug/kg
Kp = partition coefficient, I/kg
C * concentration of dissolved phase, ug/l.
Here, the sorbed concentration of Cu was estimated as 8 x 105 ug/kg (800 mg/kg).
15-172
-------
Case Discussion
This case illustrates the use of site-specific data and the use of information on
the environmental fate of contaminants in the design of a surface water monitoring
program. Site data are needed to locate waste sources and to determine the likely
flow paths by which contaminants reach rivers. An understanding of the general
behavior of the contaminants of interest and of the factors which influence their
fate is helpful in determining where samples should be collected and what
parameters, particularly master variables, should be measured. Collecting data on
such parameters (e.g., pH, suspended solids) ensures that the necessary information
is available to interpret the data.
15-173
-------
CASE STUDY 28: USE OF BIOASSAYS AND BIOACCUMULATION TO ASSESS
- POTENTIAL BIOLOGICAL EFFECTS OF HAZARDOUS WASTE ON
AQUATIC ECOSYSTEMS
Point Illustrated
• Measurements of toxicity (i.e., bioassays) and bioaccumulation can be used to
assess the nature and extent of potential biological impacts in off-site areas.
Introduction
A study was conducted to determine whether leachate discharged into surface
waters had adversely affected biota in a stream adjacent to a waste site and in a
nearby lake. The components of the study included chemical analyses of the
leachate, surface waters, sediments, and tissue samples; toxicity testing of the
surface waters; and surveys of the structure and composition of the biological
communities. Tissue analyses are important for determining contaminant bio-
accumulation and assessing potential human exposure through consumption of
aquatic organisms. Toxicity testing is important for determining potential lethal
and sublethal effects of contaminant exposure on aquatic biota. Although
ecological analysis of community structure and composition is also an important
component of biomonitoring, it will not be discussed here since the focus is on the
relationships between the leachate source, the distributions of contaminants near
the waste site, and the toxic effects and bioaccumulation of the contaminants in the
tissues of local fauna.
Site Description
The 5-acre facility is an industrial waste processing site which accepts wastes
from nearby plastic manufacturing and electroplating industries. Liquid wastes are
dewatered on site prior to removal to an off-site disposal area. The principal wastes
processed at the faclity include several organic compounds and metals.
The site contains a wastewater impoundment with numerous seeps and
drainage channels that transport leachate into an adjacent river (Figure 15-53). The
river flows from northeast to southwest, and is joined by a tributary stream before
15-174
-------
entering a nearby lake. The RFA indicated an oily sheen associated with a strong
chemical odor on the surface of the stream below the treatment pond, and further
reported numerous violations of the NPDES permit. Subsequent analyses of samples
taken from the drainage channels and seeps flowing into the river showed high
concentrations of organic and trace metal contaminants, principally bis(2-
ethylhexyl) phthalate, ethylbenzene, phenol, copper, cadmium, and zinc.
Sampling Program
Six stations were sampled to assess possible toxicity and bioaccumulation of
released substances (Figure 15-53). Station 6, located upstream of the release, was
selected as a reference location for the stream. Station 17 was selected as a
reference location for the lake because it is distant from the river mouth and
because prevailing winds from the northwest direct the river discharge along the
southeast shore of the lake away from the station. Stations 7, 15, and 18 were
selected to determine the extent of toxic impacts on river and lake biota.
Water, sediments, and tissues of bottom-dwelling fishes (brown bullhead
catfish. Ictalurus nebulosus) were collected at each station. Concentrations of bis(2-
ethylhexyl) phthalate, ethylbenzene, phenol, copper, cadmium, and zinc were
measured in each matrix. Analyses were conducted according to U.S. EPA guidelines
for sediments, water, and tissues. Water quality variables (dissolved oxygen,
temperature profiles, and alkalinity), total organic carbon in sediments, and lipid
content of tissues were also measured.
Three independent bioassays were conducted on each water sample. The test
species and endpoints used in the bioassays were those recommended in the U.S.
EPA protocol for bioassessment of hazardous waste sites (Tetra Tech, 1983). Growth
inhibition in the alga Selanastrum capricornutum. and mortality in the crustacean
Daphnia maana were determined using U.S. EPA (1985) short-term methods
for chronic toxicity testing. Inhibition of enzyme-mediated luminescence in the
bacterium Photobacterium phosphoreum (i.e., the Microtox procedure) was
measured according to the methods established by Bulich et a[. (1981).
