EPA 910/9-88-235
copy 2
SUMMARY OF HISTORICAL PUGET SOUND
CONTAMINANT MASS LOADING ANALYSIS
SUBMITTED TO:
U.S. ENVIRONMENTAL PROTECTION AGENCY
REGION 10
PREPARED BY
COOPER CONSULTANTS, INC.
1750 - 122TH N.E., SUITE C-225
BELLEYUE, WASHINGTON 98004
AND
ENVIROSPHERE COMPANY
10900 - 8TH STREET
BELLEYUE, WASHINGTON 98004
OCTOBER, 1985
8601A
-------
TABLE OF CONTENTS
Page
1.0 INTRODUCTION 1
2.0 POLLUTANT LOADING ESTIMATES DEVELOPED FOR PUGET SOUND . . 2
2.1 WATER QUALITY MANAGEMENT PROGRAM FOR PUGET SOUND . . 2
2.2 TOXICANT PRETREATMENT PLANNING STUDY (TPPS) 2
2.3 TOXIC CHEMICALS AND BIOLOGICAL EFFECTS
IN PUGET SOUND 7
3.0 INTERPRETIVE EVALUATION 11
8601A
-------
TABLE OF CONTENTS (Continued)
LIST OF TABLES
Table
No. Page
1 ESTIMATED HEAVY METAL MASS BALANCE FOR CENTRAL PUGET
SOUND (ROMBERG ET AL. 1984) 3
2 ESTIMATED ORGANIC MASS BALANCE FOR CENTRAL PUGET SOUND
USING ARITHMETIC MEANS (ROMBERG ET AL. 1984) 5
3 ESTIMATED ORGANIC MASS BALANCE FOR CENTRAL
PUGET SOUND USING SOME GEOMETRIC MEANS
(ROMBERG ET AL. 1984) 6
4 ESTIMATED TRACE METAL INPUTS INTO PUGET SOUND
(QUINLAN ET AL. 1985) 8
5 ESTIMATED INPUTS OF SELECTED ORGANIC CONTAMINANTS
INTO PUGET SOUND 9
6 COMPARISON OF METAL LOADINGS CALCULATED FOR
CENTRAL PUGET SOUND 12
8601A
ii
-------
1.0 INTRODUCTION
Effective management decisions concerning the control of contaminant
discharges into Puget Sound require identification of the contributing
sources and the quantification of the contaminant mass loading
associated with each source.
The objectives of this report are to:
o Develop an updated summary of the available historical
information generated by a number of recently completed
studies.
o Point out the differences and major difficulties encompassed
in performing mass loading calculations.
o Provide, to the extent possible, the reasons for the
discrepancies in these studies.
The discussion which follows presents the mass loading summaries
developed in these historical investigations and provides an
interpretive evaluation of this information within the context of the
objectives listed above.
8601A
-------
2.0 POLLUTANT LOADING ESTIMATES DEVELOPED FOR PUGET SOUND
The discussion in this section summarizes the contaminant mass loading
in Puget Sound estimated in three historical studies. The mass loading
analysis presented in the following reports are discussed: Water
Quality Management Program for Puget Sound (Jones and Stokes, 1983);
Metals/Toxicants Pretreatment Planning Study (Romberg et al., 1984);
and Toxic Chemicals and Biological Effects in Puget Sound (Quinlan et
al., 1985).
2.1 WATER QUALITY MANAGEMENT PROGRAM FOR PUGET SOUND
This study presented wet season, dry season, and annual estimates of
loadings associated with NPDES permitted municipal and industrial
dischargers. Pollutant mass loadings associated with nonpoint sources
were not reported.
Data were generally not available for all permitted discharges to a
geographic basin. The information reported was primarily for
conventional pollutants; priority pollutant data was sparse. Due to
these limitations, total pollutant loading and the relative importance
of each source was not assessed in the Jones and Stokes report. For
these reasons no data from this report are presented in the summary.
2.2 TOXICANT PRETREATMENT PLANNING STUDY (TPPS)
In the TPPS report (Romberg et al. 1984), the central basin of Puget
Sound was the primary focus of the contaminant mass balance estimates.
Mass loading data were presented for both trace metals and synthetic
organic components.
