United States Environmental Protection Agency
CBP/TRS 19/88
June 1988
903 R88120 U.S. Environmental Protection Agency
Region III Information Resource
Center (3PM52)
£41 Chestnut Street
hia, PA 19107
EPA Staff Papers
Presented at the
Chesapeake Bay
Research Conference
March 1988
TD
225 !^9Q^Pr Program
E72
Chesapeake
Bay
-------
/I .K •' *
^
, -, " U.S. Enviionmental Protection Agency
* ' •" Region ill luioimation Resource
Center (2PM52)
841 Chestnut Street
Philadelphia, PA 19107
ENVIRONMENTAL PROTECTION AGENCY STAFF PAPERS
PRESENTED AT THE
CHESAPEAKE BAY'RESEARCH CONFERENCE
MARCH 1988
Controllong Energy DissapaLlou Prom R!\/<-. .- In flow
as a Factor in Managing "stuarine Water Quality
The Role of Organic and Inorganic Carbon
in the Dissolved Oxygen Regime of the Chesapeake Bay
by
T. A. Wasler, Office of Marine & Estuarine Protection
Chesapeake Bay Sediment Monitoring for Water Quality
Model Development
by
Lewis C. Linker, Chesapeake Bay Program
Implications of Toxic Materials Accumulating in the
Surface Microlayer in Chesapeake Bay
by
Hermann Gucinskl, Anne Arundel Community College
John T. Hardy, Oregon State University
H. Ronald Preston, Chesapeake Bay Program
Potential Biological Effects of Modeled Water Quality
Improvements Resulting from Two Pollutant Reduction Scenarios
by
Kent Mountford, Chesapeake Bay Program
Robert C. Reynolds, Computer Science Corporation
-------
Controlled Energy Dissipation from River Inflow as a Factor
in Managing Estuarine Water Quality
by
T. A. Wastler, Senior Science Advisor
Office of Marine and Estuarine Protection
U.S. Environmental Protection Agency
Washington, D.C. 20460
INTRODUCTION
River inflows into estuarine systems may carry with them large
amounts of energy depending on their size and volume, and
the constant tidal action provides an additional input of energy
carried by the rhythmic movement of water into and within an es-
tuary. The potential and kinetic energy of rivers is often
tapped for hydroelectric power during their passages toward the
sea, and where tidal range and velocity are sufficient, the
energy of tidal action has been used to generate electrical power.
As the flow from a river and tidal flow from the sea meet in an
estuary they usually are moving in different directions, with the
kinetic energy of their passage also moving in opposite direc-
tions. Upon the meeting of two such inflows the kinetic energy
is rapidly dissipated, sometimes with dramatic visible effects.
For example, with an incoming tide there may be a zone of
significant wave action, i.e., a "tide rip", over a limited area
during a time of very light winds. Conversely, on an ebb tide
there may be a zone of very smooth water with a clearly visible
line of demarcation between two water bodies as the kinetic
energy from river and tide move smoothly along together. In
either case the kinetic energy from both sources is dissipated in
an uncontrolled manner with effects generally on the mixing char-
acteristics of the system, particularly on stratification and
resuspension of sediments.
-------
If the dissipation of this energy in the Chesapeake Bay could be
controlled in some manner, it might very well be possible to con-
trol the extent of stratification in parts of the Bay so as to
enhance the aquatic environment for living resources management
and public health. This investigation is a preliminary investiga-
tion of the technical feasibility of doing this in the Chesapeake
Bay. It does not address the broad issues of administrative
feasibility, social desirability, and legal responsibility, nor
the even broader scientific issue of whether or not we know
enough about the physical, chemical, and biological processes of
the Bay to be able to do this in a responsible manner.
The specific questions to be addressed in this study are:
1. Has a significant change in estuarine stratification and water
quality resulting from a river inflow change ever been
documented?
2. What physical conditions would be necessary to do this in the
Chesapeake Bay and do they presently exist?
3. Can the propagation of kinetic energy and its dissipation in
the Bay be demonstrated?
4. Is it feasible to develop a mathematical description of the
kinetic energy flow in the Bay sufficiently quantitative to serve
as the basis for regulating river discharges into the Bay?
5. Since such a management scheme would involve real time
control of river flows into the Bay, what information on
conditions in the Bay would be needed as a basis for using such
an approach?
6. Is this approach worth pursuing further, and, if so, what
must be done?
The following sections of this paper address each of these
questions in order.
HAS A SIGNIFICANT CHANGE IN ESTUARINE STRATIFICATION AND WATER
QUALITY RESULTING FROM A RIVER INFLOW CHANGE EVER BEEN
DOCUMENTED?
In the early 1940's at Charleston,S.C., a hydroelectric dam was
built near the head of tidewater on the Cooper River, the major
river input to Charleston Harbor. To feed this dam, the flow of
the Santee River was diverted to the power pool behind the
structure. The net result was an increase in mean river flow to
Charleston Harbor from 300 cfs to 15,000 cfs. The first notice-
able effect of this change was an increase in the cost of dredg-
ing from $400,000 per year to $4,000,000 per year over a period
of several years, even at this cost it was impossible to keep
any slips perpendicular to the main channel clear.
A subsequent investigation(Wastler and Walter 19G9) showed that
the increased river flow had changed the main part of the Harbor
from an unstratified to a salt wedge estuarine circulation
system. This study concluded that reduction of the mean flow to
less than 8000 cfs would break the stratification and reduce the
input of sediment to the Harbor. The flow of the Santee was in
-------
part rediverted to its original waterway to reduce the mean flow
to these levels, and the stratification was indeed broken.
The Harbor at Charleston is once again a natural deepwater system
with a significant reduction in annual dredging costs. Sedimen-
tation was reduced to nearly its original levels and other as-
pects of water quality were also improved, in large part due to
considerable improvements in waste treatment, which were also re-
commended because of the tremendous decrease in flushing time
that would occur with the breaking of the salt wedge.
This example involved only an overall reduction of mean river
input into the system, and no effort was made to control the
river input in synchronization with tidal flow. However, this
example does demonstrate that dramatic changes can be caused by
changes in stratification resulting from changes in river flow and
that these changes could be quantified and predicted, at least in
the example given.
WHAT PHYSICAL CONDITIONS WOULD BE NECESSARY TO DO THIS IN
THE CHESAPEAKE BAY AND DO THEY PRESENTLY EXIST?
In the Charleston Harbor situation there was a single major
river inflow into the Bay and it clearly was a primary factor
in controlling the extent of stratification. There was also
a dam at the head of tidewater which had the capability of
regulating the river discharge into the estuary. While it might
be possible to visualize other physical conditions with which
it might be possible to regulate river flow so as to alter strat-
ification, is clear that these conditions are suffucient.
In the Chesapeake Bay system, the major single river inflow is
the Susquehanna, which provides about 87 percent of the riverine
input to the Bay above the confluence of the Potomac and 50 per-
cent of the entire riverine input to the entire Bay. Conowingo
Dam is about 9 miles above the head of tidewater and receives the
flow from over 90 percent of the Susquehanna drainage basin. The
pool behind Conowingo Dam is not sufficient for long-term storage
but the dams operated in the Basin by the Corps of Engineers do
have a total amount of storage sufficient to supply a minimum
flow of about 30,000 cfs over a several month period with proper
routing. Thus, the physical structure necessary to provide river
flow regulation does exist at the head of tidewater.
The question of whether or not the Susquehanna does have a suffi-
cient impact on stratification of the maJn stem of the Bay to aJ-
low effective regulation, particularly at low flows, must'aiso be
addressed. To do this, the results of the main stem monitoring
program were examined in regard to the extent of stratification
of the main stem throughout the year and how well this correlated
with the annual cycle of river flow. Table I exhibits the ratios
of salinity in the Main Stem of the Bay for the full period of
reported data for the present monitoring program, i.e., June 1984
to July 1987, at approximate mile points closest to the routine
sampling stations. The geographical area covered is from Havre de
-------
Grace to the mouth of the Potomac, the reach of the Bay in which
the Susquehanna accounts for nearly 90 percent of the river flow
entering the Bay. Data are presented on a monthly basis and each
value represents only one, or at the most an average of two sam-
ples .
There are several striking features of the data presented in
Table I as far as the impact of the Susquehanna flow on strati-
fication is concerned. First, the data show absolutely no evidence
of any type of seaonal overturn and redevelopment. There are
changes in the degree of stratification during a year, but these
do not appear to be related to any kind of seasonal regime, nor
to any obvious environmental scenario. Stratification in 1984 ap-
pears to be stronger than in either 1985 or 198G; however, there
is less than half a year of record in 1984, so it is difficult to
make an equitable comparison. Second, the change of the degree of
stratification down the Bay from Havre de Grace does not appear
to to be very great for a specific year or season, which suggests
that whatever forces are causing the degree of stratification ob-
served persist far down the Bay. Third, the degree of stratfica-
tion is extreme throughout the Bay and for much of the year. At
salinity ratios of around 0.0, Charleston Harbor behaved as a
salt wedge estuary. However, the size and structure of the Chesa-
peake are different from Charleston, and it is not known whether
the use of salinity ratio as a surrogate for estuarine behavior
is valid. Fourth, the asterisks (*) by certain values in the
table indicate times and places where the Bay exhibited more than
one pycnocline. These are quite common in the data set, and sug-
gest the existence of a consistent phenomenon responsible for the
condition.
Table II presents some data which may offer an insight into the
reasons for the existence of multiple pycnoclines in the Bay as
well as indicate a source of some of the organic material respon-
sible for the depleted oxygen levels in the bottom layer of the
water column. This Table compares water column densities in the
Potomac near its mouth with those in the Bay itself above, at,and
below the confluence of the Potomac with the Bay. For ease in
reading the density values have been adjusted so that they appear
in the table as small numbers; 'however, what is significant are
their relative values, and these are unchanged. The data
presented are for the entire year of 1985, and what is important
is to note the relative densities of the Potomac water and those
of the Bay at and close to the confluence.
Remember that water(e.g., Potomac water) tends to ride over other
water it meets of higher density(e.g..usually Bay bottom water)
and slide under water it meets of lesser density(e.g., usually
Bay surface water). The data show that in nearly all cases the
surface water in the Potomac would tend to ride out over or mix
with the surface waters of the Bay, although on two occasions,
3/19/85 and 5/G/85 the relative densities were such that the
Potomac water would tend to slide under the surface water of the
Bay. If these conditions persisted for any length of time, a
significant amount of Potomac surface water could be introduced
-------
into the bottom waters of the Bay where it would decay without
benefit of reaeration, thus causing possibly severe oxygen deple-
tion. For most of the year it appears that the surface layers
of the Potomac tend to end up in the surface waters of the Bay
but that a midlayer representing some kind of blending of Potomac
bottom water and Bay bottom water may very well form in the Main
Stem of the Bay. The small but significant differences between
the densities of the bottom and midlayers in the Bay suggest the
existence of such a mechanism.
This look at the salinity structure as shown by the monitoring
data suggests that there is no obvious, direct relationship
between the Susquehanna River flow and stratification in the Bay.
A more detailed look at the salinities of the bottom layer
indicated that these did tend to vary consistently with river
discharge, but a quantitative relationship could not be estab-
lished with the data available. Nevertheless, the wide range of
values of salinity ratios in no coherent pattern over a year sug-
gests that the Bay responds quite rapidly to changes in river
discharge and that samples taken a minimum of two weeks apart are
not adequate to establish the nature of the response. More de-
tailed measurements are needed to resolve this problem.
CAN THE PROPAGATION OF KINETIC ENERGY AND ITS DISSIPATION IN
THE BAY BE DEMONSTRATED?
The quantity of kinetic energy in a flowing body of water is a
function of its mass and the square of its velocity. In this
case there are two bodies of water involved, that associated with
the river discharge into the Bay and that associated with the
tidal flow. "Since water behaves as an incompressible fluid, the
interaction of the two bodies should be measurable as a
perturbation of the observed tide height by the river discharge.
The methods described by Wastler(Wastler 19G9) and used by
Wastler and Walter in the Charleston Harbor analysis allow the
estimation of the energy present at the dominant periods
exhibited in each record at each sampling point. These methods
involve the calculation of cross-spectra from river discharge and
tide height records and the interpretation of the results in
terms of the physical parameters involved. The effect of the pre-
sence of some potential energy due to the slope of the estuary
from its head toward the sea is eliminated by using only
deviations from the mean tide height at each tide gage; the
potential energy of the water would, however, be small in
relation to the kinetic energy in any event.
For the present case records of hourly discharges from Conowingo
Dam were available for the first nine months of 1983. These were
run against hourly tide gage records for the same period at Havre
de Grace, Matapeake, and Solomons, affording a coverage from the
head of tidewater nearly to the mouth of the Potomac. A range of
river discharges based on daily means from 5000 cfs to 103,000
cfs were covered by these records, thus incorporating an excellent
range of flows from the Susguehanna. The river discharge records
had an extremely strong diurnal component typical of hydroelectric
-------
dam operation. The analysis showed that the propagation of ener-
gy from the river discharge was extremely rapid and consistent at
all river disacharges. This amounted to 44 hours to the Solomons
gage, or roughly two days from Havre de Grace to the mouth of the
Potomac. This is, of course, the speed of energy propagation, not
the time of water travel. It should be noted that the rapidity
of energy propagation from the river flow could have some bearing
on the similarity of the degree of stratification down the Bay,
as shown in Table I. The diurnal component of river flow was
still present at Solomons with about 25 percent of the original
energy still present. Long-period components of the river
discharge (i.e., those of periods greater than 7 days) were not
large in the river record, and these were not distinguishable in
the records at Solomons. This suggests that the long-period energy
of the river discharge is dissipated in the upper part of the
Bay.
These results indicate that energy from the Susquehanna flow
makes its way at least halfway down the Bay in a quantifiable
form, that it tends to be distributed throughout the upper part
of the Bay in its presently uncontrolled regime, and that its
time of propagation is quite rapid. Thus it appears that there
is enough energy available to use if there is a means of
controlling it.
IS IT FEASIBLE TO DEVELOP A MATHEMATICAL DESCRIPTION OF THE
KINETIC ENERGY FLOW INTO THE BAY SUFFICIENTLY QUANTITATIVE TO
SERVE AS THE BASIS FOR REGULATING RIVER DISCHARGES INTO THE BAY?
Kinetic energy balance equations for estuarine flow and for river
discharges are standard textbook exercises, but apparently no one
has ever put the two togather. As part of this investigation a
synthesis of the two was attempted, not to develop an exact solu-
tion, but to establish the nature of the equation and the para-
meters necessary for its solution. A harmonically varying river
discharge was introduced to determine its effect.
The results indicated that a mathematical description of
the process is feasible, but there are some problems. First,
there appear to be a large number of acceptable mathematical
solutions depending on a number of unknown quantities which can
only be measured in the field. That is, it would be necessary to
experimentally vary the discharge from Conowingo in a preset pat-
tern and measure what happens. Second, with opposing harmonic
f]ow patterns, there is a real possibility of a feedback process
developing with the formation of a disastrously large standing
wave in the Bay. Again, this may be a mathematical artifact, and
field investigation may demonstrate that such is the case.
In short, yes, it is feasible to develop a quantitative mathemat-
ical description of what would be required, but it must be
based on accurate field studies in its development stage, and it
must be carefully and thoroughly evaluated in the field before it
is used in an operational mode.
-------
SINCE SUCH A MANAGEMENT SCHEME WOULD INVOLVE REAL TIME CONTROL
OP RIVER PLOWS INTO THE BAY, WHAT INFORMATION ON CONDITIONS IN
THE BAY WOULD BE NEEDED AS A BASIS FOR USING SUCH AN APPROACH?
Before such a management scheme could be implemented, it would
be necessary to have a very clear understanding of what changes
in river discharge would do to the salinity structure of the Bay.