15-175
-------
vrt
C
o
01
C
"5.
•o
fl
C
J2
Q_
nn
in
i
un
15-176
-------
Results
Results of the survey indicated that concentrations of organic contaminants m
the surface waters were generally less than U.S. EPA water quality criteria, but that
concentrations of inorganic contaminants generally exceeded water quality criteria
at Stations 7, 15, and 18 (Table 15-19). In comparison with the reference stations,
significant sediment contamination was evident at Stations 7, 15, and 18 for the
three trace metals (Table 15-20). Tissue concentrations of organic substances
exceeded detection limits for bis(2-ethylhexyl) phthalate at Stations 7 and 15, and
for ethylbenzene at Station 7 (Table 15-21). However, trace metal concentrations in
tissues were highly elevated at Stations 7 and 15, but only slightly elevated at
Station 18.
The bioassay data showed a considerable range in sensitivity, with the algal
bioassay being the most sensitive (Table 15-22). Consequently, the bioassay results
were normalized to the least toxic of the reference stations (i.e., Station 6) to
compensate for the wide range of sensitivity among the test species (Table 15-23).
Overall, the bioassay results showed a high degree of agreement with contaminant
concentrations in water and sediments (Figure 15-54, Table 15-19 and 15-20).
Stations 7 and 15 showed highly toxic results, and Station 18 indicated moderate
toxicity. Only the algal bioassay indicated significant, but low, toxicity at Station 17
(the lake reference station).
In summary, the results indicated that the organic contaminants were less of a
problem than the trace metals in terms of bioaccumulation and potential toxicity.
Most of the observed toxicity was attributed to trace metal contamination, which is
consistent with the elevated concentrations of trace metals measured in the water,
sediments, and tissues.
Case Discussion
This case study provides an example of a biomonitoring program designed to
characterize the relationship between a contaminant source, contaminant
concentrations in sediments and water, bioaccumulation in tissues, and receiving-
water toxicity. It should be recognized that in many instances, the relationship
15-177
-------
TABLE 15-19
MEAN CONCENTRATIONS Ug/i) OF ORGANIC SUBSTANCES AND TRACE METALS
IN LEACHATE AND SURFACE WATERS3
Chemical Class
Base Neutral
Volatile
Acid Extractable
Metals
Chemical
Bis(2-ethylhexyl)
phthalate
Ethylbenzene
Phenol
Copper
Zinc
Cadmium
Station
Seep
LI
600
100
1500
4300
35,000
4800
River
6
2
1
<1
<1
17
<1
River
7
11
1
18.37
489
4290
146
Lake
15
10
<1.
<1
56
1100
49
Lake
18
1
1
<1
26
37
<1
Lake
17
2
2
<1
2
35
<1
Water Quality
Criteria0
Acute
940
32,000
10,200
18
320
3.9
Chronic
3
NAC
2560
12
47
1.1
'River and lake alkalinity = lOOmgCaCCtyL
"Trace metal criteria adjusted for alkalinity
'Not available for this substance
15-178
-------
TABLE 15-20
MEAN SEDIMENT CONCENTRATIONS (ug/kg DRY WT) OF ORGANIC
SUBSTANCES AND TRACE METALS
Chemical Class
Base Neutral
Volatile
Acid Extractable
Metals
Chemical
8is(2-ethylhexyl)
phthalate
Ethylbenzene
Phenol
Copper
Zinc
Cadmium
Station
Seep
LI
NA1
NA
NA
NA
NA
NA
River
6
216
10
<30
3
11
<0.1
River
7
1188
34
<30
1663
28,314
19
Lake
15
1080
20
<30
190
7260
6
Lake
18
108
14
<30
88
24
<0.1
Lake
17
216
8
<30
7
23
<0.1
aNot applicable (NA).
15-179
-------
TABLE 15-21
MEAN LIVER TISSUE CONCENTRATIONS (ug/kg WET WT) OF ORGANIC
SUBSTANCES AND TRACE METALS
Chemical Class
Base Neutral
Volatile
Acid Extractable
Metals
Chemical
Bis(2-ethylhexyl)
phthalate
Ethylbenzene
Phenol
Copper
Zinc
Cadmium
Station
Seep
LI
NA'
NA
NA
NA
NA
NA
River
6
<25
<5
<30
118
983
115
River
7
95
9
<30
1600
28,400
1600
Lake
15
86
<5
<30
750
8500
639
Lake
18
<25
<5
<30
237
2139
190
Lake
17
<25
<5
<30
180
1420
125
aNot applicable (NA).