Estimates of heavy metals provided general information on the relative
contribution of various sources. The mass balance for these
constituents is presented in Table 1. By far the dominant mass of most
metals was associated with the large amount of marine water that moves
in and out of Puget Sound. Overall, little excess metal load from
8601A
-------
Table 1 ESTIMATED HEAVY METAL MASS BALANCE FOR CENTRAL PUGET SOUND
Sources
Inputs3
Marine Waters/Advection
Non-Point Source
River Drainage
Shoreline Erosion
Atmospheric Inputs
Point Sources
Industrial
Municipal
CSOs
Dredge Disposal
TOTAL INPUTS
Outputs8
Advection
Sedimentation
TOTAL OUTPUTS
VARIANCE6
3A11 input and lose terms were
bv/arlanno in Hoflnorl >< . Total
Mass Loading
As
390(80)
28(6)
58(12)
3(.6)
5(1)
K.2)
*0 . 1
'0.1
485
400(95)
21(5)
421
+ 13
estimated as
Input-Total
Cu
75(39)
49(25)
12(6)
6(3)
32(17)
17(9)
<1
2(1)
193
157(70)
68(30)
225
-17
discussed
Output ,
in mt/year (percent of total in parenthesis)
Pb
50(33)
30(20)
7(5)
40(26)
4(3)
15(10)
<1
4(3)
151
46(37)
79(63)
125
+ 17
in the
nn
Hg
0.6(67)
0.2(22)
NDAC
<0.1
0.1(11)
'0.1
0
*Q.l
0.5(55)
Q.4(45)
0.9
+0.3
text.
Ag
3.33
2(22)
'1(11)
*0.1 ( 1)
1(11)
2(22)
'0.1
< 9
3(60)
2(40)
5
Zn
510(71)
89(12)
46(6)
4
24(3)
34(5)
<0. 1
"am
715
570(75)
194(25)
764
+ 44 -7
Total
Metala
1029(66)
198(13)
123 (8)
69(4)
66(4)
54(3)
<2
14(1)
1554
1176(76)
365(24)
1541
+ 1
Total Input
NDA=No data available.
-------
land-based sources was carried out of the Sound. The majority of the
land-originating metals appeared to be retained in the central basin
via sedimentation. These land-based sources were dominated by rivers
and shoreline erosion. The total anthropogenic inputs to the central
basin represent the following relative contribution to the total annual
metal loadings: 43 percent for silver, 42 percent for lead (probably
much higher if the surface runoff is combined with the riverine
discharges), 16 percent for copper, 11 percent for mercury, 10 percent
for zinc, and 2 percent for arsenic.
Attempts to estimate a mass balance for organic compounds in the
central basin were unsuccessful. In virtually all cases, estimated
inputs were extremely small, even insignificant in comparison with the
estimated output. This extreme variance was unexpected since the same
method was used for calculating both the metal and organic mass
balances. Possible explanations are:
(1) The organics data were far more variable than the metals data,
due to large environmental patchiness; this indicates that
more sampling is required to obtain a representative
population of the ambient concentrations under various spatial
and temporal coverage;
(2) that assumptions about compound stability, sedimentation, and
water transport were in error; and
(3) that there was some large unknown source of organic
toxicants.
The preliminary organic chemical loading estimates and mass balances
prepared using arithmetic and geometric means are presented in Tables 2
and 3.
The TPPS loading estimates were also based upon limited data,
especially for nonpoint sources. As stated above, data were incomplete
for NPDES permitted dischargers, particularly for the priority
8601A
-------
TABLE 2 ESTIMATED ORGANIC MASS BALANCE FOR CENTRAL PUGET SOUND USING ARITHMETIC
MEANS (ROMBERG ET AL., 1984)
SOURCES
Acid
Base
Neutrals
Pesticides
Volattleg
PA,H CPAH CLA CBD Phth ni-Octyl DDT PCB MISC.
INPUTS
Advection .024
(Effectlon Oceanic Loading)
6.16
59.4 2424
Non-Point Sources
River Drainage
Shoreline Erosion
Atmospheric Inputs
Point Sources
.211
.416 .463 .009 .009 10.24
.066 .0020 .132
.021 .024 .042 .54
Industrial
Municipal
CSOs
Dredge Disposal
Subtotal Estimated
Inputs
Unknown/Unaccounted
forc
Advection
Sedimentation
Total Outputs
Variance (X)b
^All units in mt/yr"1
^Variance is defined as:
.08
.66
.003
.0004
.978
12.05
12.99
.04
13.03
Lge-
Total
0.13
.0005
.0135
11.16
11.05
.13
11.18
Lge-
1
8
82
88
1
90
.039
.62
.004
.031
.27
.36
.8
.83
.63
Lge-
Input - Total
.115
.192
1.105
29.69
27.11
3.69
30.80
Lge-
Output x
.001
.1904
.0026
.203
11.10
11.27
.03
11.30
Lge-
100
.0009 .