It would also be necessary to understand what salinity structure
was needed at specific points in the Bay to provide the suitable
degree of stratification to protect and enhance the habitat for
living resources and to maintain a predetermined level of water
quality. With these considerations in mind, it can then be
stated that high-frequency real time measurements of salinity are
the only water quality information really necessary. However,
the only cost effective technique to obtain this type of
information presently available is automatic remote sensing
buoys. The major cost of such devices is in the buoy itself and
in the data transmission system, so it would be reasonable to
collect other useful information such as Temperature, Dissolved
Oxygen, Turbidity, pH, and other parameters of interest to
management authorities and responsible investigators.
IS THIS APPROACH WORTH PURSUING FURTHER, AND,IF SO, WHAT MUST
BE DONE?
This paper has emphasized the unknowns and the problems involved
in developing and implementing a management approach based on
flow regulation for positive control of the aquatic environment.
These problems and unknowns are real in the context of this
management approach certainly; but they are no less real in the
context of any other management strategy.
A major problem identified has been that of establishing a firm
relationship between river flow and stratification. Certainly it
is necessary to know this to differentiate between "wet years"
and "dry years" in any quantitative sense for any type of model.
Establishing such a relationship is a first and major step in any
further investigations, whether of this management approach or
any other approach toward water quality control.
For example, it would be impossible to develop a complex water
quality model of the Bay without accounting for the daily pattern
of flows out of Conowingo, the effect of such variations in flow
on water quality, the development of multiple stratification
layers in the Bay, and the occasional underflow of the Potomac
into these layers. The development of any such model would go
far toward evaluating the feasibility of controlled river flow
regulation as a management tool.
This approach should be examined further, not as a unique and
separate entity, but as part of the total array of management
techniques that might be feasible.
-------
BIBLIOGRAPHY
Wastler.T.A. Spectral analysis: applications in water pollution
control. Washington,DC: U.S. Department of the Interior; 1969.
Wastler.T.A.; Walter,C.M. Statistical approach to estuarine
behavior. J. of the Sanitary Engineering Division, ASCE. 94:
No.SA G, Proc. Paper 6311, 1175-1194; 1963.
-------
Table I
Date
Salinity Ratios in the Main Stem of the Chesapeake Bay
(June 1984 - July 1987)
Miles from Havre de Grace
20
30
40
50
GO
70
80
90
100
1984
June
July
Aug
Sept
Oct
Nov
Dec
1985
Jan
Feb
Mar
Apr
May-
June
July
Aug
Sept
Oct
Nov
Dec
198G
Jan
Feb
Mar
Apr
May
June
T .. T ..
jiu.y
Aug
Sept
Oct.
Nov
Dec
1987
Jan
Feb
Mar
Apr
May
June
July
0
0
0
0
0
0
0
0
0
r»
O
0
0
0
0
0
0
r»
u
0
0
0
-
-
-
.94
-
-
.41
-
-
-
-
.79
.52
.78
.76
.85*
.GO
.79
-
.42
-
-
-
-
-
.71
.92
.75
.67*
-
.09
-
.77
.08
.45
.08
.45
-
-
0.22
0.22
0.97
0.94
-
-
-
-
0.40
0.20
0.53
O.GG
0.59
-
0.82*
0.8G
-
O.G9
-
O.G2
0.2G
0.29
0.42*
0.52
O.G8
O.G9
-
0.70
-
0.47
-
0.74
0.70
-
O.G9
0.78*
O.G2
0
0
0
0
0
0
0
-
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.42
.36
.33
.68*
.82*
.60
-
.83
.02
.52*
.71*
.70*
.61*
.07*
.83*
-
-
.03
-
.47
.24
.25*
.51
.50*
.G8
.GG*
.88*
.07*
-
.58
-
.81
-
.24
.GO
.GO
.72
0.
0.3G 0.
0.38 0.
0.70* 0.
0.82* 0.
0.74* -
0.
-
-
0.07 0
0.54 0
0.08 0
O.G7 0
0.65 0
0.72* 0
0.82* 0
0.90 0
0
0.80
-
0.58*
0.45*
0.43*
0.52*
O.G7*
O.G7
O.G8
0.85
0.74*
-
O.G3
-
0.79
0.74
0.57
0.72
0.73
O.G3
G3
46 0
63 0
81 0
72 0
75* 0
-
-
.66
.62*
.67*
.70
.75
.76
.74
.80
.91
-
-
-
O.G8
0.41
O.G7
O.G1
O.GO
O.G8*
0.71
0.72
0.8G
-
0.7G
-
0.78
0.04*
O.G2
O.G9*
0.71
0.
.45 0.
.62 0.
.74 0.
.68 0.
0.
.79
0.
0.84
0.82 0.
0.61*0.
0.78*0.
0.72 0.
0.68*0.
0.67*0.
0.78 0.
0.88 0.
0.
0.74*
0
0.51*
0.52 0
0.57 0
0.5G*0
0.70*0
0.69 0
0.7G*0
0.79*0
0.74 0
- 0
0.71
- 0
0.79
O.G8 0
0.51*0
0.70 0
0.79 0
O.G4 0
65
49*
41*
71
78*
74*
-
85*
-
76
68*
05
68
75*
76
77
81
86
-
.84
-
.72
.55*
.68
.67
.64
.70
.74
.75*
.85
-
r» r\
: O£.
~
.77
.80
.GG*
.09
.78
0
0
0.38
0.71 0
0.76 0
0.78* 0
-
0.88
-
0.82*
0.74*
0.77*
0.7G*
0.70*
0.71*
0.76
0.92
0.86
-
-
0.68
0,70
0.57*
O.GO
0.7G
O.G9
0.80
0.88
0.74
-
0.80
-
0.79
O.G7
0.70
0.71
0.75
O.G8
.78
.57
-
.84*
.69
.83*
-
0.85
-
0.85
0.68
0.82
0.79
0.78*
0.76*
0.75
0.99
-
0.84
-
0.73
O.G9*
O.G3
O.G2*
0.89
O.G7*
0.78
0.8G
0.74
-
0.81
-
O.G4
0.72
O.G5
0.73
0.78
-
Data Source: Chesapeake Bay Program monitoring data
-------
Table II
Comparison of Water Densities at the Mouth of the Potomac River
with Those in the Main Stem of the Chesapeake Bay Above and
Below the Potomac Confluence*
Date
Bay above Potomac
Potomac Mouth
(CBS. 2)
1/14/85
2/11/85
3/4/85
3/19/85
4/8/85
4/22/85
5/6/85
5/20/85
6/3/85
G /' 1 7 / 85
U
M
L
U
M
L
U
M
T
Lt
U
M
L
U
M
L
U
M
L
U
M
T
Li
U
M
L
U
M
L
TT
U
M
T
1
1
1
1
1
1
1
1
1
1
1
1
- 0
1
1
0
1
1
1
1
1
0
1
1
1
1
.278
-
.456
.391
-
.515
.096
-
.455
.171
.288
.435
.155
.446
.708
.969
.119
.392
.978
.189
.563
.025
.150
.394
.979
-
.504
.007
.175
.434
(LE2
1
1
1
1
0
1
1
1
1
1
1
1
0
1
1
1
1
1
0
1
1
0
1
o
1
1
.3)
.254
-
.444
.265
-
.451
.977
.086
.204
.216
-
.411
.109
.260
.488
.956
.090
.286
.044
.321
.511
.898
.008
.364
.967
-
.273
.882
.042
.260
Bay at
Confluence
(CBS. 3)
1
1
1
1
1
1
1
1
1
1
1
0
1
1
1
1
1
1
1
1
1
1
1
1
.320
-
.552
.457
-
.585
.006
-
.456
.241
-
.451
.198
.626
.856
.950
.227
.569
.053
.389
.513
.156
-
.450
.040
.365
.616
.054
-
.402
Bay beli
Potomac
( CBS . 4 )
1.350
1.429
1.490
1.320
-
1.467
1.145
-
1.308
X
X
X
X
X
X
1.059
1.305
1.513
1.092
1 .449
1.G91
1.244
1.316
1.502
0.973
1.333
1.5G1
1.082
1.248
1.434
10
-------
Table II (Continued)
Bay above Potomac Bay at Bay below
Date Potomac Mouth Confluence Potomac
(CB5.2) (LE2.3) (CBS.3) (CB5.4)
7/8/85
7/22/85
8/6/85
8/19/85
9/9/85
9/23/85
10/7/85
U
M
L
U
M
L
U
M
T
U
U
M
L
U
M
T
Jj
U
M
L
U
M
L
0
1
1
0
1
1
1
1
1
1
1
1
1
1
1
1
1
, 1
1
11/12/85U 1
M
T
Li
1
.991
.356
.611
.976
.418
.609
.036
-
.369
.034
.399
.605
.003
.240
.429
.220
.429
.669
.347
-
.660
.422
-
.733
0
1
0
1
1
1
1
1
0
1
1
1
1
1
0
1
.991
-
.284
.939
-
.301
.034
-
.159
.012
-
.193
.879
-
.487
.211
-
.300
.287
-
.340
.852
-
.464
1
1
1
1
1
1
1
1
1
1
0
1
1
1
1
1
•1
JL
1
1
1
.034
-
.727
.036
.102
.465
.112
-
.419
.099
.256
.559
.826
.076
.389
.267
-
.767
.320
-
.649
.044
.247
.627
1.054
-
1.586
1.220
-
1.556
1.215
-
1.567
1.105
-
1.284
1.003
-
1.432
x
X
X
1.386
-
1.G39
x
X
X
* Values in table are (Density x 1000) - 1000, which gives more
readable numbers.
U - upper layer
M - middle layer( - means there is no middle layer)
L - bottom layer
x - no oat a
11
-------
-------
Role of the Organic and Inorganic Carbon Systems in the Dissolved
Oxygen Regime of the Chesapeake Bay
by
T. A. Wastler, Senior Science Advisor
Office of Marine and Estuarine Protection
U.S. Environmental Protection Agency
Washington, D.C. 20460
INTRODUCTION
Examination of the three years of water monitoring data on the
main stem of the Chesapeake Bay( collected from June 1984 to July
1987 as part of the Chesapeake Bay Program), suggested that there
is much closer coupling of the organic and inorganic carbon systems
in the main stem of the Bay than had been previously suspected.
A quantifiably close coupling of the two organic systems in the Bay
would allow prediction of the behavior of one system from the be-
havior of the other, and, since dissolved oxygen is the
connecting link between the two systems, it should be possible to
predict the behavior of the dissolved oxygen regime from
knowledge of either the organic or inorganic carbon systems.
Since the composition and dynamics of the inorganic carbon system
can be determined from measurements of temperature, salinity,
and pH alone, this would make it quite simple to predict periods
of low dissolved oxygen from relatively straightforward measure-
ments .
THEORETICAL BACKGROUND
The total carbon budget of any water body is composed of the sum
of the components of the organic carbon system plus those of the
inorganic carbon system. In freshwater the major concern is the
organic carbon system, since the inorganic carbon system is driven
13
-------
primarily by the solubility of gaseous carbon dioxide in fresh
water. In seawater, however, the inorganic carbon system is quite
complex ,as can be attested by all students of marine chemistry.
The best and most detailed discussion of this subject is in
Harvey's classic text "The Chemistry and Fertility of Seawater"
(Harvey 19G3), and only aspects of this system immediately
relevant to anoxia in the Chesapeake Bay will be mentioned here.
The components of the inorganic carbon system are hydrogen ions,
carbonate ions, bicarbonate ions, free carbon dioxide, and undis-
sociated carbonic acid. The exact amounts of each component pre-
sent in a specific situation are determined by the temperature,
salinity, pH, and pressure. As pH values increase, the amount
of free carbon dioxide decreases and the equilibrium is shifted
toward the carbonate ion. Also, as algae consume free carbon dioxide,
the pH tends to increase, thereby causing the amount of free carbon
dioxide that can exist in the equilibrium to decrease. The kinetics
of the equilibrium are not extremely rapid as far as the
regeneration of free carbon dioxide from the carbonate is concerned,
therefore it is possible that conditions could exist in the ocean
in which algal growth is limited by the availability of free carbon
dioxide even in the presence of a large excess of total inorganic
carbon. It has been observed in some estuarine situations that algal
growth has been inhibited by a lack of carbon dioxide at pH 9
(Kinne 1972).
In fresh water, on the other hand, a very large amount of free
carbon dioxide can be dissolved in the water since the concentra-
tion of dissolved carbonate ion is quite small and does not
significantly affect the chemical equilibrium. Thus, the
conditions in which the availability of inorganic carbon may be-
come a limiting factor in algal growth would not be expected in
fresh water systems.
Rates of organic decay and oxygen consumption in fresh water sys-
tems have been studied in considerable detail in various water
quality control and public health applications; the classic for-
mulation by Streeter and Phelps of the oxygen sag in streams is
accepted as the standard theoretical description of the natural
decay processes involved(Fair and Geyer 19G3). In salt water the
rate of decay appears to vary with the concentration of the sea-
water. In concentrations of up to about 25 percent seawater, the
rate constant for organic carbon decomposition is greater than in
fresh water, but in higher concentrations it is depressed.
The total amount of carbon present in a water body at a specific
time ana place consists of the sum of those amounts present in the
organic carbon system plus those present in the inorganic carbon
system. Between times and places there may be exchanges between the
two systems (which would not change the total amount of carbon
present) but there may also be external sources or sinks of
carbon in either organic or inorganic form. In examining the
monitoring data on the Bay, consideration must be given to any such
external sources or sinks of carbon, particularly to the sediment
oxygen demand, which has been regarded as a significant factor in
14
-------
development of anoxia in the Bay. Resuspension of organic matter
from sediments into the water column is a potentially important
source of organic carbon in the water column where water column
properties such as dissolved oxygen could be affected.
The primary connecting link between the organic and inorganic
carbon systems is oxygen, particularly the free oxygen dissolved
in the water column. Oxygen is used up in the decay process of the
dissolved organic matter with the formation of carbon dioxide;
carbon dioxide is used by algae to produce organic matter with
the release of oxygen. The algae themselves consume some oxygen
in respiration and when algal cells die their decaying tissue
uses up oxygen with the formation of carbon dioxide. Thus oxygen
is an integral part of both systems and is directly involved in
the transference of carbon from one system to the other.
The basis for this investigation is the application of the stan-
dard textbook concepts and relations described above to the
specific problem of quantifying the coupling of the inorganic and
organic carbon systems in the bottom waters of the Chesapeake Bay
and gaining some insight into the quantification and prediction
of the hypoxia of those waters.
STUDY METHOD
The method used in this investigation was a simple materials
balance at pairs of monitoring stations occupied in the Chesapeake
Bay Program monitoring effort between June 1984 and July 1987. The
overall approach was to use the classical theory of the oxygen
depletion in a natural body of water as enunciated in the
Streeter-Phelps analysis in concert with the accepted inorganic
carbon system behavior as summarized by Harvey. The parameters
required for the analysis were Temperature, pH, Salinity, Dis-
solved Oxygen, Total Organic Carbon(TOC), Dissolved Organic Car-
bon(DOC), depth of the pycnocline, and the thickness of both
surface and bottom layers. Particulate Organic Carbon(POC) can be
calculated from TOC and DOC, and total inorganic carbon and
free Carbon Dioxide can be calculated from pH, Temperature, and
Salinity using the equilibrium constants and tables presented in
Harvey. Thus it was possible to use only pairs of stations for
which all of these parameters were available.
The major geographical area of interest is the reach of the Bay
between the Bay Bridge at Annapolis and the mouth of the Potomac
River. This includes the part of the Bay having the greatest
depths and the most severe oxygen depletion problems. The
analysis was therefore concentrated in this reach, with the
analysis limited to the period between May and November 198G,
which is the only period for which sufficient data were avail-
able. Individual samples were used to calculate rate processes
between stations, since the two week or greater sampling period
made the aggregation of samples for this purpose highly suspect.