15-180
-------
TABLE 15-22
MEAN LC50 AND EC50 VALUES (PERCENT DILUTION) FOR SURFACE-WATER
BIOASSAYS3
Bioassay
Algae
Daphnia
Microtox
Endpoint
Growth inhibition
(EC50%)'
Mortality (LC5o%)a
Decreased
luminescence
(EC5o%)*
Station
Seep
L1
NA°
NA
NA
River
6
>100C
>100
>100
River
7
0.4
3.3
5.6
Lake
15
10.0
18.5
15.0
Lake
18
24.9
100.0
43.4
Lake
17
75.0
90.0
>100
aPercent dilution required corresponding to a 50 percent response
"Not applicable (NA) because leachate toxicity was not tested
'Response of > 100 indicates that samples were not toxic at all dilutions tested
"Percent dilution corresponding to 50 percent mortality
15-181
-------
TABLE 15-23
RELATIVE TOXICITY OF SURFACE-WATER SAMPLES8
Bioassay
Algae
Daphnia
Microtox
End point
Growth inhibition
(EC50%)
Mortality (LCso%)
Decreased
luminescence
(EC50%)'
Station
Seep
L1
NA°
NA
NA
River
6
0.0
0.0
0.0
River
7
99.6
96.7
94.4
Lake
15
90.0
81.5
85.0
Lake
18
75.1
0.0
56.6
Lake
17
25.0
10.0
0.0
'Relative toxicity s 100 x [(Reference Station - Impacted Station)/Reference Station]
"Not applicable (NA) because leachate toxicity was not tested
15-182
-------
i
i
•^f
g
o
§
100
90 -
80
70
60
50
40
30
20
10
0
\
Rfvw?
18
/\
17
Stnoon
100
90 -
80 -
70
80
50
40
30
20
10 -
0
Rfv«r9
zz
I
I
19
Station
Dophnio
Lota* 18
Mleratox
Figure 15-54. Bioassay Responses to Surface Water Samples
15-183
17
-------
between contaminant concentrations in the water and toxicity will not be as clear-
cut as described in this example. Consideration of the chemical composition in
leachate samples, mass balance calculations, and transport and fate mechanisms
may indicate that sediments are the primary repository of contaminants. In such
instances, sediment bioassays rather than receiving-water bioassays may be better
suited for characterization of potential toxic effects on local fauna.
References
Bulich, A.A., M.W. Greene, and D.L Isenberg. 1981. Reliabiltv of the bacterial
luminescence assay for determination of the toxicitv of pure compounds and
complex effluent, pp. 338-347. In: Aquatic toxicology and hazard assessment.
Proceed ings of the fourth annual symposium. ASTM STP 737. D.R. Branson and K.L
Dickson (eds). American Society for Testing and Materials, Philadelphia, PA.
TetraTech. 1983. Protocol for bioassessment of hazardous waste sites. EPA-600/2-
83-054. Lafayette, CA. 42 pp. + appendices.
U.S. Environmental Protection Agency. 1985. Short-term methods for estimating
the chronic toxicitv of effluents and receiving"waters to freshwater organisms.
EPA/600/4-85/014. U.S. EPA, Environmental Monitoring and Support Laboratory,
Cincinnati, OH. 162pp.
15-184
-------
CASESTUDY29: SAMPLING OF SEDIMENTS ASSOCIATED WITH SURFACE
RUNOFF
Point Illustrated
• Contaminated sediments associated with surface runoff pathways
(rivulets or channels) are indicative of the migration of chemicals via
overland flow.
Introduction
This facility is a secondary lead smelting plant which began operation in 1976.
The plant reclaims lead from materials such as waste automotive batteries,
byproducts of lead weight manufacture, and wastewater sludges. Lead grid plates
from salvaged batteries are temporarily stored on site in an open pile prior to being
re-melted. It is therefore appropriate to conduct some form of runoff sampling to
monitor migration of contaminants from the site via this route.
Facility Description
The facility covers approximately 2,000 ft? and is situated in an area primarily
used for farming. A creek flows adjacent to the plant and drains into a major river 6
miles west of the site. Population is sparse with the nearest town 4 miles to the
south. In the past, there have been four on-site impoundments in operation and
two landfills. In addition, blast furnace slag, lead grid plates, and rubber chips from
the recycled batteries have been stored in two on-site waste piles.