~ 11.
.069
.00015 .
.07005 17
137
206
.006 1
.006 208
004
62 3.78
069 .015
0068 .058
.1088 2348.09
.58
.83 183.03
.85 12.2
.68 195.23
Lge+ Lge- Lge+
.0008 .057
.00005 .001
.020 .011
.1078 .0958
2.794
.00157 2.49
.005 .4
.0066 2.89
Lge+ Lge-
1.
17.
.
.00018
.1742 20.
136079
.066 136100
.0031
.0691 136100
B
74
234
024
338
.66
.
.49
.
Lge-t- Lge-
Total Input
cUnknown/unaccounted for term is employed to attempt a balance with estimated outputs,
-------
TABLE 3 ESTIMATED ORGANIC MASS BALANCE FOR CENTRAL PUGET SOUND USING SOME
GEOMETRIC MEANS (ROMBERG ET AL., 1984)
Mass Loadings in mt/year
Neutrals
Sources
Inputs8
Marine Hatera/Advectiond
Non-Point Sources
River Drainage
Shoreline Erosion
Atmospheric Inputs
Point Sources
Industrial
Municipal
CSOs
Dredge Disposal
Subtotal Estimated Inputs
Unknown/Unaccounted forb
Outputs8
Advectiond
Sedimentation'1
TOTAL OUTPUTS
VARIANCE (X]c
aAll input and loss terms
^unknown/Unaccounted for
cVariance is defined as:
Acids
.022
.211
0.08
.66
0.0003
0.0004
< 1
10
11.0
.022
11.02
I-qe-
Bases
_
.013
.0005
< .02
2.8
2.87
.0066
2.88
f,qe-
are estimated based
term is employed to
Subtotal
PAHs CPAIIs
0.0022 -
.416
.039
1.62
.004
.031
2.11 1.
6 5
7.96 2.
.246 3.
8.21 6.
I.qe- L
-
463
335
115
192
1
87
39
26
qe-
CLAs
.009
.001
.190 11
.0026
.203 20
3
PIITHs
8.4
.009
.004
.6
.069
.0068
.1
3.31 14.1
.0068 .524
3.32 14
Lqe-
on available data and
attempt a balance with
Ijlput -Total Output
x 100
.
.6
Med +
Pesticldes/PCns
DI-OCTYL DDTa
1400 0.022
10.2 .021
___ .
3.78 .0008
.015 .00005
.058 .020
1414 .064
89.7 .00066
2.93 .0023
92.6 .0030
Lqe-t- Lqe +
scientific assumptions.
estimated outputs.
PCBa
0.0022
.024
.057
jOOl
.011
.095
.55
.464
.185
.649
Lqe-
Misc Volatile
0.044 748
.042 .54
1.8
18.
.23
.00018 .024
.086 769
.00082 2B7
.011 .0046
.012 287
Lqe+ I,qe +
dValues based on geometric mean values for whole water and sediments. All other loadings are same as the original
detailed values used to develope the simplified version presented in Table 41 of the main text.
-------
pollutants. However, in the TPPS program, detailed data were obtained
for the METRO municipal discharges. Incorporation of these data and
efforts to estimate nonpoint source contributions resulted in
substantially improved estimates of pollutant loading to central Puget
Sound over the Jones and Stokes report.
2.3 TOXIC CHEMICALS AND BIOLOGICAL EFFECTS IN PUGET SOUND
Quinlan et al. 1985 developed mass loading estimates for five source
categories and three major classes of chemical contaminants, including
trace metals, polynuclear aromatic hydrocarbons (PAHs) and
polychlorinated biphenyls (PCBs). Sources included: rivers, shoreline
erosion, atmospheric deposition, municipal sewage effluents and
industrial waste effluents. Smaller sources, including combined sewer
overflows and storm drains, small industrial dischargers and other
nonpermitted discharges and spills were not included. The loading
estimates computed for five Puget Sound subregions are presented in
Tables 4 and 5.
Although available data for the selected contaminant groups are
sufficient to develop quantitative loading estimates, a number of
limitations to the data were identified. For the trace metals,
literature values for mass inputs exist for most of the sources
considered; however, samplings of many of these sources are very
limited.