The monitoring data show that the Bay between the Bay Bridge and
the Potomac is stratified throughout the year, with the ratio of
15
-------
surface to bottom salinity being generally about O.G to 0.8,
indicative of a strongly stratified system. The monitoring data
show no evidence of a seasonal overturn in the main stem of the
Bay; in fact, the data show the existence of more than two layers
within this reach for part of the year. The materials balance
for this reach therefore reflects the existence of at least a
two-layered system, with the flow in the bottom layer up the Bay,
and that in the surface layer down the Bay.
Some mixing between layers is expected even in a strongly
stratified system. Thus, in addition to being transferred chemi-
cally or biochemically from one carbon system to the other, there
is some mechanical transfer between layers for each constituent.
An overall materials balance for the Salinity, which is a
conservative parameter, provides ratios of surface layer to
bottom layer average velocity between the stations, as well as a
ratio of the vertical transport velocity to either the surface
layer or bottom layer velocity. An average effective cross-section
for the two stations is assumed; the cross-sectional area
above the pycnocline is assumed to be rectangular, and that below
the pycnocline is assumed to be triangular with the altitude
equal to the observed thickness of the bottom layer.
The first step was to compute the overall carbon balance for both
layers and for each layer separately. This provided an estimate
of the external sources and sinks of carbon, both organic and
inorganic. Balances on each carbon system for each layer then
gave estimates of the external gain or loss to each system.
Next, for the bottom layer material balances were made for DOC,
TOC,dissolved oxygen, and free and total carbon dioxide, assuming
decay processes of the first order, following the Streeter-Phelps
formulation for the decay of organic matter in the absence of
reaeration. Calculations were also made in some cases using a
second order rate process assumption. In the bottom layer
calculations it was assumed that any change in POC between
stations was the result of scouring or deposition of sediment .
containing adsorbed organic matter. An increase in DOC between
stations in the absence of a known external point source, is as-
sumed to be from organic matter dissolved off POC or interstitial
water elutriated from the sediment bed load.
RESULTS
A first matter of concern is the extent to which the sediment
load (or "bed load" in stream pollution terminology) may affect
the dissolved oxygen regime in the Chesapeake Bay, particularly
that part of the Bay which is subject to anoxic conditions in the
bottom layer, and which is the area selected for examination in
the present study. The technique of materials balancing, which
considers all sources of carbon to the water column, affords an
opportunity to look at external carbon sources, and in the case of
the bottom layer of the water column, particularly what is put
into or taken out of that layer. Table I summarizes the results
of this analysis of the monitoring data. The data show a mixed
16
-------
pattern of scouring and settling of Participate Organic Carbon and
also a similarly mixed pattern of addition or loss of Dissolved
Organic Carbon to the water column. A significant point to note
in this table is that the magnitude of addition or loss to
the water column from the sediment load is of the same order of
magnitude as the smallest reporting number of the data. This
indicates that the quantitative extent of the impact of the sedi-
ment load is so small as to not be detectable in the monitoring
data. Therefore, in analyzing the monitoring data, separate con-
sideration of the sediment load as a factor in hypoxia in the
bottom layer is not appropriate, since its impact is subsumed in
the existing water column data.
Table Ha. compares the results of calculating the values of
Dissolved Organic Carbon using the inorganic carbon balance with
the observed values reported in the monitoring program. Table
lib. shows differences between the observed values of Dissolved
Oxygen depletion between two monitoring stations and the
calculated increase in Carbon Dioxide between the same two
stations. The pattern of differences is again mixed, but again
the differences are within the order of magnitude of the mea-
surements. The bottom line conclusion is that there does not appear
to be a basis for deciding that there is a significant difference
between the two sets of data. An analysis of variance of the data
in Tables Ila. and lib. showed that in each case the data could be
considered to be of the same set at the 99 percent probability
level.
Table III shows the results of calculating first order deoxygenation
rate constants from the organic carbon balance (Table Ilia.) and
from the inorganic carbon balance (Table Illb.) using the monitoring
data. The results show a large variation in both cases. For the or-
ganic balance the mean valueis -0.28, while the inorganic balance
based on free Carbon Dioxide gives a value of -0.17. In each case
the mean temperature is 21.0 degrees C. The value obtained in a
large number of practical cases for freshwater streams is -0.23 at
20.0 degrees C. based on BOD, not on DOC. However, it may be
noted that the two values for rate constants found in this study
average -0.23. It is worthy of note that at least two months of the
monitoring data showed dissolved oxygen values so low as to raise
questions as to their usefulness in rate constant calculations, and
it was these data that resulted in the lowest rate constant values.
The calculations leading to the results presented in Table III
suggested that the anoxia development in the bottom layer of the
Main Stem could be explained analytically through application of
a first order Streeter-Pheips deoxygenation formulation. To do
this would require more reliable data on the organic carbon com-
ponents of the water column and more extensive data on the course
of oxygen concentration change in the bottom layer during the
Spring and Fall transition periods to and from hypoxic conditions.
However, the close coupling of the organic and inorganic carbon
systems suggested that it might be possible to use the Streeter-
Pheips formulation to calculate the Dissolved Oxygen values for
1984 and 1985, for which the organic carbon data were mis-
17
-------
sing, by estimating an initial value for the Dissolved Organic
Carbon from the inorganic carbon system and then applying the
Streeter-Phelps approach to predict the depletion of Dissolved
Oxygen in the bottom layer.
Table IV presents the results of such an analysis. This table
shows a comparison between observed Dissolved Oxygen values in
the bottom layer and those calculated as described above.
The calculated and observed Dissolved Oxygen values are in all
cases in agreement within the limits of experimental accuracy.
The results presented here use the rate constants calculated
using the 1980 data to reproduce the actual Dissolved Oxygen
values of 1984 and 1985, not only for the Summer months, but
throughout the year where data are available. This analysis is
basically an academic exercise to illustrate the close coupling
between the organic and inorganic carbon systems and has no prac-
tical application in its present form. However, the method of
calculation presented,e.g., a materials balance and use of the
Streeter-Phelps formulation to predict the onset of low Dissolved
Oxygen in the bottom layers of the Bay, could provide a reasonable
basis for management decisions on combatting the anoxic conditions
in the Bay at very little cost compared to large numerical models.
CONCLUSIONS
1. An extremely close coupling between the organic and inorganic
carbon systems has been demonstrated; this relationship is
sufficiently quantitative that the components of one system can
be estimated from the components of the other.
2. Sediment -load is not a significant factor in predicting the
behavior of the oxygen of the bottom layer of the water column.
3. The behavior of the oxygen system of the bottom layers of the
Bay can be predicted using a first-order formulation of the
Streeter-Phelps type with very little expenditure for development
of a practical model.
4. To develop a quantitative model based on this approach, the
present monitoring program needs to be redesigned to acquire more
data during the Spring and Fall when DO concentrations are in
transition and to provide additional data on both carbon systems.
High frequency data on pH, Temperature, Salinity, and DO in the
bottom layer would minimize the need for other data, and in all
likelihood make it possible to predict the onset of anoxia from
these data alone. A cost effective way to do this would be to
convert the existing manual monitoring program to one using con-
sistently reliable remote sensing buoys.
3. The question of whether or not free Carbon Dioxide might
be a limiting nutrient under salinity conditions existing in the
Bay should be explored.
18
-------
Table I
Changes in Organic Carbon in the Bottom Layer from Sediment Load
(Chesapeake Bay Mainstem between Bay Bridge and Potomac; Miles
from Havre de Grace at midpoint of section between stations; all
data from 193G monitoring results)
a. Participate Organic Carbon(mm/1)
+ - scouring
- - settling
Mile May June July August September October November
20
50
GO
70
80
90
100
0.07
0.13
0.20
0.02
0.01
0.03
0.03
-0.17
-0.09
-0.01
-0.03
-0.04
0.10
-0.05
-0.20
-0.17
-0.05
-0.07
-0.09
0.00
-
-0.18
-0.02
0.11
-
-0.03
-0.02
-
-0.11
-0.13
0.02
-0.03
0.01
O.OG
0.00
0.15
0.01
0.03
-0.30
0.00
0.03
0.05
0.01
-0.03
-0.03
0.50
-
0.02
-
b. Dissolved Organic Carbon(mm/l)
+ - addition to the water column
- - loss from the water column
Mile
May
June July August September October November
20
50
GO
70
80
90
100
0.35
0.3G
0.17
-0.1G
0.07
0.3G
0.47
0.27
0.-15
0.12
-0.21
0.22
0.91
0.15
-0.
0.
0.
-0.
-0.
-0.
-
55
05
00
38
44
14
-0
0
0
-0
0
.78
.14
.38
-
.15
.OG
-
-0
-0
0
-0
-0
-0
0
.55
.52
.10
.42
.09
.17
.43
-1.
0.
-0.
-2.
-0.
0.
0.
20
23
07
02
08
13
54
-1.2G
0.30
-0.90
-0.1G
-
0.72
-
19
-------
Table II
Coupling Between Organic and Inorganic Carbon Systems in the
Bottom Layer of the Main Stem Chesapeake Bay
a. Difference Between Observed Bottom Layer Dissolved Organic
Carbon Values and Those Calculated from the Inorganic Carbon
Balance(mm/1)
Mile May June July August September October November
-1.50 -1.00
0.00 0.01
0.01 0.10
0.17 -3.00
0.00
0.01 0.14
100 0.00 0.00 - - 0.03 0.10
b. Difference Between Calculated Carbon Dioxide Generated and
Observed Oxygen Consumed in the Bottom Layer(mm/1)
Mile May June July August September October November
-0.40 0.15
O.OG 0.20
0.33 -0.30
0.54 -0.82
0.05
0.05 -0.25
100 0.03 0.08 - - 0.05 0.17
20
50
GO
70
80
90
-0.
-0.
0.
0.
0.
0.
08
02
03
15
00
01
0
0
0
0
0
0
.03
.05
.01
.06
.07
.10
0
0
-0
-0
0
0
.08
.07
.02
.07
.23
.01
0.
-0.
-0.
-
0.
0.
01
01
03
05
OG
0.
0.
0.
0.
0.
0.
16
01
02
16
01
03
20
50
GO
70
80
90
0
-0
0
0
0
0
.13
.07
.12
.12
.02
.05
0
0
0
0
0
0
.08
.05
.07
.OS
.02
.10
-0
0
0
0
0
-0
.22
.04
.02
.08
.18
.10
0
0
-0
0
0
.23
.08
.10
-
.03
.05
0
0
0
0
0
0
.21
.28
.03
.12
.02
.10
20
-------
Table III
First Order Deoxygenation Rate Constants Calculated
from Chesapeake Bay Main Stem Monitoring Data
for the Bottom Layer Between the Bay Bridge and the Potomac
a. Rate Constants Calculated from the Organic Carbon Balance
Using the Streeter-Phelps Algorithm(per day)
Mile May June July August September October November
20
50
GO
70
30
90
-0
-0
-0
-0
-0
-0
.77
.40
.37
.18
.15
.25
-0.71
-0.28
-0.18
-0.40
-0.16
-0.25
-0.93
-0.17
-0.07
-0.14
-0.19
-0.10
-0
-0
-0
.06
.24
-
-
-
.03
-0
-0
-0
-0
-0
.41
.24
.20
.28
-
.19
-0
-0
-0
-0
-0
-0
.45
.20
.26
.48
.19
.15
-0.16
-0.27
-0.32
-0.67
-
-0.19
100 -0.28 -0.25
Average(arithmetic mean) - -0.28
-0.20
b. Rate Constant Based on Inorganic Carbon Balance — Free
Carbon Dioxide Generation(per day)
Mile
May
June July August September October November
20 -0.94
-0.05 -0.07
50
GO
70
80
90
-0.03 <
-0.44
-
-0.89
-0.05
-0.27 -0.28
-0.25 -0.01
-0.02
-
-0.1G
100
-0.28. -0.19
Average(arithmetic mean) - -0.17
-0.02
-0.05
-0.02
-0.00
-0.01
-0.14
-0.32
-0.32
-0.02
-0.15
21
-------
TABLE IV
COMPARISON OF OBSERVED DISSOLVED OXYGE VALUES
IN THE BOTTOM LAYER WITH THOSE CALCULATED FROM
THE INORGANIC CARBON BALANCE
(STATIONS FROM POTOMAC TO BAY BRIDGE)
DATE STATION CALCULATED DO OBS.DO
FREE C02 TOTAL COS
RATE CONSTANT 8 20.OC - -.23
BOTTOM VELOCITY (FT/SEC) -
08/00/84 CBS. 2
08/06/84 CB5.1
08/06/84 CB4.4
08/06/84 CB4.3C
08/06/84 CB3.3C
1.425
1.015
.785
-.005
-.005
BOTTOM VELOCITY (FT/SEC) -
09/10/84 CBS. 2
09/10/84 CB5.1
09/10/84 CB4.4
09/10/84 CB4.3C
09/10/84 CB3.3C
-.114
-.173
-.199
-.196
-.033
BOTTOM VELOCITY(FT/SEC) -
09/24/84 CBS. 2
09/24/84 CB5.1
09/24/84 CB4.4
09/24/84 CB4.3C
09/24/84 CB3.3C
2.571
2.091
1.731
.001
2.661
BOTTOM VELOCITY(FT/SEC) -
10/22/84 CBS. 2
10/22/84 CB5.1
10/22/84 CB4.4
10/22/84 Cfl4.SC
10/22/84 CB3.3C
4.518
4.355
4.032
2.330
1.573
BOTTOM VELOCITY(FT/SEC) -
12/10/84 CBS. 2
12/10/84 CB5.1
-12/10/84 CB4.4
12/10/84 CB4.3C
12/10/84 CB3.3C
8.250
8.526
7.754
8.412
8.677
BOTTOM VELOCITY (FT/SEC) -
04/22/85 CBS. 2
04/22/85 CB5.1
04/22/85 CB4.4
04/22/85 CB4.3C
04/22/85 CB3.3C
7.982
4.941
3.837
2.943
1.788
.820
1.967
1.552
1.319
.525
.522
.180
-.195
-.250
-.275
-.271
-.107
.479
2.986
2.506
2.142
.407
3.055
.691
5.182
5.023
4.704
3.005
2.257
.459
8.339
8.622
7.853
8.515
8.789
.154
7.850
4.796
3.688
2.789
1.G29
1.430
1.020
.790
.000
.000
.120
.050
.020
.020
.180
2.570
2.090
1.730
.000
2.660
5.020
4.860
4.540
2.840
2.090
8.300
8.580
7.810
8.470
8.740
8.100
5.070
3.970
3.080
1.930
22
-------
TABLE IV (CONTINUED)
DATE STATION CALCULATED DO OBS.DO
FREE C02 TOTAL COS
BOTTOM VELOCITY(FTXSEC) - .258
4.540
3.830
4.420
3.740
3.330
3. GOO
2.010
2.040
.580
.000
3.8GO
3.100
1.550
.030
.000
1.2GO
.940
.720
.610
1.150
2.880
2.580
2.4GO
1.830
.200
.010
.310
.310
.430
.070
05/06/85 CBS. 2 4.234
05/06/85 CB5.1 3.508
05/06/85 CB4.4 4.094
05/06/85 CB4.3C 3.405
05/06/85 CB3.3C 2.981
BOTTOM VELOCITY (FT/SEC) -
05/20/85 CBS. 2 3.443
05/20/85 CB5.1 1.851
05/20/85 CB4.4 1.879
05/20/85 CB4.3C .418
05/20/85 CB3.3C -.165
BOTTOM VELOCITY (FT/SEC) -
06/03/85 CBS. 2 3.903
06/03/85 CB5.1 3.144
06/03/85 CB4.4 1.593
06/03/85 CB4.3C .073
06/03/85 CB3.3C .044
BOTTOM VELOCITY (FT/SEC) -
OG/17/85 CB5.2 .856
06/17/85 CB5.1 .546
06/17/85 CB4.4 .328
06/17/85 GB4.3C .220
06/17/85 CB3.3C .764
BOTTOM VELOCITY (FT /SEC) -
07/08/85 CB5.2 3.099
07/08/85 CB5.1 2.796
07/08/85 CB4.4 2.675
07/08/85 CB4.3C 2.043
07/08/85 CB3.3C .405
BOTTOM VELOCITY (FT/SEC) -
07/22/85 CBS. 2 .171
07/22/85 CB5.1 -.109
07/22/85 CB4.4 -.103
07/22/85 CB4.3C .025
07/22/85 CB3.3C -.320
5.121
4.440
5.038
4.375
3.992
.246
3.918
2.331
2.365
.907
.335
.520
4.475
3.718
2.161
.645
.618
.287
1.355
1.033
.812
.702
1.241
.934
5.439
5.107
4.9G8
4.315
2.601
.495
.095
-.182
-.174
-.045
-.388
23
-------
TABLE IV (CONTINUED)
DATE STATION
CALCULATED DO
FREE C02 TOTAL COS
BOTTOM VELOCITY(FT/SEC) -
08/06/85 CB5.2
08/06/85 CB5.1
08/06/85 CB4.4
08/06/85 CB4.3C
08/06/85 CB3.3C
.514
-.609
-.276
-.525
.098
BOTTOM VELOCITY(FT/SEC) -
09/09/85 CB5.2
09/09/85 CB5.1
09/09/85 CB4.4
09/09/85 CB4.3C
09/09/85 CB3.3C
2.038
.103
-.236
-.453
-.612
BOTTOM VELOCITY (FT/SEC) -
10/07/85 CBS. 2
10/07/85 CB5.1
10/07/85 CB4.4
10/07/85 CB4.3C
10/07/85 CB3.3C
3.135
3.371
3.305
2.919
1.599
BOTTOM VELOCITY (FT/SEC) -
11/12/85 CBS. 2
11/12/85 CB5.1
11/12/85 CB4.4
11/12/85 GB4.3C
11/12/85 CB3.3C
6.056
6.155
5.924
6.604
7.272
BOTTOM VELOCITY (FT /SEC) -
12/09/85 CBS. 2
12/09/85 CB5.1
12/09/85 CB4.4
12/09/85 CB4.3C
12/09/85 CB3.3C
7.963
7.727
7.591
7.686
' 7.469
.444
.356
-.280
.047
-.212
.400
.680
2.283
.337
-.005
-.226
-.395
.697
4.010
4.237
4.165
3.773
2.438
.747
6.302
6.405
6.176
6.856
7.532
.145
8.184
7.971
7.844
7.947
7.740
OBS.DO
1.250
.100
.420
.150
.750
2.810
.840
.490
.260
.070
3.720
3.950
3.880
3.490
2.160
6.130
6.230
6.000
6.680
7.350
8.110
7.890
7.760
7.860
7.650
24
-------
BIBLIOGRAPHY
Fair.G.M.;6eyer,J.G. Water supply and waste-water disposal. New
York, NY-.John Wiley and Sons; 1963: Chapter 28.