Sediment Sampling
Four sediment samples (020, 022, 025, and 027) were collected from surface
runoff pathways and a creek which receives runoff from the site. Figure 15-55
shows the locations of the runoff pathways relative to the facility and the four
sampling points. Additional sediment samples were collected from the creek at
various points upstream and downstream of known overland leachate seeps and
surface water runoff routes. The program design enabled comparison between
15-185
-------
. CLOSED
IMPOUNDMENT 3
CLOSED SLAG
STORAGE AREA —
CLOSED RUBBER CHIP
STORAGE AREA
CLOSED SLAG
STORAGE AREA
• DRILL HOLES
WASTE AREA
WELLS
FEET
200
Figure 15-55. Surface Water and Sediment Sample Locations
15-186
-------
concentrations at different sections of the creek and background locations in
relation to the runoff pathways.
Results
Table 15-24 presents the concentrations of lead and arsenic measured on the
four surface runoff pathways and at location 029, which represents an upstream
background concentration (Figure 15-55). It is clear that highly elevated levels of
lead were detected in all four of the runoff pathway samples. The highest
concentration of lead, 1,900 ppm, was detected in the western-most portion of the
site. Runoff pathway sediment at the northern end of the facility, adjacent to the
slag storage area, recorded 1,600 ppm of lead. Concentrations of this order
represent a substantial source of sediment contamination.
TABLE 15-24
ARSENIC AND LEAD CONCENTRATIONS (PPM) IN RUNOFF
SEDIMENT SAMPLES
Sampling Location
Contaminant
Arsenic
Lead
Case Discussion
This case illustrates the importance of monitoring surface runoff pathways,
because they can represent a major route of contaminant migration from a site,
particularly for contaminants likely to be sorbed on or exist as fine particles. This
type of monitoring is especially useful for units capable of generating overland
flows. Such monitoring can establish the need for corrective.measures (e.g., surface
runon/runoff controls and/or some form of waste leachate collection system).
#020
11.0
1300
#022
9.6
1900
#025
2.0
1600
#027
8.9
1700
Background
#029
<0.1
11.0
15-187
-------
CASE STUDY 30: SAMPLING PROGRAM DESIGN FOR CHARACTERIZATION OF A
WASTEWATER HOLDING IMPOUNDMENT
Points Illustrated
• Sampling programs should consider three-dimensional variation in
contaminant distribution in an impoundment.
• Sampling programs should encompass active areas near inflows and
outflows, and potentially stagnant areas in the corner of an
impoundment.
Introduction
This study was conducted to assess whether an active liquid waste
impoundment could be assumed to be of homogenous composition for the purpose
of determining air emissions. This case shows the design of an appropriate
sampling grid to establish the three-dimensional composition of the impoundment.
Facility Description
The unit being investigated in this study is a wastewater impoundment at a
chemical manufacturing plant. The plant primarily produces nitrated aromatics and
aromatic amines. Raw materials include benzene, toluene, nitric acid, and sulphuric
acid. Wastewater from the chemical processing is discharged into the
impoundment prior to being treated for release into a nearby water body. The
impoundment has an approximate surface area of 3,750 m2 and a depth of 3 m.
Sampling Program
For the most part, sampling involved the collection of grab samples using an
extended reach man-lift-vehicle. The program was designed to collect samples at
different locations and depths in the impoundment.
15-188
-------
Sampling Locations and Procedures--
Samplinq Grid - The wastewater impoundment was divided into 15 segments
of equal area. Within this grid, eight sampling locations were selected which
included all pertinent areas of the impoundment, such as active portions near the
inflows and outflows, potential stagnant areas in the corners, and offshore points
near the center line of the impoundment.
It was decided to take samples from four depths in the liquid layer and one
from the bottom sediments at each of the eight locations. Figure 15-56 shows the
impoundment schematic and sampling locations.
Liquid Sampling - A total of 32 liquid grab samples were taken. These were
analyzed for the following parameters: all identifiable volatile organic compounds
(VOCs) and semivolatile organic compounds (SVOCs) using gas chromatograph/mass
spectroscopy; and selected VOCs and SVOCs by gas chromatography using a flame
ionization detector.