For the PAHs, data are available for many of the sources, providing a
usable data set, but the majority of this data was unsupported by
replicate samplings to allow for the establishment of precision and
temporal consistency of measurements. In addition, because many of the
PAH measurements were made at or very near the detection limits of the
instrumentation available, considerable uncertainty must be accorded
the PAH source data.
8601A
-------
TABLE 4
ESTIMATED TRACE METAL INPUTS INTO PUGET SOUND (mt/yr)
(excluding advective fluxes)
Source
Type
METAL
As
Cd
Cr
Cu
Pb
Hg
Ag
Zn
Rivers
64
19
89
108
55
4
4
384
TOTAL METALS 726
]_/ ND =
SOURCE:
no data.
Quinlan et al .
Shoreline
34
17
68
75
54
3
1
305
557
1985, Table
Atmospheric
11
0.5
ml/
32
121
0.1
0.1
27
192
23, p. 74.
Municipal
1.5
1.5
16
24
25
0.2
3
51.0
119
Industry
63
2
18
56
15
0.1
2
47
203
8601A
-------
TABLE 5
ESTIMATED IMPACTS OF SELECTED ORGANIC CONTAMINANTS
INTO PUGET SOUND (mt/yr)
5a: ESTIMATED CPAH INPUTS TO PUGET SOUND (mt/yr)
(excluding advective fluxes)
Riverine
Shoreline
Atmospheric
Municipal
Industrial
Whidbey
Basin
1.69
0
0.17
0.011
NQ
Main
A
0.43
0
0.28
0.33
NQ
Basin
B
0.46
0
0.34
0.12
NQ
Southern
Sound
0.10
0
0.76
0.012
NQ
Hood
Canal
0.09
0
0.62
0
NQ
Straits
2.49
0
0.50
0.024
NQ
5b: ESTIMATED PCB INPUT TO PUGET SOUND (mt/yr)
(excluding advective fluxes)
Whidbey
Basin
Main
A
Basin
B
Southern
Sound
Hood
Canal
Strait
Juan de
inner
of
Fuca
outer
Riverine 0.053 0.009 0.024
Municipal 0.019 0.282 0.06
0.004 0.003 0.003 0.047
0.021 negligible 0.014 0.019
]_/ Total dissolved and particulate.
NQ Insufficient data available to quantify loading.
SOURCE: A - Quinlan et al., 1985, Tables 24 and 25, p. 76-77.
B - Romberg et al., 1984
8601A
-------
For the PCBs, measured values are available for only a few rivers and
some municipal discharges. PCBs, like the PAHs, occur at very low
levels, and most source measurements have not used the sophisticated
analytical procedures necessary to achieve adequate quantification. As
a result, PCBs may be present in many additional sources but have not
yet been identified. Additionally, the result that estimated inputs of
CPAH from rivers were the largest overall source calculated by Quinlan
et al. appears to be questionable given the lack of major known sources
to the rivers themselves (Quinlan et al. 1985, page 76). A similar
caution seems warranted for the atmospheric inputs which also appear to
be large in nonurban areas. These limitations reported by Quinlan
et al. (1985) are consistent with the discussion provided by Romberg
et al. (1984) and provide additional rationale for explaining the large
variance observed in the mass balance estimates attempted in the TPPS
study.
8601A
10
-------
3.0 INTERPRETIVE EVALUATION
The total Puget Sound metals loading data reported by Quinlan et al.
(1985) indicate that metal inputs are dominated by riverine and
shoreline erosion consistent with the findings of Romberg et al. (1984),
Comparison of the central basin metal loading estimates of Quinlan
et al, (1985) and Romberg et al. (1984) are shown in Table 6 indicate
that for most metals, loadings computed in the two studies are
comparable. The discrepancies for arsenic and mercury loadings
calculated for industrial sources appear to be due to the use of
different data sources and assumptions regarding the distribution
between the dissolved and particulate phase. For the loading estimates
developed by Quinlan et. al., specific industrial source values were
generally from more recent surveys. These data were considered to be
more reliable than past summary data, and were used in the industrial
loading estimates. With no data available for comparison, but assuming
the same general geochemical reactions occurred, metals in industrial
effluents were assumed to be fractionated in Puget Sound in a manner
similar to those from municipal effluents and rivers.
Atmospheric loading estimates are comparable because the same
methodology was employed in both studies. The approach used to
estimate riverine and shoreline erosion metal concentrations was
different for the two studies; thus reported metal loadings was also
different, but are within a factor of two except for arsenic. The
relatively large difference in arsenic loading between the two studies
is due to the fact that the reported soils arsenic concentration of 100
ug/g reported by Dexter et al. (1981), was used by Romberg et al.