Harvey,H.W. The chemistry and fertility of sea waters.Cambridge,
UK: Cambridge University Press; 1963: Chapter 10.
Kinne.O..editor. Marine ecology, vol l.part 3. New York, NY:
Wiley-Interscience;1972: Chapter 9.
25
-------
-------
CHESAPEAKE BAY SEDIMENT MONITORING FOR WATER QUALITY
MODEL DEVELOPMENT
Lewis C. Linker
EPA Chesapeake Bay Program
410 Severn Avenue
Annapolis, Maryland 21403
INTRODUCTION
The Chesapeake Bay Program has a comprehensive modeling strategy consisting of three
mathematical water quality models: the Watershed Model, the Steady State Eutrophication
Model, and the Time Variable Eutrophication Model. The Watershed Model covers the
entire 64,000 square miles of the Bay drainage basin and simulates pollutant loads
delivered to the Bay from various land use, population, and point source treatment
scenarios. Essentially complete, the Watershed Model is undergoing a series ol"
refinements which will be concluded in 1989. The Steady State Model is designed to give
an initial estimate of the relationships among nutrients, eutrophication, and anoxia, and to
provide an initial evaluation of proposed nutrient control strategies. The Steady State Model
was completed in the spring of 1987 and is fully successful in its application. The Time
Variable Model will improve nutrient control strategy evaluation by projecting the degree
and timing of the Bay response to control actions. The Time Variable Model will be
capable of short-term simulations of critical episodic events (e.g. pycnocline tilting) and
long-term simulations of about 30 years. Work was initiated on the Time Variable Model
in October 1987, and will be completed in 1991.
Steady State Model results provided important guidance for the development of the Time
Variable Model (HydroQual, 1987). Among the findings of the Steady State Model:
o The decline in dissolved oxygen (DO) in the bottom waters of the Bay
27
-------
o Only management actions that reduce SOD and sediment nutrient flux
improve Bay water quality to any significant degree.
The strong linkage between Bay sediments and water quality requires a multilayer sediment
model to be incorporated into the Time Variable Model framework.
The sediment submodel will have three components: net deposition of paniculate organic
matter (POM), diagenesis of POM to dissolved inorganic components within the sediment,
and nutrient flux, the movement of the dissolved inorganic nutrients from the sediment to
the water column (Figure 1). Detailed modeling of Bay vertical processes requires an
intensive sediment monitoring program to provide necessary data for model formulation,
calibration, and verification. The sediment monitoring program described below will begin
in April 1988 and continue for one year.
The sediment monitoring program is a cooperative effort of the Chesapeake Bay Program's
Modeling and Monitoring Subcommittees with expert assistance from HydroQual, Inc.,
the participants of the Sediment Processes and Sediment Modeling Workshop, and the
U.S. Army Corps of Engineers. The generous cooperation of the principle investigators
participating in the sediment monitoring program is gratefully acknowledged. They are:
Walter Boynton, Michael Kemp, Johnathan Garber, Peter Sampou, and Jeff Cornwell,
University of Maryland; Richard Wetzel, Larry Hass, and Bruce Neilson, Virginia Institute
of Marine Science; David Burdige, Old Dominion University; and Grace Brush, Johns
Hopkins University.
EXISTING SEDIMENT DATA
Since 1984, the Maryland Department of the Environment (then the Office of
Environmental Programs) has supported an integrated sediment, water column, and
phytoplankton monitoring program called SONE (Sediment Oxygen and Nutrient
Exchange) (Boynton et al., 1987). SONE focuses on the exchange of material between
sediment, water column, and phytoplankton in the upper and mid-Bay (Figure 2).
Incubated sediment cores are used to measure SOD and nutrient flux, and vertical arrays of
sediment traps are used to measure movement of material between the sediment and water
column. This'ongoing study is foundational to the sediment monitoring^program. The
SONE study allows the use of an existing data set for a large portion of the Bay, and
informed judgment as a guide for the sediment monitoring program. Other important
studies of Bay vertical processes can be found in an excellent review and synthesis by
Garber (1987).
REQUIRED SEDIMENT DATA
The close linkage between data collection and model development requires correct
anticipation of data needs. Effort is concentrated on the mainbay and on lower estuary sites
of major tributaries.
Sediment station locations
The sampling plan has a total of 25 stations located along the mainstem channel axis and in
the lower tributaries (Figure 2). Stations 2, 4, 5, 7,9, 10, 11, 12, 13, and 14 are existing
SONE program stations (Boynton et al., 1987). Stations 6,7, and 8 are the mid-Bay
transect stations and include shallow lateral stations and a deep water station of a previous
study (Malone et al., 1986). Stations 20,21, and 22 are lower Bay transect stations.
Lateral transect stations are used in the mid-Bay and lower Bay to capture aspects of the
vertical exchanges between nutrient generating deep waters and adjacent biologically
28
-------
UJ Z>
H-J
<0
£ O
o
UJ
CO
CO
M
co
o
UJ
<
O
i-o.
CO
CO
UJ
tr
CO
UJ
X
-J
u_
C9 •
CO
UJ
z
o
Q
t
CO
00
o>
O
O
DC
Q
O
4)
•o
O
0)
s
o
CO
OT
c
4)
C
O
0.
o
o
© 1
=3
O)
il
?s
00
UJO
cou.
29
-------
productive shallow shelves (Malone et al., 1986). The mid-Bay transect stations are
approximately centered in the area of summer anoxia. Southern transect stations are in an
area of summer hypoxia (U.S. EPA, 1983).
There is very little existing information on lower Bay sediment/water column exchanges.
Lower Bay stations 20,21,22,24, and 26 are located according to homogeneous sediment
types identified by Wright et al. (1986) as: estuary mouth shoals and spits, Bay-stem
plains, Bay-stem channels, lower estuary or shallow bay muddy beds, and Bay mouth
shoals respectively. The remaining eight stations are located relative to the 2-D model
segmentation scheme. All sediment stations are located close to CBP water column
monitoring stations.
Net Deposition
The net deposition component of the model simulates the input of inorganic and organic
matter to the sediments. Remineralization of organic matter within the sediment provides
material which contributes to SOD and dissolved nutrient fluxes. Inorganic matter
contributes to sedimentation and advects organic inputs down into the sediments.
Measurements of the net input of organic and inorganic matter to the bottom will be used to
calibrate net deposition.
Two vertical array sediment traps will measure the net flux of solids across various water
column layers and to the bottom. Each vertical array trap has three pairs of particle
collectors located above the pycnocline, just below the pycnocline, and just above the
sediment, as in the SONE program (Boynton et al., 1987). Particulate organic nutrients
(participate organic carbon [POC], participate organic phosphorus [POP], particulate
organic nitrogen [PON], particulate organic silica [POSi]), total solids, BOD, chlorophyll a
and pheophytin, and general algal classification will be determined from the collected
particulates. Algal identification of three broad functional groups (diatoms, non-diatom
eucaryotes, and picoplankton), will provide information on phytoplankton occurrence, cell
size, and settling rate. Sediment traps are located at two stations: station 7, an existing
SONE station in the mid-Bay transect, and station 21 in the lower Bay transect. The
sampling schedule results in 25 sampling periods a year with intensive weekly sampling in
summer months between July and mid-September. Data from the sediment traps will be
used for calibration of net deposition.
Sediment particulate organic profiles will be measured to determine the particulate organic
material (POC, POP, PON, POSi, particulate organic sulfur [POS]) deposited over the
years modeled in the long term simulations. The vertical distribution of bulk POM will be
matched with coincident determinations of average sedimentation rates in duplicate cores.
Radionuclide (14C) and pollen dating techniques (Brush, 1984; Brush et al., 1982) will be
used to determine average sedimentation rates, with particular emphasis on the profile
between 1950 and the present. From this, long term average net deposition of refractory
organic matter will be determined. The vertical profiles of particulate organics will be
determined at all stations. Vertical profiles of pore water concentration will further
characterize sediment composition at most sediment stations (Table 1).
Diagenesis
The diagenesis component of the model simulates the transformation of POM inputs to
dissolved inorganic nutrients. There are three fractions of diagenic material: a labile
fraction, a refractory fraction, and an inert fraction. These are empirical classifications
based on a fraction that is fast reacting (labile) and is in thermodynamic equilibrium, a
30
-------
Figure 2. Location of sediment monitoring stations.
| = SONE station. Sone stations have four flux sample
periods, except for stations 2 and 7.
CBP station. CBP stations have two
flux sample periods,except for station 21.
High frequency stations; six flux sample
periods.
Sediment trap.
31
-------
fraction that is slow reacting (refractory) but has kinetic rates that are important to the
model, and a fraction (inert) that does not react within the time frames of the model. The
kinetic rates for these three fractions must be determined for the diagenesis component of
the model. Organic material is transformed at rates which are a function of temperature, the
degree of anoxia, the amount of organic matter present, the presence of other chemical
constituents, and the sedimentation rate.
The sediment in the Bay contains a considerable amount of the refractory organic fraction
that has not undergone complete decomposition. This heterogeneously distributed organic
portion of the sediment continues to exert considerable SOD and contribute to nutrient flux
from the sediments. Long term diagenesis will be determined from main-Bay and southern
Bay tributary stations. Diagenesis rates will be determined by long-term sulfate depletion
studies on sediment slurries from three sediment depths: a surface (0-2 cm) sample of
recently deposited material, sediment of a "medium" age collected at a depth between 6 and
8 cm, and older deep sediments collected between 12 and 14 cm. Sediments will be
incubated for 50 to 250 days. Total carbon dioxide (£CO2), NHLt, NO2, NO3, PO4, Si,
, ZH2S, CKU, and pH will be measured over time.
Nutrient flux
The flux component of the model completes the sediment cycle by returning inorganic
nutrients to the water column. Model calibration requires extensive temporal and spatial
coverage of SOD and nutrient flux measured under ambient bottom water conditions.
Shipboard measurements of incubated intact cores will be the primary method of data
collection. Cores will be collected and maintained at ambient conditions. Some
observational measurement will be made of the effects of bioturbation. Fluxes of O2,
NH4, NO2, NOs, PO4, Si, SOs, ICO2, CHi, and ZH2S will be measured. Hydrogen
sulfide flux will be measured at stations with overlying water DO < 1.0 mg/L.
Nutrient flux studies have three sampling frequencies. Eight stations (4, 5, 9, 10, 1 1, 12,
13, 14; the SONE stations) have a sampling frequency of four periods. Three stations
(2, 7, 21) have a high sampling frequency of six sampling periods. The remaining
stations (3, 4a, 6, 7a, 8, 16, 17, 18, 20, 22, 23, 24, 25, 26) have a low sampling
frequency of two periods. Denitrification measurements have the same sampling
frequencies as nutrient flux measurements but are limited to nineteen main Bay stations.
The sampling scheme is not systematic with respect to time; rather, sampling periods are
established to coincide with an annual cycle of benthic processes. Sampling frequencies
are based on the experience gained from the SONE program. All sediment stations with
four or more sampling periods are sampled according to the following SONE sampling
periods: "(1) a period (April-May) when the early spring phytoplankton bloom occurs, and
nutrients (particularly nitrate) are high in the water column , (2) a period influenced by the
presence of a large macrofaunal community (spring-early summer), (3) a period during
which macrofaunal biomass is low but water temperature and water-column metabolic
activity are high and anoxia is prevalent in deeper waters (August), and (4) a period in the
fall when anoxia is not present and the macrofaunal community biomass is low but
reestablishing" (Boynton etal.,1987).
To improve temporal coverage of sediment nutrient flux and denitrification, three high
frequency stations have two additional sampling periods. Additional sampling periods
include: (1) a period in the winter (December - early March) when anoxia is not present,
metabolic activity is low and nitrate is increasing in the water column, and, (2) a period in
the early fall (September) just prior to the break-up of anoxia .
32
-------
The majority of the stations (14 out of a total of 25 stations) have a low sampling frequency
of two periods; one sample period when anoxia is prevalent in deeper waters (August), and
one when anoxia is not present (spring-early summer). Location and frequency of the
various field and laboratory measurements are described in Table 1.
Paired with the measurement of ambient nutrient flux are measurements of anoxic fluxes.
Anoxic fluxes are measured at all stations with the same frequency as ambient fluxes.
Anoxic fluxes are designed to measure rapid, short-term changes in flux due to the die-off
of benthic infauna and the rapid chemical changes caused by decreasing oxygen
concentrations and oxidation-reduction potential. Pore water concentrations of NH4, NOi,
NOs, PO4, Si, SOs, ICO2, CH4, U^S, Fe, Mn, and pH will be measured with short-
term anoxic incubations of surficial (6-2 cm) sediment.
Denitrification is a major sink for nitrogen in the Bay and has seasonal and region-specific
properties which must be delineated for successful sediment model development (Twilley
and Kemp, 1985). Three methods for measuring denitrification will be used. At all
stations sampled, an acetate inhibition method wUl be used with the same sampling
frequency as for nutrient flux. Acetate inhibits the final reaction (with N2 the product) of
the denitrification reaction path. This allows the analysis of an intermediate product
concentration without the problem of background contamination. Nitrification potential
will be measured at a limited number of stations by N-serve treated control sediment
slurries, an inhibition technique that prevents nitrification in the treated slurries. Control
slurries will be compared to untreated test cores. Calibration of the denitrification and
nitrification measurements will be with 15N labeled nitrate and ammonia respectively.