Sediment/Sludge Sampling - The bottom layer was sampled using a Ponar grab
sampler. The same analyses were performed on the eight sediment/sludge samples
as on the liquid samples.
Meteorological Monitoring - The ambient meteorological conditions were
monitored throughout'the sampling period, including wind speed, wind direction,
and air temperature. A video camera was also used to record the movement of
surface scum on the impoundment.
Table 15-25 summarizes the sampling locations and analyses, including
locations where QC data were collected.
Results
From the sampling program, it was discovered that approximately 99 percent
of the organic compounds (by weight) were contained in the bottom sludge layer.
15-189
-------
117'
343'
D
A
(*
H
E 0
© - PROPOSED SAMPLING LOCATIONS
O
X
o
Ul
EXISTING
N
PLANT SUMP EFFLUENT
INFLUENT/LAGOON
EFFLUENT
BOILER BACK WASH EFFLUENT
PRIMARY PLANT EFFLUENT
Figure 15-56. Schematic of Wastewater Holding Impoundment Showing
Sampling Locations
15-190
-------
TABLE 15-25
SUMMARY OF SAMPLING AND ANALYSIS PROGRAM FOR
WASTEWATER IMPOUNDMENT
Location
A-1
A-2
A-3
A-4
A-5
8-1
8-2
8-3
B-4
8-5
C-1
C-2
C-3
C-4
D-1
D-2
D-3
D-4
D-5
DeQth
(Feet)
0-1
2
4
6
Bottom
Sediment
0-1
2
4
6
Bottom
Sediment
0-1
2
4
Bottom
Sediment
0-1
2
4
6
Bottom
Sediment
Sample Analyses
GOFID
VGA
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
GGMS
VGA
X
X
X
X
X
X
X
X
TOC
X
X
X
X
X
X
X
X
POC
X
X
X
X
Onsite
Parameters*
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
GGFID
svoc
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
GC/MS
svoc
X
X
X
X
X
X
X
X
a Includes pH, turbidity, specific conductance, and dissolved oxygen measurements.
X Indicates locations where QC samples were collected.
15-191
-------
TABLE 15-25 (continued)
Location
E-1
E-2
E-3
E-4
E-5
F-1
F-2
F-3
F-4
F-5
G-1
G-2
G-3
G-4
G-5
H-1
H-2
H-3
H-4
H-5
Depth
(Feet)
0-1
2
4
6
Bottom
Sediment
0-1
2
4
6
Bottom
Sediment
0-1
2
4
6
Bottom
Sediment
0-1
2
4
6
Bottom
Sediment
Sample Analyses
GC/FID
VOA
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
GC/MS
VOA
X
X
X
X
X
X
X
X
TOC
X
X
X
X
X
X
X
X
POC
X
X
X
X
Onsite
Parameters*
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
GGFID
svoc
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
GC/MS
svoc
X
X
X
X
X
X
X
X
a Includes pH, turbidity, specific conductance, and dissolved oxygen measurements.
X I nd i cates I ocati ons where QC sam pi es were col I ected.
15-192
-------
Vertical and horizontal variation in the composition of the lagoon was apparent.
The degree of horizontal variation was relatively small, but sample point "A"
showed consideraby higher concentrations of 2,4-dinitrophenol than the other
locations. This could have resulted from a recent discharge from the outflow at the
southern end of the impoundment. Vertical variation in composition showed a
general trend of increasing concentration with depth, but certain chemicals tended
to have higher concentrations at mid-depth in the impoundment.
Case Discussion
This case provides an example of a sampling program at an areal source
designed to yield accurate information for characterizing air emissions from the
unit. The study illustrated the importance of characterizing the organic
composition of the lagoon in three dimensions and considering variations resulting
from inflow and outflow areas.
It should be mentioned that this study did not consider variation in the
chemical composition of the impoundment with time. To obtain tLs information, it
would be necessary to conduct subsequent sampling programs at different times.
From this study, it is apparent that chemical composition varies both horizontally
and vertically, and is likely to change depending on inflows and outflows of wastes.
This sampling program is therefore limited to effectively characterizing composition
at a single point in time.
15-193
-------
CASE STUDY 31: USE OF DISPERSION ZONE CONCEPTS IN THE DESIGN OF A
SURFACE WATER MONITORING PROGRAM
Point Illustrated
• Estimation of the dispersion zone of contaminants downstream of a
release point can be used to help design a surface water monitoring
program.