(1984), whereas Quinlan et al. (1985) used a concentration of 10 ug/g
based upon literature values for average earth's crust composition and
observed street dust concentrations.
8601A
11
-------
TABLE 6
COMPARISON OF METAL LOADINGS CALCULATED
FOR CENTRAL PUGET SOUND
Municipal Discharge
Loading (mt/yr)
Metal
As
Cu
Pb
Hg
Ag
Zn
METRO I/
1
17
15
0.1
1
24
URS y
1.3
21.2
22.2
0.19
2.3
34
Industrial Discharge
Loading (mt/yr)
METRO I/
5
32
4
0.1
1
24
URS y
62
51
5.79
0.01
0.42
27.5
Riverine
Loading
METRO I/
28
49
30
0 .2
2
89
URS y
12
24
13.3
0.905
0.6
74
Shoreline
Erosion
METRO I/
58
12
7
NDA
1
46
URS y
8
17
12.7
0.8
0.3
69
Atmospheric
Loading
METRO I/
3
6
40
0.1
0.1
4
URS y
2.9
12.0
31.1
0.02
0.03
6.8
V Romberg et al. 1984.
2/ Qulnlan et al. 1985.
NDA No Data Available.
8601A
-------
The loadings of organic constitutents computed in the two studies
presented in Table 5 indicate that for the CPAHs, the agreement is good
between the two studies; however, for PCB's the difference in loading
estimates computed in the two studies vary in a range of two to five
times. For the municipal inputs, the discrepancy appears to be due to
different flow estimates and concentrations used by Quinlan et al.
Similarly, an average PCB value was applied to all Puget Sound rivers
to compute the loadings presented by Quinlan et al. 1985. The TTPS
values are probably more accurate because actual flow and concentration
data were used.
Summary
Available data have been used to compute preliminary contaminant mass
loading to Puget Sound. These estimates are based upon best available
data, however, they must be regarded only as approximate loading
values. The computations do indicate the relative contributions from
different sources and are graphically presented in Figures 1 and 2.
Where comparisons are possible among the different studies, mass
loading evaluations, study findings and limitations appear to be
consistent. However, data are generally not available to quantify,
with accuracy, the loadings from contaminant sources to Puget Sound.
The difficulty lies both in our ability to characterize the quantity
and quality of non-point source contributions and the limitations
associated with data available for point source discharges.
8601A
13
-------
I
_
s
t>
2
I
*
O>
c
TJ
8
eo
340
320 -
300 -
280
260 -
240 -
220 -
200 -
180 -
160 -
140 -
120 -
100 -
80
80
40
20
FIGURE 1
Total Mass Loading for Selected Metals (Mt/yr) i)
I i
I I
c o
in c
i !
! !
II
f
m
^
\
\
p
I
1) Include! A* Cu, Ph. HO, Ag, tnd Zn
2) No Municipal or Industrial Dttt Avtllablt lor Hood Can*/.
' I
|lt
1 I 2
c
l
i f
" 8
|
i
o
W-
Whldbey
Basin
Main
Basin
Southern
Sound
Hood 2)
Canal
Straits of Juan
De Fuca
-------
FIGURE 2
CD
C
TJ
03
a
S
a
n
m
o
a.
3
.o
Total PCB and PAH Mass Loading i) (Mt/yr) ?
2.5 -
2.0-
1.5
1.0-
5.
.1 i
|
2
£
PI
1
P
'' '
K
1^
i
i^
??
jii
^
^
c
&
o
. 1
ft
£ ^
'{0
'£
. "sf-
~^&,
%fr.\
1 ;' ^3 '^^
'
i
2
c
i
pm
u
1
c £ a
560
^ O P
* 5 1
£
i
i
^
s^
^
%
y^
%,
^xx
iH
i
p
p
^
J^
1
^
p
;- 'rJ-'
Q>
|
o "S
E 5-
^ i
§
%
jj
(
^
>^,
2) Southern 2) Hood 5) Straits of Juan 2)
Basin Canal De Fuca
1) Dale not available to quantify Industrial
PAH loading. PCB loading quantified
only lor municipal and riverine
sources.
2) No shoreline erosion or Industrial data
available
3) negligible (003 Ml/yr)
4) negligible (0.033 Mt/yr)
5) negligible lor municipal; no Industrial
dala available
- contribution from PCBs
. contribution from PAHs
------- |