Details of sampling periods and times are in Table 1.
CONCLUSION
Work on acquiring data for calibrating the sediment submodel will be initiated in April,
1988. The sediment monitoring program is based on the consensus recommendations of
the expert panel of the Sediment Processes Workshop, and the Modeling and Monitoring
Subcommittees of the Chesapeake Bay Program. The sediment data outlined above are
essential to the Time Variable Model of the Chesapeake Bay.
33
-------
TABLE 1. SEDIMENT PLAN ACTIVITIES, STATIONS, AND SAMPLING
FREQUENCY.
ACTIVITY
Nutrient Flux
Denitrification
Sediment
Traps
Paniculate
Organic Profile
Pore Water
Profile
Recent Rates of
Sedimentation
Long Term
Diagenesis
METHODS
ambient
shipboard
anaerobic
transitive
acetlyene
blockage
nitrification
(N-serve)
15N
vertical array
particle traps
depth profile
of POM
pore water
concentration
pollen dating
sulfate
depletion
STATIONS
3,4a,6,7a,8, 16,17,18,
20,22,23,24,25,26
4*,5*,9*,10*,11*,12*,
13*,14*
2*,7*,2
same as above
2*,7*,21
3,4a,7a,16,18,20,21
4,14
2,7,21
2,7,21
2 stationsA
7*,21
all stations
2,3,4,4a,5,6,7,7a,8,9,10,
11,12,14,16,18,20,21,22
all stations
2,3,4,4a,6,7,7a,8, 14,
16,17**,18,20,21,22,
SAMPLING
FREQUENCY
2 periods
4 periods
6 periods
same as above
6 periods
2 periods
4 periods
6 periods
6 periods
2 periods
25 periods
1 period
1 period
1 period
1 period
23**,24**,25**,26**
* SONE stations
A Not Determined
** two sample periods for surface sediments
34
-------
REFERENCES
Boynton, W.R.; Kemp, W.M.; Garber J.; Barnes, J.M.. Sediment and water column
interaction in the Chesapeake Bay. In: State of the Chesapeake Bay Second Annual
Monitoring Report. 1987.
Brush, G.S.; Martin, E.A.; DeFries, R.S.; Rice, C.A. Comparisons of 210Pb
and pollen methods for determining rates of estuarine sediment accumulation. Quater.
Resear. 18:196-217; 1982.
Brush, G.S. Stratigraphic evidence of eutrophication in an estuary. Wat. Res. Resea.
20(5):531-541; 1984.
Garber, J.H. Benthic-Pelagic Coupling in Chesapeake Bay. In: Perspectives on the
Chesapeake Bay: Recent Advances in Estuarine Sciencies. Eds: M.P. Lynch and E.G.
Krome, Chesapeake Research Consortium. Gloucester Point, VA. 1987.
HydroQual, Inc. Development of a coupled hydrodynamic/water quality model of the
eutrophication and anoxia process of the Chesapeake Bay. EPA Contract No. 68-03-3319.
Annapolis, MD. 1987.
HydroQual, Inc. Workshop Number 2: Sediment processes and sediment modeling
workshop, December 2-3, 1987. USCOE Contract No. DACW39-88-C-0004. U.S. Army
Corps of Engineers, Baltimore District. Baltimore, MD. 1988.
Malone, T.C.; Kemp, W.M.; Ducklow, H.W.; Boynton, W.R; Tuttle, J.H.; Jonas, R.B.
Lateral variation in the production and fate of.
Twilly, R.R.; W. M. Kemp. Preliminary results on the significance of sediment
denitnfication to the fate of nitrogen in the Chesapeake Bay. Final report to the EPA
Chesapeake Bay Program. Annapolis, MD. 1985.
Wright, L.D.; D.B. Prior, C.H. Hobbs; R.J. Byrne; J.D. Boon; L.C. Schaffner; M.O.
Green. Spatial variability of bottom types in the lower Chesapeake Bay and adjoining
estuaries and inner shelf. Estuarine, Coastal and Shelf Science, Vol. 24, P. 765-784.
1987.
35
-------
-------
IMPLICATIONS OF TOXIC MATERIALS ACCUMULATING IN THE SURFACE
MICROLASER IN CHESAPEAKE BAY
Herman Gucinski(l), John Hardy(2), H. Ronald Preston (3)
(1) Anne Arundel Community College Arnold, Maryland 21012
(2) General Science Department Oregon State University
Corvalis, Oregon 97330
(3) U.S. EPA Chesapeake Bay Liaison Office,
Annapolis, MD 21403
ABSTRACT
The aquatic surface micro-layer is subject to the spontaneous,
thermodynamically driven enrichment of naturally occurring surface
active molecules (surfactants). These in turn may serve as ligands
or solvents for toxic metals, chlorinated, saturate and
polyaronatic"hydrocarbons. Toxic substances that are themselves
surface active, such as TBT (tributyltin) and its derivatives will
readily accumulate in the microlayer as well. There is a growing
body of evidence that shows enrichment ratios of one to several
orders of magnitude in samples of toxic substances taken from
estuarine, coastal, and lake waters on a worldwide scale. In
Chesapeake Bay, limited sampling has shown elevated levels of
metals and hydrocarbons (alkanes and polyaromatics) on the upper
tidal Potomac River and at three northern Bay stations
(Susquehanna, Elk, and Patapsco Rivers). Sources implicated appear
to include aerial deposition and surface run-off. Additional
sampling during autumn 1987, i.e. a time least likely to suggest
pesticide presence, nevertheless showed detectable levels of 23
different toxic organics and pesticides in upper bay microlayer
samples. Bulk water samples, by comparison, rarely had detectable
levels of the same compounds. Only few investigators have sought
to demonstrate the susceptibility of exposure to and effects on
aquatic organisms that are neustonic in at least part of their
life cycle. Preliminary work by others in the North Sea, southern
California coastal waters, and by one of us on Puget sound and in
37
-------
Gulf Stream waters near the straits of Florida showed selectively
high mortality to some indicator organisms. We have tentatively
identified eggs and larvae of the bay anchovy (Anchoa mitchilli),
atlantic silverside (Menidia menidia), hogchoker (Trinectes
maculatus) as well as the copepod (Acartia tonsa) as candidate
vertebrate and invertebrate indicator organisms for bioassays.
Suitable bioassay techniques for in-situ and laboratory testing of
microlayer water need to be developed in order to assess both the
biological significance of toxic materials enrichment and the
degree to which this enrichment is found. The resultant
information should provide further assessment relating the role of
microlayer contamination to the decline of Chesapeake Bay living
resources. If significant effects can be traced to sources
contributing toxics to the surface microlayer and if these
concentrations affect the living resources, their control measure
should be explored, evaluated and implemented.
DJTRDDUCTION
The boundary between the atmosphere and the aquatic
environment is an important biological habitat and a collection
point for pollutants. The eggs and larvae of many fish and
shellfish species float on, or come in contact with, the water
surface throughout their early development. The aquatic surface
microlayer (surface microlayer), operationally defined as 50-100
um thick, serves as a concentration point for metal and organic
contaminants that have low water solubility or are associated with
floatable particles. Recent studies have linked aquatic surface
contamination with negative biological inpacts. In Puget Sound
(Hardy et al., 1988a), Southern California (Cross et al., 1987),
and the North Sea (Kocan et al., 1987), fish eggs exposed to
contaminated surface microlayer exhibited reduced viability.
We report here work on the evaluation of microlayer samples
taken from six stations in the upper Chesapeake Bay (Figure 1) and
identify living resources potentially impacted. Additionally, we
explore necessary features of a plan for assessing the biological
significance of the observed enrichment of toxic contaminants.
Organic molecules, collectively called surfactants
(surface-active agents), are thermodynamically driven to remain at
the interface because they lower the surface free energy. Dominant
molecules appear to be biogenic, are'long-chain, and of high
molecular weight. The enriched interface is capable of trapping
other molecules, both dissolved and particulate, toxic and benign.
This matrix forms a substrate for bacterial growth as seen by 1 to
4 orders of magnitude higher bacterial counts in the microlayer
than the subsurface water. In estuarine waters the time scale for
such initial enrichment is short, a time frame of hours or tens of
hours is likely (Gucinski, 1985, 1986; Olson, 1983; Crow et al.,
1975; Sieburth, 1982; Hartwig and Herr, 1984).
38
-------
Many taxa, known to be neustonic offshore, also occur-in the
Bay. Table 1 is a listing of taxa clearly found to be neustonic
in marine waters of the Mid-Atlantic Bight (Grant, 1979).
Comparing his findings to species lists published by Wass (1972)
and Lippson et al. (1979), one can identify those species
occurring in the Bay, and putatively classify them as neuston as
well. Exceptions do exist. For example, the larval stage of the
blue crab (Callinectes sapidus) is neustonic in offshore waters,
but once transported into Bay waters, its zoeae and megalopae are
pelagic (Provenzano et al., 1982; 1983). Marine neuston studies
cannot be extrapolated to the low salinity zones of the Bay.
Clearly, a need exists to characterize the neuston of Chesapeake
Bay.
On the mid-Atlantic shelf the larvae of commercially valuable
species such as menhaden, hake, cod, bluefish, lobster, and blue
crab, occur in greater concentrations in surface water than in
subsurface water (Grant, 1979, Castagna, 1977). The copepod
Acartia tonsa, various species of amphipod, the bay anchovy,
atlantic silverside and hogchocher, are all known to have egg,
larval, or adult stages that contact the surface microlayer. The
exposure risks that these selected species undergo when toxic
materials are enriched or resident in the microlayer need to be
assessed. In addition, to estimate exposure duration, information
is needed on the residence times in the microlayer for
zcoplankton, larvae, and eggs.
Studies conducted in the spring of 1986 (Hardy et al., 1987)
suggested that serious surface microlayer contamination,
consisting of complex mixtures of chemicals, occurs in Chesapeake
Bay. Contamination levels suggested the presence of three major
zones —the upper Bay, with high levels of contamination; the
Potomac River with moderate to high levels; and the southern and
eastern shore, with a different chemical mixture and generally low
levels of contamination. Our data result from autumn sampling and
cover an extended range of contaminants identified as potentially
harmful by the EPA.
We hereby gratefully acknowledge support of the EPA
Chesapeake Bay Program and the following : L. Antrim, R. Smith, K.
Durrell, R. Allard, T. Wilson, S. Jordan, M. O'Malley, E.
Crecelius, T. Fortonan, W.Steinhauer.
METHODS
Neuston collections were limited to waters of the upper
Chesapeake Bay (entirely within the state of Maryland) and did
not include night-time tows. The rectangular mouth (0.46 x 0.3 m)
zooplankton net, had a mesh size of 100 um and sampled a
surface-water volume of 9 cubic meters when immersed to an average
depth of 0.05 m. In a few cases the volume sampled was estimated
39
-------
at about 4 cubic meters rather than calculated. Samples were
preserved in buffered 5% formalin for later analysis.
Saitples of the sea-surface microlayer were collected from 6
sites in Chesapeake Bay between Septetttoer 10 and 12, 1987 using
the rotating teflon drum microlayer sampler and repeated during
October, 1987. because of the loss of some samples in shipment.
Methods followed in general those described previously for similar
samples (Hardy et al., 1986; Hardy et al., 1987; Hardy et al.,
1988). In addition, samples were analyzed for concentrations of
tributyl tin by hydride generation and atomic absorption
detection. Analyses performed at EPA contract labs followed
standard methodology as specified by the Office of Water
Regulation and Standards.
Microlayer sampler collection efficiency determinations were
made with Lycopodium spores, a hydrophobic particulate
representative of aerially deposited materials and radio-labelled
14C DDT dissolved in oleyl alcohol, used to represent a dry
surfactant that forms monolayers.
Table 2 summarizes the results for particulate recovery, done
on seawater, except for a single run on tapwater at 0 mN/m
spreading pressure.
Counts of 14C DDT indicated a recovery which increased with
increasing film pressure and was 78.7, 80.3 and 88.2%, for mean
surface pressures of 4.6, 19.6 and >23 mN/m, respectively.
Collection efficiencies of 2 microlayer samplers tested here are
less than 100%, and vary both as to the substance recovered as
well as with the presence of slick forming molecules, as
determined by surface pressure or surface tension measurement.
Within slicks, such efficiencies exceed 85%, and are much less,
from 60% to nearly 80% when surface waters are "free" of
organized, coherent films.
RESULTS
Our analysis of neuston tows collected this summer and autumn
from four sites in Chesapeake Bay indicates the presence of at
least 20 abundant taxa dominated by the copepod Acartia tonsa
(Table 3). All tows were conducted in daytime and only
zooplankters were collected. The limited scope of the study (no
replicate sampling) and late season of the sampling could not
comprehensively represent overall neuston abundance and diversity.
Results indicate that surface-dwelling organisms occur at very
high densities in the areas sampled with a mean density > 7000
individuals/cubic meter. For comparison, densities of total
zooneuston in Puget Sound collected with a similar net were about
100 to 400 individuals/cubic meter with copepods dominating the
community. In Chesapeake Bay, the copepod Acartia tonsa could
represent an important prey item for surface feeding fish or other
organisms.
40
-------
Samples collected during September and October 1987 (Station
locations are shewn in Figure 1) showed elevated concentrations of
metals in the raicrolayer. Of particular interest, in terms of
potential toxicity, were the concentrations and/or enrichments
(microlayer/bulkwater concentrations) of silver (indicative of
sewage inputs) at Stations 7, 8, and 12, copper at all stations,
and arsenic, lead and zinc at Station 3. The total microlayer
concentrations of Ag+Cu+Cd+Pb+Zn ranged from 59 to 642 ug/1. We
are presently reconfirming the results of these measurements.
Pesticides and other organic compounds were enriched In the
microlayer compared to the bulkwater samples at several sites.
Microlayer concentrations are shown in Table 4. Enriched
microlayer concentrations occurred at two or more stations for the
following: carbophenothion, demeton, diazinon, di-butyl
phthalate, EPN, ethion, famphur, fensulfothion, and kepone. In
general, Stations 7 and 8 were most contaminated; e.g. the
microlayer at Station 8 was particularly enriched in dieldrin.
Dieldrin has been found in urban run-off at concentration from 8
to 100 ng/1, values in the same range as seen here (EPA, 1982).
October 1987 samples (Table 5), analyzed by Battelle, had
high concentrations of organic contaminants in the microlayer at
most stations. Concentrations of organotin ranged from 30 to 349
ng/1 in the microlayer and 60 to 90 ng/1 in the two bulkwater
samples analyzed.
Concentrations of total aromatic hydrocarbons in the surface
microlayer of Chesapeake Bay ranged from 0 to 20 ug/1. Table 6
lists concentrations at or above the detectable level. Spike
recovery measurements on the samples using surrogate aromatic
hydrocarbons suggested that only 35 to 100% (mean 69%) of the
aromatic hydrocarbons were recovered, i.e. our reported values
represent roughly 69% of the actual concentrations present in the
sample. Aromatic hydrocarbon concentrations were low or below
detection at Stations 2 and 3, significant (potentially toxic) at
Stations 7, 11, and 12, and very high at station 8 (Susquehanna
River). Concentrations of total saturate hydrocarbons ranged from
3.8 to 66.5 (mean 21.3) ug/1 (Table 7). Highest concentrations
occurred at Stations 3, 7 and 8.
Pesticide and chlorinated organic compounds were largely
undetected in surface microlayer samples taken in October, with
the exception of dieldrin. Dieldrin occurred in concentrations of
1 to 18 ng/1 except at Station 2, where it was absent. As was the
case with most of the other contaminants, high concentrations of
pesticides were found at Stations 3, 7, and 8.