Introduction
When a contaminant is initially released to a body of water, the concentration
of the contaminant will vary spatially until fully dispersed. In streams, the
contaminant will disperse with the surrounding ambient water as the water moves
downstream and will eventually become fully dispersed within the stream.
Downstream of this point, the contaminant concentration will remain constant
throughout the stream cross-section, assuming that streamflow is constant and that
the contaminant is conservative (e.g., nondegradable). The area in which a
contaminant's concentration will vary until fully dispersed, referred to here as the
dispersion zone, should be considered when determining the number and location
of sampling stations downstream of the release point.
Facility Description
A facility that processed zinc, copper and precious metals from ores operated
along a stream for five years. The plant was closed after being cited for repeated
fish kills, reportedly due to failures of a tailings pond dike. At present, the site is
covered with tailings containing high concentrations of copper, zinc, cadmium,
arsenic, and lead. There is no longer a tailings pond. This is the same facility
described in Case Study 27.
Site Setting
The site is located on coarse colluvium (hill-slope deposits of weathered
bedrock) and fine-grained alluvium. These deposits are typically 50 feet thick.
Metamorphic rock (phyllite) underlies the unconsolidated materials. Ground water
15-194
-------
moves laterally in the gravel formations from the steep valley walls toward the
stream.
The site is located about 400 feet from the stream. Two drainage ditches cross
the lower portion of the site and merge prior to leaving the site. The ditch carries
the combined flow and discharges directly into the stream (Figure 15-57). No other
tributaries enter the stream within 2 miles of this location. Downstream of the
release point, stream width and depth remain fairly constant at 45 and 3 feet,
respectively. Mean stream velocity is 0.5 feet per second and channel slope is 0.0005
feet per foot.
Sampling Program
A surface water monitoring program was designed as part of a Phase I
investigation to determine the extent of contamination in the stream. Existing data
from previous sampling had shown high concentrations of metals in the drainage
ditch sediments (e.g., 5,170 mg/kg Cu and 11,500 mg/kg Zn). Ground-water data
from the plant's well showed concentrations of Cu (7 ug/0 and Zn (54 ug/i). The
contribution of metals to the stream by ground-water discharge was considered to
be relatively minor.
Based on a review of the plant history and the available water quality and
sediment data, a monitoring program was designed. The potential pathways by
which metals could reach the stream appeared to be direct discharge from the
drainage ditch, discharge of contaminated ground water, and storm water runoff
over the general facility area. Plant records indicated that typical flows in the
drainage ditch at its confluence with the stream varied from 1 to 3 cubic feet per
second (cfs) in the spring. During extreme flood conditions, the flow in the ditch
exceeded 20 cfs. In the summer, flows in the drainage ditches at all locations were
less than 0.5 cfs. Resuspension of contaminated sediments in the ditches during
storm runoff appeared to be the most likely pathway for metals to reach the
stream. The specific metals of concern were identified as As, Cd, Cu, Pb and Zn,
based on the processes used at the plant and the composition of the ores which
contained some arsenopyrites (with As, Cu), galena (Pb), and sphalerite (with Zn,
Cd).
15-195
-------
ArMof
Fonw
Tailings
Pond
Site Operation*
Drainage Ditch
Sampling Station
N
t
Scale
n
160 tot
Figure 15-57. Sampling Station Locations for Surface Water Monitoring
* Located approximately 1030 feet downstream of the confluence of the ditch
with the stream.
15-196
-------
The available soil and water quality data from previous sampling were
reviewed to help determine the likely fate of the metals. The pH of soils in the area
is about 6.5 and they contain about 0.5 percent organic matter by weight. Under
such conditions, the metals, particularly Pb, would be expected to adsorb onto the
soil particles. In the on-site tailings piles, the pH of core samples ranged between
3.3 and 4.9. Low soil pH values had been measured in sediments in the drainage
ditch just downgradient of the tailings pile. The pH of the stream during the
previous sampling was 6.9. The suspended solids concentration was 10 mg/l.
Estimates of the distribution of metals between the dissolved and adsorbed
phases for a range of partition coefficients (Kp) are shown in Table 15-26. For
example, if Kp = 104 and the suspended solids concentration was 10 mg/l, 90
percent (0.9) of the metal present would be in the dissolved phase. This information
indicated that even though a metal (e.g., lead) was known to strongly sorb, a
significant amount could still be transported in the dissolved phase. Thus, both
water and suspended solids should be analyzed for metals. The complete list of
parameters selected for measurement in the Phase I investigation and the rationale
for their selection are outlined in Table 15-27.