DISCUSSION
In recent years, toxicity tests of sediment contamination
have involved the development of an environmental quality triad
(chemical, bioassay and infauna) to determine environmental impact
41
-------
{Long and Chapman, 1985). Chapman and Long (1983) argued that for
accurate evaluation of sediment quality, at least three categories
of measurement must be evaluated. These are: (I) concentrations
of toxic chemicals, (2) toxicity of the environmental samples
(bioassay), and (3) evidence of modified resident biota, -
particularly the infauna.
The same approach can be used for determining the effects of
surface microlayer contamination on neustonic eggs and larvae.
Thus, accurate evaluation of surface waters may involve at least
five categories of sampling, testing, and evaluation. These are:
1) collection of surface microlayer samples, 2) determination of
concentrations of toxic chemicals, 3) collection and. enumeration
of representative resident neuston species populations, 4)
toxicity tests (laboratory bioassays), and 5) toxicity to repr-
esentative neustonic organisms (field bioassays). Certainly all
five measures are necessary to get an accurate picture of the
physical, biological and chemical parameters that contribute to
aquatic surface quality. At this stage, the toxicity evaluation
should involve both laboratory bioassays with standard organisms,
such as the sea urchin, and selected field bioassays, with
important resident organisms such as the bay anchovy or crab
larvae. To determine the toxicity of surface microlayer in
Chesapeake Bay a dual approach, consisting of both in-situ and
laboratory bioassays may be useful. In-situ studies simulate
natural conditions, but are often unable to determine controlling
variables (e.g., temperature and salinity). Organisms for field
bioassays are not always available which leads to unproductive
field time. If no difference exists in the results of toxicity
tests using fresh versus frozen surface microlayer samples, the
samples could be collected throughout the year and tested on
seasonal spawning species when eggs are available. Nevertheless,
laboratory studies are needed because they allow better control
and provide a basis for accurate hypothesis testing. Both types
of bioassay are necessary.
Floating pelagic fish eggs are particularly suitable for
studies of aquatic surface toxicity. For example, eggs of anchovy
(Hunter, 1981), sole (Hardy, 1987), and mackerel (Longwell, 1976,
1980) have been used successfully as sensitive indicators of
toxicity. The eggs of such pelagic spawners are often distributed
in extremely patchy but dense concentrations. Eggs are frequently
present in only 5% of the neuston net trawls, but when found, are
often in densities of 17 to 31 eggs/L. This is the equivalent of
up to 46,000 eggs per 10 square meters of water surface. These
patches originate from intensive spawning activity and gradually
disperse (Hunter, 1981).
Two fish, in particular, are important ecosystem components
and produce, floating eggs in large numbers. Neuston net tows at
the South Island of the Chesapeake Bay Bridge-Tunnel indicated
maximum egg densities in mid-June and mid-July for the hogchoker
(Trinectes raaculatus) and the bay anchovy (Anchoa mitchilli),
(Birdsong, pers. comm.). Blue crab (Callinectes sapidus) occurred
42
-------
in high densities, near the surface in mid-July to raid-August.
Also, the zoeal larvae of the blue crab concentrate at the surface
and can be cultured in the laboratory. We recommend the bay
anchovy and the blue crab larva as appropriate species for
assessing surface microlayer toxicity in Chesapeake Bay. The
anchovy egg is a representative pelagic fish egg that contacts the
surface during an approximately four-day period during
development, is widespread and euryhaline, and can be collected in
large numbers during the summer using neuston net tows. Blue crab
larvae represent the reproductive stage of an important commercial
shellfish resource. They are typical of crustacean neuston that
probably feed on the high densities of microorganisms at the water
surface (Zaitsev, 1971). Also, they can be cultured and used for
toxicity tests in the laboratory.
Our research in Puget Sound suggested that toxicity to
pelagic fish eggs and other organisms resulted from a complex
mixture of contaminants, with no single compound or group of
compounds responsible for the overall toxicity (Hardy et al.,
1988). We do not yet have toxicity measurements for Chesapeake
Bay microlayer. To obtain a relative measure of toxicity one may
enter the data from this study into our microlayer toxicity model
(Hardy et al., 1988). The results of the cumulative impact
suggest surface contamination in Chesapeake Bay may be responsible
for a reduction in the survival of neuston, including the hatching
success of pelagic fish eggs. Predicted toxicity would be highest
at stations 3, 7 and 8. This estimate, based on a limited data
set, is uncertain, but is probably conservative because it does
not take into account the possible effects of the organotin found
in our samples. A conprehensive study, including simultaneous
measurements of toxicity and concentrations of contaminants,
should be conducted in Chesapeake Bay.
43
-------
REFERENCES
Cross, J.N., Hardy, J.T., Hose, J.E., Hershelman, G.P., Antrim,
L.D., Gossett, R.W. and Crecelius, E.A, 1987. Contaminant
concentrations and toxicity of sea-surface microlayer near
Los Angeles, California. Mar. Environ. Res. (in press).
Crow, S.A., D.G. Ahern, W.L. Cook, and A.W. Bourquin. 1975.
Densities of bacteria and fungi in coastal surface films as
determined by membrane adsorption procedure. Limnol.
Oceanogr. 20:644-646.
EPA. 1982. Results of the Nationwide Urban Run-off Program. Vol
II, Appendices. Washington, D.C.
Grant, G.C. and J.C. Olney, S.P. Berkowitz, J.E. Price, P.O.
Smyth, M. Vecchione, and C.J. Womack. 1979. Middle Atlantic
Bight zooplankton: second year results and a discussion of
the two-year BLM-VIMS Survey. Virginia Institute of Marine
Science, Gloucester Point, Virginia 23062. Special Report
in Applied Marine Science and Ocean Engineering, No. 192.
Chapter 4, 236 p.
Gucinski, H. 1985. . Correlation of biophysical surface
characteristics with hydrodynamic properties of adhesive
biofilms. Ph.D. Thesis, Roswell Park Mem. Inst. SUNY at
Buffalo.
Gucinski, H. 1986. The effect of sea surface microlayer
enrichment on TBT transport. Oceans '86. Proceedings Vol. 4
Organotin Symposium, pp. 1266-1274.
Hardy, J.T., E.A. Crecelius, L.D. Antrim, S.L. Kiesser, and
V.L. Broadhurst. 1987. Aquatic Surface Microlayer
Contamination in Chesapeake Bay. Contract Report to Maryland
Department of Natural Resources, Energy Administration, Power
Plant Research Program, Annapolis, MD 39 p.
Hardy, J.T., S. Kiesser, L. Antrim, A. Stubin, R. Kocan,
and J. Strand. 1988a. The sea-surface microlayer of Puget
Sound: ^Part I. Toxic effects on fish eggs and larvae. Mar.
Environ'. Res. 23.
Hardy, J.T., E.A. Crecelius, L.D. Antrim, V.L. Broadhurst,
C.W. Apts, J.M. Gurtisen, and T.J. Fortman. 1988b. The
sea-surface microlayer of Puget Sound: Part II.
Concentrations of contaminants and relation to toxicity.
Mar. Environ. Res. 23.
Hardy, J.T., J.A. Coley, L.D. Antrim, and S.L. Kiesser.
1988c. A hydrophobic large-volume sampler for collecting
aquatic surface microlayers: characterization and comparison
to the glass plate. Can. J. Fish. Aquatic Sci. 45(5).
Harts/ig, E.O. and F.L. Herr. 1984. Chemistry and biology
of the sea-surface interface relationships to remote sensing.
Off. Naval Res. Workshop, Sanibel Island, Florida, Oct.
23-25, 1984.
Hunter, J.R. 1981. Feeding ecology and predation of marine
fish larvae. In: R. Lasker (ed.), Marine fish larvae, pp.
33-77. Washington Sea Grant Program.
Kocan, R.M., M.L. Landolt, and K.M. Sabo. 1982. Anaphase
aberrations: a measure of genotoxicity in mutagen-treated
fish cells. Environ. Mutag. 4:181-189.
44
-------
Lippson, A.J., M.S. Haire, A.F. Holland, F. Jacobs, J. Jensen,
R.L. Moran-Johnson, T.T. Polgar, and W.R. Richkus. 1979.
Environmental atlas of the Potomoc Estuary. Prepared for the
Maryland Department of Natural Resources, Power Plant Siting
Program, by Martin Marietta Environmental Center, Baltimore,
Maryland, 280 p.
Longwell, A.C. and J.B. Hughes. 1980. Cytologic,
cytogenetic and developmental state of Atlantic mackerel eggs
from sea surface waters of the New York -Bight, andprospects
for biological effects monitoring with ichthyoplankton.
Rapp. P.V. Reun. Cons. Int. Explor. Mer, 179:275-291.
Olson, M. 1983. Surface energies and chemical analysis of
the initial stages of marine microbiological fouling.
Trident Scholar Proj. Rep. No. 127, US Naval Academy,
Annapolis, MD.
Provenzano, A.J., Jr., J.R. McConaugha, K.B. Philips, D.F.
Johnson, and J. Clark. 1983. Vertical distribution of first
stage larvae of the blue crab, Callinectes sapidus, at the
mouth of Chesapeake Bay. Estuarine Coastal Shelf Sci.
16(5):489-499.
Provenzano, A.J., Jr., J.M. McCanaugha, and D.F. Johnson.
1982. Significance of the neuston^ layer in the dispersal of
larvae of the blue crab Callinectes sapidus. J. Shellfish
Res. 3(1):99.
Sieburth, J.M., 1982. Microbiological and organic-chemical
processes in the mixed layer and the surface skin of the sea.
Nato Adv. St. Symp., Durham, NH, July 1982.
Wass, M.L. (Ed.). 1972. A checklist of the biota of lower
Chesapeake Bay. Special Scientific Report No. 65. Virginia
Institute of Marine Science, The College of William and Mary,
Gloucester Point, VA 290 p.
Zaitsev, Y.P. 1971. Marine neustonology. Acad. Sci. Ukr.
SSR. (Trans, from Russian). National Marine Fisheries
Service-and the National Science Foundation, Washington,
D.C., National Technical Information Service, Springfield,
Virginia. 207 pp.
45
-------
TABIE 1: PELAGIC ZOOPLANKTON POUND IN NEOSTON TOWS
AND ALSO POUND IN LOWER CHESAPEAKE BAY (after
Grant, 1979 , Wass, 1972 , Lippson et al, 1979)
CNIDERIA (comb jellies)
Liriope sp.
COPEPODA (zooplankton)
Acartia sp.
Acartia tonsa
Calanus finrnarchicus
Candacia armata
Centropages furcatus
Centropages hamatus
Centropages typicus
Corycaeus sp.
Eucalanus sp.
Eucalanus pileatus
Labidocra sp.
Labidocera aestiva
Oithona spp.
Oncaea venosta
Paracalanus parvus
Paracalanus quasimodo
Pontella sp.
Pontella meadii
Pseudocalanus sp.
Temora longicornis
Temora stylifera
Temora turbinata
ECHINODEPMATA (including
seastars, etc.)
unident. cphiuroids
CHAETOGNATHA (arrow worms)
Sagitta elegans
Sagitta enflata
Sagitta hispida
Sagitta tenuis
unident. chaetognaths
MOLLUSCA (mussels & clams)
Dosinia discus
Melampus bidentatus
Spisula solidisima
MYS3DACEA (mysid shrimp)
Heteromysis formosa
Mysidopsis bigelowi
Necmysis americana
OJMACEA
Oxyurostylis
ISOPODA
Edotea triloba
Idotea metallica
AMPHIPODA
Corophium sp.
Stenothoe sp.
DECAPODA (principally crabs)
Callianassa sp.
Callinectes sp.
Crangon septemspinosa
Homarus americanus
Homola barbata
Latreutes fucorum
Leptochela sp.
Li
oinia sp.
Munida sp.
Pinnixa cylindrica
Portunus sp.
PISCES (fish, eggs,
larvae)
Astroscopus guttatus
Cynoscion regalis
Hippocampus sp.
Menidia menidia
Scomberesox saurus
Scophthalmus aquosus
Sphoeroides sp.
Syngnathus sp.
Syngnathus fuscus
Urophyics sp.
Urophycis regius
46
-------
TABLE 2: LYCOPODIUM SPORE SAMPLING EFFICIENCY AS A
FUNCTION OF SPREADING PRESSURE
Surface Spreading Pressure mN/m
Sampler 0 <0.82 4.4 18.8
Drum 64.4 18.4* 74.3 89.8
Glass plate 31.1 55.2 37.6 99
*Visual observation shaved spores pushed away from sampler,
suggesting some surfactant contamination that produced
its own spreading pressure.
TABLE 5: BUTYL TIN (ng/1 AS INORGANIC TIN) IN CHESAPEAKE
BAY MICROLAYER AND BULKWATER SAMPLES COLLECTED
OCTOBER, 1987
Location
*Baltimore Harbor
Susquehanna River
Upper Potomac
*Matapeake
*Choptank River
*Point Lookout
Butyl Tin
(ng/1)
Microlayer Bulkwater
80
299
349
200
70
30
90
60
*Data requires additional verification.
47
-------
TABLE 3: NEOSTON (XJJCENTRATIONS IN THE UPPER CHESAPEAKE BAY
Choptank River
Number,
TAXA: per m
Acartia tonsa 9362
copepod nauplii 942
barnacle nauplii 626
Bosmina
longirostris 312
Caitptocercus
rectirostris 208
insecta 184
shrimp larvae 184
Moina micrura 184
Podon
polyphemoides 92
Chydorus sp. 92
Matapeake
Number-
TAXA: per m
Acartia tonsa 2167
copepod nauplii 557
Bosmina
longirostris 1052
Podon
polyphemoides 62
Cyclops vernalis 62
Cyclops
bicuspidatus 371
Diaphanosana 62
Elk River
Number-
TAXA: per m
Acartia tonsa 495
Diaphanosana 21
Susquenanna River
Number-
TAXA: per m
Acartia tonsa 4800
insecta 1280
Chydorus sp. 320
Cyclops vernalis 2880
unident'd, dmaged 3200
Percent
Total
76.71
7.72
5.13
2.56
1.70
1.50
1.50
1.50
0.75
0.75
Percent
Total
50.01
12.85
24.28
1.43
1.43
8.57
1.43
Percent
Total
95.38
4.05
Percent
Total
38.46
10.26
2.56
23.08
25.64
48
-------
TABLE 4: CONCENTRATIONS (ug/1) OF PESTICIDES AND ORGANIC
COMPOUNDS IN THE SURFACE MICROLASER AND BULKWATER OF
CHESAPEAKE BAY SEPTEMBER, 1987. ANALYZED BY EPA.
M=*HCROIAYER, B=BULK SEA WATER
Station:
Compound:
Baltimore Susquehanna Matapeake Potomac
Harbor River River
M
B
M
B
M
B
M
B
azinphos ethyl
azinphos methyl
captafol
carbophenothion
chlorfevinphos
coumaphos
crotoxyphos
demeton
diazinon
dichlorvos
m-di-butyl-
phthalate
EPN
ethion
famphur
fensulfothion
kepone
leptophos
monocrotophos
phosmet
terbufos
tetrachlorovinphos
thio-bis-methane
trichlorofon
nd*
nd
<0.2
nd
<0.5
nd
<1
nd
<0.5
nd
50
<0.5
<2
nd
<1
nd
nd
nd
<1
nd
nd
nd
nd
nd
nd
<0.2
nd
<0.5
nd
<1
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
<1.2
nd
nd
nd
<1!
~ <1
nd
<0.5
<0.5
nd
<1
~ <1 <
nd
_ <0.5
nd
nd
<2
_ <0.5
<1
~ <0.12
<0.5
nd
_ nd
<1.2 <
nd
nd
-
-------
TABLE 6: AROMATIC HYDROCARBONS IN THE SURFACE MICROLASER
OF CHESAPEAKE BAY, OCTOBER, 1987
Baltimore Choptank
Compound: (ug/1)
Phenanthrene 0 . 1
Fluoranthene 0 . 2
Pyrene 0 . 1
Benz (a) anthracene 0.0
Crysene 0.0
Benzo (k) f luoranthene 0 . 0
Benzo (e) pyrene 0.0
Total (all compounds) 0.4
X = Present, but below detection
(ug/1)
0.0
0.1
X
X
X
X
X
0.1
limit of
Matapeake
(ug/1)
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.05 ug/1.
Susquehanna River
Conpound: (ug/1)
AcenaphtheneOTI
Fluorene 0.4
Phenanthrene 3.4
Anthracene 1.0
C1-P/C1-A 1.0
Dibenzothiophene 0.1
Fluoranthene 5.3
Pyrene 3.5
Benz(a)anthracene 1.4
Chrysene 2.8
Benzo(b)fluoranthene 0.8
Benzo(k)fluoranthene 0.1
Benzo(e)pyrene 0.2
Benzola)pyrene 0.1
Total (all ccrpounds)20.2
AROMATIC HYDROCARBONS SCANNED EOT NOT DETECTED:
Naphthalene Cl-Fluorene Cl-D
Cl-Naphthalene C2-Fluorene C2-D
C2-Naphthalene C3-Fluorene C3-D
C3-Naphthalene C4-Fluorene C4-D
C4-Naphthalene C2-P/C2-A Benzo(g,h,i)perylene
Acenaphthylene C3-P/C3-A Perylene
Biphenyl C4-P/C4-A Indeno(l,2,3-CD)pyrene
Dibenz(a,h)anthracene
50
-------
TABLE 7:
O2X3NTRATICNS (ug/1) OF SATURATE HYDROCARBONS IN
SURFACE MICROLASER OF CHESAPEAKE BAY, OCTOBER 1987
Compound:
Station:
11 12A 12B Blank
Heptadecane
Pristane
Octadecane
Phytane
Nonadecane
Eicosane
Henicosane
Docosane
Tricosane
Tetracosane
Pentacosane
Hexacosane
Heptacosane
Octacosane
Nonacosane
Triacontane
Hentriacontane
Dotriacontane
Tritriacontane
Tetratriacontane
DTP
Isopronoid 1380
Farnesane 1470
Isopronoid 1650
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
2,
0,
0,
0,
0,
0,
0,
0,
0,
0,
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.9
.1
.1
.0
.8
.0
.0
.0
.0
.0
.0
.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.3
0.2
0.7
0.3
1.9
0.2
4.5
0.3
2.4
0.0
0.2
0.0
0.0
0.0
0.0
0.0
0.2
0.4
0.5
0.0
1.2
1.1
2.3
5.5
9.8
11.8
10.7
7.8
4.8
3.2
3.4
1.2
1.8
0.2
0.7
0.0
0.0
0.0
0.0
0.0
0
0
0
0
0
0
0
0
1
2
4
2
6
1
18
1
10
0
1
0
0
0
0
0
.0
.0
.0
.0
.0
.0
.3
.6
.9
.0
.2
.1
.3
.6
.3
.5
.5
.4
.7
.0
.0
.0
.0
.0
0.0
0.0
0.0
0.0
0.1
0.3
0.4
0.4
0.2
0.0
0.0
0.3
0.5
0.2
1.1
0.5
0.9
0.0
0.2
0.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.8
0.0
0.9
0.0
0.0
0.0
2.9
0.0
0.0
0.0
0.1
0.3
0.2
0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.0
0.4
0.4
1.0
0.0
1.1
0.0
0.7
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
Total 3.9 11.0 66.5 51.2
Mean for All Stations 24.0
5.8 7.0 3.8 0.0
Station # Location:
2 Point Lookout
3 Upper Potomac
7 Baltimore Harbor
8 Susquehanna River
11 Choptank River
12a Matapeake
12b Matapeake, Bulkwater
Scanned but not detected:
Decane Undecane
Dodecane Tridecane
Tetradecane Pentadecane
Hexadecane
51
-------
Rgure 1. Station Locations
Susquehanna
River fl
52
-------
POTENTIAL BIOLOGICAL EFFECTS OF MODELED WATER QUALITY
IMPROVEMENTS RESULTING FROM TWO POLLUTANT
REDUCTION SCENARIOS
Kent Mountford, PhD
Senior Scientist
U.S. EPA Chesapeake Bay Program Liaison Office
Annapolis, MD 21403
Robert C. Reynolds
Programmer/Analyst
Computer Sciences Corporation
Chesapeake Bay Program Liaison Office
Annapolis, MD 21403
INTRODUCTION
The use of water quality models to make environmental projections is popular in contem-
porary management circles. It is important, however, to relate model output in practical terms
to the attainment of beneficial uses; especially restoration of both a balanced ecological
system and the recreational and commercial resources which make man eager to pay the bill
for his transgressions. This paper is an approach to satisfy that immediate need in
Chesapeake Bay and is companion to App and Fitzpatrick (1988) elsewhere in this volume.
In an effort to efficiently allocate limited resources, the Environmental Protection Agency
(EPA), together with a cooperative state/federal management structure called the Chesapeake
Bay Program (CBP), sponsored development of a state of the art steady-state coupled hydro-
dynamic/water quality model (HydroQual, Inc. 1987). This computer model, known as the
Chesapeake Bay Program Steady-State Model, is used as a management tool to calculate and
project water quality conditions based on a number of management scenarios. The model
develops estimates of selected environmental conditions in the Bay and presents them as an
average summer "steady-state" condition (defined as the 62-day period covering July and
August). Many users and reviewers of the voluminous and complex model output have had
difficulty relating this product to meaningful improvements in the Bay's living resources.
In this paper, we present some of the mechanisms by which the estuarine ecosystem and
principally the benthos could respond to water quality improvement as restoration of
Chesapeake Bay progresses. The proposed approach is general rather than scenario-specific,
and can be applied to future runs of this and other water quality models. The results of these
processes, as they might be expressed in the natural system, are based on the best available
estimates made by qualified scientists in the areas of their expertise.
ACKNOWLEDGEMENTS:
The estimation of living resource effects which have resulted or might result from
changing water quality in Chesapeake Bay is not a unique idea. In particular, relating
53
-------
hypoxic volume and the exposure of benthic habitat to low oxygen was explored by Taft and
others (USEPA 1983a). In this paper, we emphasize that with respect to the model and
habitat change estimates we are working with the excellent data of other researchers. Our
contribution is to bring this work together to create a better understanding of potential
environmental response. Any errors attendant to this process are ours alone and should not
reflect on the primary investigators who have shared their ideas with us.
In particular, we wish to acknowledge detailed discussions with the following scientists and
researchers: Fred Holland and Mary Tyler, Versar/ESM; Denise Breitburg and Kevin
Sellner, Academy of Natural Sciences of Philadelphia; Jonathan Garber, Roger Newell and
Larry Sanford, University of Maryland; Jim Fitzpatrick, HydroQual, Inc.; Steve Jordan,
Maryland Department of Natural Resources; Charles App, Lewis Linker, Alan Beck and
David Hanson, U.S. Environmental Protection Agency. Charles Spooner, EPA/CBP
Director and Technical Coordinator Edward Stigall, generously supported us in accomplish-
ing this work. Nina Fisher, Computer Sciences Corporation., produced the graphics and
layout.
THE PROBLEM AND OUR STRATEGY
Two principal nutrients, nitrogen and phosphorus, have been strongly associated with the
decline of environmental quality in Chesapeake Bay; specifically with eutrophication and the
resulting expansion of deep water hypoxia (USEPA, 1983b). Despite efforts to reduce
nutrient loads to the estuary, there is a continuing need for controls on both point and non-
point nutrient sources, a need recognized in the 1987 Chesapeake Bay Agreement which
commits to a Baywide 40% reduction of both nitrogen and phosphorus by the year 2000.
The escalating cost for such controls, especially at a time when federal deficits demand fiscal
restraint, make decisions on the allocation of limited resources extremely painful. For analy-
sis, we have chosen to use one of the model calibration years —1984. This was a relatively
high freshwater flow year which resulted in strong vertical water column stratification and
severe deep water oxygen depletion (Seliger et al. 1985). Over the years, the distribution
and duration of severe hypoxia has defined the survival and habitat range for many
ecologically important species in the Bay (N. Mountford et al. 1977).
We will evaluate potential effects on habitat conditions which could result from imple-
menting either of two control scenarios of primary interest; a 40% reduction in total phos-
phorus loadings basin-wide (TP) and a 40% reduction in both total nitrogen and total phos-
phorus (TN+TP). Phosphorus and nitrogen removal alternatives are, of course, separated
by a substantial incremental cost to basin taxpayers. At the same time, the improvements
suggested by the model output are perceived by many agency decisionmakers to be relatively
modest. For this reason, in evaluating the scenarios we will portray potential environmental
results in terms of the incremental benefit from upgrading treatment from TP to TN+TP.
Whenever we seek to "force" the system a small distance further along the path of restora-
tion, the apparent resistance is relatively great and the apparent benefits not so dramatic as
the previous increment. We suggest that biological mechanisms inherent in the Bay may
loop back and enhance the relatively modest changes implied by the model results, produc-
ing improvements of real environmental significance.
We will deal with two principles, both of which will be expanded in turn:
1. The model indicates oxygen distribution with depth in the water column
would improve as the proposed pollution controls are implemented,
resulting in increased habitat suitable for living resources.
2. Such improvements in habitat with depth could result in pycnocline
tilting events impacting less benthic habitat than occurs under present
conditions in the Bay.
54
-------
HABITAT EXPANSION FROM IMPROVEMENTS IN OXYGEN
DISTRIBUTION AT DEPTH
The model estimates dissolved oxygen (DO) for each segment and at different depths in the
water column. For current conditions, these estimates represent quite well the observed
summer oxygen at different depths in Chesapeake Bay. For various pollution control scena-
rios these projections change, generally producing higher oxygen values which penetrate
deeper in the water column. This pattern is shown schematically (Figure 1) for two hypo-
thetical conditions. When acceptable DO conditions extend to an increased depth according
to model calculations, a strip or perimeter of bottom previously exposed to unfavorable con-
ditions could become suitable habitat.
Figure 1: Schematic representation of incremental habitat improvement "A" scenario: unfavorable
conditions extend high in the water column. "B" scenario: unfavorable conditions are re-
duced and are found only at greater depth. "A-B" differential: represents an increment of
Bay bottom which is now potential new habitat
Figure 2 is a hypsographic curve summarizing the bottom area of several hundred modeled
segments in Chesapeake Bay. Starting with segments having a modeled depth of 20 m, the
total additional bottom area encountered by rising up each meter in the water column can be
read along the abcissa. At right is the entire modeled area of the estuary, 7500 km2. The
"shape" of this curve is typical for water bodies with a relatively constrained deep region and
rapid rise to a broad "shelf region of substantially shallower depth.
The modeled structure of Chesapeake Bay superimposes "boxes" of varying dimension on
the Bay's natural contours. This scheme does not capture accurately the extremely deep
holes and narrow trench areas which exceed 20 m and even 30 m in a number of areas. The
estuary is, however, modeled (HydroQual, Inc. 1987) to contain the same volume as
55
-------
HYPSOGRAPHIC CURVE
en
uj 5
z
i
ID
Q
10
15
20
THOUSANDS OF SQUARE KILOMETERS
BOTTOM SURFACE AREA BELOW A GIVEN DEPTH
Chesapeake Bay. Cronin (1971) gives the accumulated area of the Bay as 6495 km2 but
when, as in the model, Tangier Sound (771 km2) and Pocomoke Sound (379 km2) are
added total area is 7645 km2, within 1.9% of the model val.ue (7500 km2). The model,
therefore, fairly represents the overall distribution of potential benthic habitat area at succes-
sive depths in the real estuary.
For this paper, output was structured to present bottom area at each of 73 unique depths
which occur in the model structure, and summer average oxygen exposure was calculated
for each spatial habitat element The matrix for 1984 and the two management scenarios
under consideration has about 2000 entries. We have therefore used Occam's razor and
summarized the essential elements as Table 1. This table presents summer average exposure
of benthic habitat to varying degrees of hypoxia and DO levels equal to the state water
quality standards for both the calibration year and scenarios of 40% load reduction in TP
and TP+TN. The expected instantaneous minimum DO can be estimated (HydroQual, Inc.
1987) from any summer average value using the relationship:
D0(min) = 0.887 D0(obs.mean)-0.937
Since this regression accounts for 92% of observed variance in the source data from
Chesapeake Bay monitoring (cf. Mountford and Mackieman 1986), we believe summer
averages (represented by the specified ranges) are a reasonable depiction of habitat
conditions for the living resources of concern.
Model output can also be interpreted as water volumes with summer average DO equal to
the specified range. These volumes can be compared among scenarios to estimate
incremental improvements in habitat available to pelagic organisms (e.g. rockfish and bay
anchovy which swim free in the water column). Such an analysis is not included in the
present paper but may be of substantial environmental significance (Coutant, personal
communication).
Vertical distribution of oxygen exposure
Data from Table 1 have been grouped in accordance with the ranges proposed above and are
graphically displayed as a case comparison in Figures 3 A, B, and C (see page ). These
56
-------
histograms represent aggregation into 1 m depth increments and are shaded to depict the
severity of hyppxia. The denser the shading, the greater the environmental stress at a given
depth and the lighter the shading, the greater the improvement. The abcissa reads in square
kilometers of bottom surface area exposed to the indicated condition.
The 1984 calibration year used for this comparison clearly has the least favorable condition.
Simple inspection shows the reduction in extreme hypoxia forecasted to result from
imposition of the 40% TP reduction scenario. In the 40% TP+TN scenario for the summer
average condition elimination of extreme hypoxia is projected. Other categories of reduced
oxygen are proportionately improved.
Table 1
Chesapeake Bay bottom (km2) exposed to varying degrees of hypoxia for 3
scenarios run on the summer-averaged Steady-State model
Range of Summer Average Dissolved Oxygen Exposure (mg/1) *
Management 0-0.5 0.5-1.0 1.0-1.5 1.5-2.0 2.0-3.0 3.0AO 4.0-5.0 >5.0
Scenario
1984
40%
40%
Calib.
TP
TP+TN
367.36
48.80
0.00
133.37
299.17
136.37
39.83
141.01
230.99
148.65
86.74
90.81
117.98
188.48
231.04
586.58
481.24
368.27
997.98
1 105.69
972.53
5107.73
5223.33
5469.47
* The DO ranges displayed are chosen to represent approximate "break-point" levels significant to the
biological community. These are arbitrary but convenient Each is supported by a brief justification:
0.0-0.5 mg/1 Inhospitable to living organisms other than sulfur bacteria
0.5-1.0 mg/1 Some benthic organisms can tolerate limited (a few days) exposure
1.0-1.5 mg/1 - Region in which hypoxic benthic phosphorus flux may decrease by 85%
1.5-2.0 mg/1 Probable acceptable exposure for demersal (sinking) fish eggs
2.0-3.0 mg/1 Stressful, especially if prolonged (> 7 days), but probably non-lethal
3.0-4.0 mg/1 Below the instantaneous water quality standard
4.0-5.0 mg/1 Meeting the instantaneous water quality standard on average
> 5.0 mg/1 Exceeding the water quality standard on average
Is the expectation of summer bottom DO above 1.0 mg/1 a plausible one? So far as we
know, the earliest survey work for the Chesapeake and lower Potomac estuaries was done
in 1912 (Sale and Skinner 1917). Their worst case condition, in the lower Potomac towards
Point Lookout (a region today also characterized by summer anoxia) was described for 21
and 22 September, 1912, with the mean for six stations at 3.40 mg/1 and a range of
1.57-4.57 (converted from ml-r1 to mg-l'1 with the ratio 1.4286 after Barnes, 1959). Sale
and Skinner "...expected the maximum reduction in the DO content of the denser bottom
layer would occur at this season of the year, since the decomposable organic matter from
plankton form, vegetation, etc., is greater during the summer months, and bacteria which
assist in the breaking down of organic matter are most active at this season, because of the
higher temperature." In this they anticipated aptly our current understanding of the process,
if not the severity experienced today.