The sampling stations were selected to determine stream water quality up-
and downstream of the site and to determine whether particulates with sorbed
metals were deposited on the stream banks or streambed. The sampling stations
and the rationale for their selection are listed in Table 15-28. The station locations
are shown in Figure 15-57. Because floods were considered a cause of
contamination incidents, samples were to be collected under both high and low
flow conditions.
The location of the downstream station (S8) was determined after estimating
the stream length that may be required for complete dispersion of the
contaminants. The following equation was used for this estimation:
0.4 w2u
DZ =
O.Sd-^gds
15-197
-------
TABLE 15-26
RELATIONSHIP OF DISSOLVED AND SORBED PHASE CONTAMINANT
CONCENTRATIONS TO PARTITION COEFFICIENT AND SEDIMENT
CONCENTRATION
Kp SS Cw/Cia
10o
101
102
103
104
1
10
100
1000
10,000
1
10
100
1000
10,000
1
10
100
1000
10,000
1
10
100
1000
10,000
1
10
100
1000
10,000
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
0.9
1.0
1.0
1.0
0.9
.0.5
1.0
1.0
0.9
0.5
0.1
1.0
0.9
0.5
0.1
0.0
After Mills et §[.,1985.
•The fraction dissolved (Cw/CT) is calculated as follows:
C 1
CT 1+KpxSxKH
where Kp = partition coefficient, f/kg
SS = suspended solids concentration, mg/l
Cw = Dissolved concentration
CT = Total concentration
15-198
-------
TABLE 15-27
PARAMETERS SELECTED FOR SURFACE WATER MONITORING PROGRAM
Parameters
Rationale
Metals - As, Cd, Cu, Pb, Zn
PH
Dissolved Oxygen, Sulfide, Fe(ll),
Fe(lll)
Alkalinity
Total Dissolved Solids
Major Cations (Ca * 2, Mg * 2, Na *, K *,
NH4*)and
Major An ions (CI-, SO4-2( NO-3)
Suspended Solids
Streamflow
Determine extent of contamination
Predict sorption behavior, metal
solubility, and speciation
Determine redox conditions which
influence behavior of metals,
particularly the leaching of tailings
A measure of how well buffered a water
is, allows consideration of the likelihood
ofpH change
Used as a water quality indicator and for
QA/QC checks
May identify other waste sources, can
influence fate of trace metals
Predict the fraction of metal in water
which issorbed
Compute mass balances and assist in
identifying sources of observed
contamination
15-199
-------
TABLE 15-28
SELECTED SURFACE WATER MONITORING STATIONS AND SELECTION RATIONALE
Station
Media
Rationale
Drainage ditch west of
site(S1)
Drainage ditches on site
(S2 and S3)
Downstream of
confluence of two
ditches (54)
Mouth of drainage
ditch (S5)
Stream (S6, S7 and 59)
Stream (S8)
Water and sediments
Determine whether off-site
drainage is significant source of
contamination
Water and sediments Identify on-site sources
Water and sediments
Water, suspended
sediment, bedload
Water, suspended
sediment, bedload
Water, suspended
sediment, bedload
Provide information for
checking mass balances from the
two drainage ditches
Determine quality of direct
discharge to stream
Determine upstream water
quality
Determine quality downstream
of site following complete
dispersion and provide data for
mass balance
15-200
-------
•
where:
DZ - dispersion zone length, ft
w = width of the water body, ft (45 ft)
u = stream velocity, ft/sec (0.5 ft/sec)
d = stream depth, ft (3 ft)
s = slope (gradient) of stream channel, ft/ft (0.0005)
g = accerleration due to gravity (32 ft/sec 2).
Using the above equation, the estimated stream length required for complete
contaminant dispersion is 1030 feet. This can serve as an approximate distance
downstream of the release point at which a sampling station should be located.
Case Discussion
This case illustrates the use of contaminant dispersion zones in the design of a
surface water monitoring program. In this example, the data indicate that
approximately 1030 feet of flow within the described stream d innel is required
before a contaminant will become fully dispersed. A downstream station should
therefore be located at or below this dispersion zone to fully characterize the
extent of the release. An adequate number of sampling stations should also be
located upstream of this point.
15-201
-------