Lowest bottom DO in the mainstem Chesapeake (55% saturation) was observed 10 miles off
the mouth of the Rappahannock on 20 September, 1912. This translates to about 4.3 mg/1
(Fair and Geyer 1963). Off Annapolis they found about 5.2 mg/1, albeit on 3 October. These
values were both substantially lower than the surface values. It is interesting the authors
here also attributed the reduction in bottom DO to: "Broken-down submerged plants, leaves,
57
-------
elutriated soil and other debris which are carried into the estuary and bay by numerous
tributaries...."
LIVING RESOURCE IMPLICATIONS
Except for the overharvesting of oysters, the living resources of this tum-of-the-century
Chesapeake were in good shape. If Sale and Skinner's data reveal the true condition of
bottom oxygen in 1912, it is heartening to consider what re-establishing these concentrations
could mean to a future Chesapeake.
In order to investigate the living resource implications of the model projections for benthic
communities in the Bay, it is necessary to interpret the water column model results in terms
of bottom surface area impacted. In this paper, we concentrate on those portions of the Bay
bottom currently exposed to varying degrees of hypoxia (< 3 mg/1 DO—summer average).
The discussion of improvements focuses on these areas.
Holland and Shaughnessy (personal communication) have interpreted data from thousands
of benthic samples collected in Chesapeake Bay and related those data to habitat conditions
observed at their sampling stations. They provided the provisional data used to construct
Table 2. These data are based on samples for 1985 but Holland indicated that 1984 was a
substantially more productive year for the benthos and recommended that the original values
be increased by a factor of 2.0 for use in a 1984 scenario.
Table 2.
Provisional Numbers Characterizing Benthic Habitat Ecology in Chesapeake Bay'
Habitat Condition
Severe Hypoxia or
Prolonged Anoxia
Benthic Organism
Biomass
CgAn2)
0.3-1.0
Annual Benthic
Production
(gAn2)
1.0-2.0
(mean 1.67 )
* Potential Capacity
for Paniculate Removal
6-20 mg/g dry wt/day
11. 25-37.52 mg/g/d
(mean 8.9 g/yr)*
Less severe exposure
<30 days (cf: today's 1.7-2.4
30-40' contour)
Consistent 1-2 mg/1 4.7-11.2
bottom DO
8.56
24.9
57.01-190.03 mg/g/d
(mean 45.1 g/yr)
165.83-552.78 mg/g/d
(mean 131.2 g/yr)
* "Potential" does not mean that the entire water column would be "cleared" of particles. Holland estimates
that shallower water columns (4-5 m) might be 60% cleared, while at 30-40 m depth only 10% of the water
column might be cleared.
' Source: Holland and Shaughnessy, personal communication, 1987.
Raising summer average bottom DO in substantial areas of the deep estuary (or in the case of
TP +TN, the entire estuary) to a level above 0.5 mg/1 and frequently above 1.5 mgyi would
permit the colonization of these areas by benthic infauna (worms, clams, etc.). This coloni-
zation presently begins each year when a larval "set" occurs following the seasonal water
column mixing which re-establishes oxygen in bottom waters. Organisms from this set are
detected by sampling in less than a month. They are extirpated, however, late each spring
when hypoxia first occurs.
Improvement in "long-lived" benthos
The areas subject to varying degrees of hypoxia shown in Figure 3 have been aggregated
into three categories to correspond roughly to the habitat groupings Holland and Shaugh-
nessy considered in their analysis of actual benthic community data. These are shown in
Table 3.
58
-------
0
2
3
4
5
6
7
8
9
10
11
14
18
17
18
19
20
CALIBRATION
1984
100 200
AREA (Sq. Km)
300
40% REDUCTION OF TP
1984
100 200
AREA (Sq. Km)
300
40% REDUCTION OF TN & TP
1984
o-
2
3-
4-
5
8
7
8
9-
10
11-
14-
16-
17-
18
19"
20-
_^^___
I I
| |
I |
100 200
AREA (Sq. Km)
300
Table 3.
Aggregation of modeled benthic hypoxia exposure into habitat categories*
Model
Scenario
1984 Calib.
40% TP
Removal
40% Removal
TN-t-TP
Severe Hypoxia or
Prolonged Anoxia
(0.0-l.Omg/l)
501.03
272.97
136.37
Less Severe Exposure
(cf: 30-40' contour)
(1. 0-2.0 mg/1)
188.48
227.75
351.80
Consistent Bottom
DO > mg/1
(2.0-4.0 mg/1)
704.56
669.72
599.31
* Areas are based on data in Table 1 and are expressed as square kilometers bottom area.
The coefficients of Holland and Shaughnessy (Table 2) are combined with the aggregated ar-
eas (Table 3) to produce Table 4, an overall benthic community interpretation for the hypoxic
regions in Chesapeake Bay and potential habitat improvements projected by the model for this
sub-region of the entire Bay.
Areas with DO remaining above 1 mg/1 could begin to sustain "long-lived benthos,"
persisting for a year or more. As organisms grow in size their filtering capacity increases
markedly. This capability assists in the aggregation of flocculents into larger effective
particle sizes ("pelletizing"), reducing resuspension through the production of mucilaginous
pseudofeces. The filtering potential of this expanded benthos under the 40% TP scenario
could remove 7.4 x 104 metric tons of paniculate material;, 24% more than benthos under the
40% TP scenario in the 1984 calibration year. For the 40% TP+TN scenario, this potential
could rise to 9.4 x 104 metric tons, 40% more than the calibration year. While Holland's
caveat in Table 2 is fully appropriate, there is good precedent to cite the importance of
benthos as a contributor to particulate removal.
Newell (personal communication) has calculated that the overharvesting of oysters in the
Maryland Chesapeake has dramatically reduced particulate clearing capacity. He suggests that
in the 1880s filtering turnover time for the Bay could have been 3.8 days; in the mid-1970s
this increased to 97.2 days and with the 1987 oyster population estimated by the Maryland
Department of Natural Resources to 486.1 days.
59
-------
Table 4
Benthic Community Interpretation Summary for Hypoxic Regions of Chesapeake Bay
in the 1984 Model Scenario and two Projected Environmental Management Strategies
Management
Strategy
1984
Calibration
Year
40% TP
Removal
Scenario
40% TP+TN
Removal
Scenario
Summer Ave.
Bottom DO
0.0- 1.0
1.0 - 2.0
2.0 - 3.0
0.0-1.0
1.0 - 2.0
2.0 - 3.0
^^^^
0.0-1.0
1.0 - 2.0
2.0 - 3.0
~
Total km2
Bottom Area
501.03
188.48
117.98
272.97
227.75
188.48
- ' " '
136.37
351.80
231.04
"
Benthic Organism
Biomass
g/m2
1.3
4.2
16.0
—
1.3
4.2
16.0
"
1.3
4.2
16.0
"
Metric Tom
651.3
791.6
1887.7
3330.6
354.9
956.6
3015.7
4327.2
(23.3%)
177.3
1477.6
3996.6
5351.5
(37.8%)
Annual Benthic
Production
g/rrrtyr
3.3
17.1
49.8
—
3.3
17.1
49.8
"" '"
3.3
17.1
49.8
"
Metric Tons
1653.4
3223.0
5875.4
10,751.8
900.8
4664.0
9386.3
14,951.1
(28.1%)
450.0
6015.8
11,505.8
17,971.6
(40.2%)
Potential Removal
Capacity for Particulars
g/m2
17.8
90.2
262.
—
17.8
90.2
262.
"
17.8
90.2
262.
"
Metric Tons
8923.1
17,000.9
30,910.8
56,834.8
4858.9
20,543.1
49,381.8
74,783.8
(24.0%)
2427.4
31,732.4
60,532.5
94,692.3
(40.0%)
Benthic Organisms
Consumed by Food
Chain in Region
M-tons/yr. k=0.33'
3583.6
4983.2
(28.1%)
5989.9
(40.2%)
* K = 0.33 represents rounding to the mid-point of the "grazing range" 30-35%, see text
Forecast biomass improvements
Relative to the calibration year, habitat improvements are projected by the model which
would permit extension of increased benthic populations into regions of the Bay covering
some hundreds of square kilometers. The augmented biomass of such a benthos for the 40%
TP scenario is estimated at 4.3 x 103 metric tons, an increase of 23.3% over the calibration.
For the 40% TP+TN scenario, biomass is projected at 5.4 x 103 metric tons, 37.8% above
the 1984 calibration year.
Production of the benthic community
Calculated production for benthic communities accounts for losses by mortality and the
grazing of predators. This estimate includes the new living tissue accumulating and the
amount passed up the food chain to higher organisms, presumably including harvestable
fish and crabs. In Table 4, we have developed production estimates. For the 40% TP case,
production could rise 28% over the deep region value for 1984. For the 40% TP+TN case,
production could rise to 1.8 x 104 tons, 40.2% greater than the calibration year.
Potential pass through to the food chain
Holland estimates that from 30-35% of benthic biomass is grazed by predator organisms
in the estuarine system, so that (assuming 33%) the 40% TP scenario could result in
5.0 x 103 metric tons moving up the food chain in Chesapeake Bay. This quantity is 28.1%
more than this region might have produced in the calibration year. For the 40% TP+TN
scenario, potential increment to the food chain is 6.0 x 103 tons, an increase over the 1984
calibration of 40.2%.
Improvement in shellfish habitat
Mature and healthy oysters, especially in low temperature conditions, can tolerate more than
a week with closed shells by respiring anaerobically (Beck and Hanson, personal
communication, 1987). Chesapeake Bay oysters, however, encounter low oxygen at high
temperatures (near 25° C) and usually just after or during their spawning period. At a time
when their metabolic rate and nutritional need is greatest and their energy reserves are least,
60
-------
exposure to hypoxic conditions is very stressful (Newell 1987). Over the past decades,
such repeated exposures have systematically eliminated the deeper oyster beds. The
absence of these beds correlates well with intrusions of severe hypoxia measured today. It
also correlates with the distribution of higher salinity and the occurrence of the major
oyster diseases MSX and "Dermo." But, it would surprise no one to find all these
prejudicial factors are synergistic. The fact remains that deeper oyster beds, once viable,
are no more.
Newell (personal communication, 1987) and others believe that deeper shellfish habitats,
once present in the pristine Chesapeake, cannot become re-established until they are
insulated from repeated (even though episodic) insults from low oxygen intrusion. This is
particularly important for juvenile oysters since swimming veliger oyster larvae appear to
die upon exposure to zero oxygen after about 18 hours. Severely hypoxic regions might
also be a "black hole" for oyster larvae spending any time there. At 20% oxygen
saturation, swimming larvae exhibit an avoidance response, swimming upwards in the
water column where they may be swept away from suitable attaching substrate and lost.
(Mann et al. 1987 and Newell, personal communication).
Using the database which supports
Table 1, we can estimate the aggregate
bottom areas impacted by DO levels
< 4 mg/1. If we consider this as the
unacceptable long-term average
condition for successful shellfish
beds, the implication is that this
marginal habitat (1394 km2 during the
1984 calibration year) decreases to
only 1170 km2 with 40% TP removal,
an improvement of 223 km2 (55,186
acres). For the 40% TP+TN scenario,
the region below 4 mg/1 summer
average decreases to 1057 km2, an
increase in the region with potential
for oyster and clam production of 336
km2 (83,099 acres). These incre-
mental acreages are depicted as a
histogram in Figure 4 in which the
"worst case" (1984) is zero and is not
shown.
Figure 4:
Increases in bottom area with potential for
shellfish production projected for two
management scenarios; 40% reduction in
total phosphorus loading (TP) and 40%
reduction in both total nitrogen and total
phosphorus (TP+TN).
CHANGE IN BOTTOM HABITAT WITH
POTENTIAL FOR SHELLFISH PRODUCTION
400
TP TWIN
SCENARIO
Reduction in benthic phosphorus release
Where DO could be raised to 1.0-1.5 mg/1, this would substantially interrupt the chemical
reactions that release phosphorus from the sediments. The release of phosphorus is never
completely interdicted. As calculated in this model, however, rates drop from about 8.0
mg-m'^d"1 in the presence of severe summer hypoxia to about 1.2 mg-m'2-d in shallower
waters not experiencing these conditions (HydroQual, Inc. 1987). Projected changes
following implementation of the two management scenarios are detailed in Table 5 and
represent substantial reductions in phosphorus release to the water column.
61
-------
Table 5
Changes projected in benthic phosphorus release from bottom areas in Chesapeake
Bay with summer average DO below 1.0-1.5 mg/1 converted to a higher oxygen
regime as a result of two pollution control scenarios
Scenario
1984 Calib.
TP 40% Rem.
TP+TN40%
km2
501.03
272.97
136.37
km2
Area
Change
-
228.06
364.66
% Change
kg/day *
-
-54.5
-72.8
Incremental
% Change
forTP+TN
-
-
-18.3
Phosphorus
Release
(metric tons)
4.008 x 103
2.458 x 103
1.529X103
P Release
% Change
-
38.7%
61.9%
* Releases in kg/d represent the differential of hypoxia mediated release at 8 mg over the "baseline"
level of 1.2 mg for mid-Bay lateral regions above the pycnocline.
Figure 5: Schematic representation for pycnocline tilting events in Chesapeake Bay; "A" scenario in
which unfavorable conditions reach high in the water column and tilting exposes large
"shelf* areas to low oxygen; "B" scenario, following an incremental improvement, shows
unfavorable conditions are lower in the water column and a tilt of the same magnitude
results in less impacted bottom area.
HABITAT IMPROVEMENTS DURING PYCNOCLINE TILTING EPISODES
A second major environmental process, one not modeled by our current tools, is the
phenomenon of pycnocline tilting. This process is shown schematically in Figure 5, version
(A) representing tilt to some fixed angle with unfavorable conditions existing higher in the
water column and version (B) with the same angle of tilt following an improvement scenario
where unfavorable conditions are only encountered at a lower depth and a substantial region
of Bay bottom is spared the transient exposure.
The pycnocline, and the horizons of varying oxygen concentration below it, are highly
dynamic zones. The term "seiche" has been loosely used to describe the transport of sub-
62
-------
pycnocline waters up onto the shallower shelf lateral to the main channel. In fact, the term
"tilt" is better applied, and such conditions usually arise when a prolonged wind set "piles"
water up against one shore and deeper water upwells into the shallows of the opposite,
windward shore. In summer, this condition frequently occurs with a southwest wind, which
sets surface water east and establishes the return circulation to upwell water with low DO
onto the western shore (towards) the wind.
Explanation for curve of bottom half-areas
The result is that approximately half the Bay's shallow bottom is affected by reduced oxygen
concentrations. In order to depict this, the bottom area data in Figure 2 has been re-plotted
symmetrically in Figure 6 A, so that half the incremental area at each successive depth is on
the eastern and half on the western flank. The abcissa measures in square kilometers, both
left and right from the center. These are curves of bottom areas totalled for all the model seg-
ments, and are not a cross section of the Bay.
WATER'S SURFACE
0
5
A 10
15
20
0
5
Z B 10
(I 15
UJ
Q 20
-------
1 METER BATHYMETRY CONTOURS FOR MD - CHESAPEAKE BAY
Figure 7: Middle (Maryland) reach of Chesapeake Bay with 1 m contours shown (Data Source:
Goldsmith and Button 1977).
middle 88 km reach of Chesapeake Bay indicates large areas can be covered by a lateral
pycnocline tilt of a few meters.
The frequency of such intrusion events has not been well documented in the past, since they
require continuous recording equipment deployed in a hostile environment. During several
recent survey periods, including deployments of a month or more (Sanford and Sellner,
1987; Breitburg, personal communication), enough events have been observed that one can
estimate in an "average" year that perhaps six events occur; four of intermediate intensity
(DO < 2 mg/1) and possibly two of severe intensity (DO < 0.2 mg/1). Both classes of events
can persist sufficiently long for environmental damage. Much more work needs to be done
with respect to such continuous data series since the frequency and duration of events deter-
mines magnitude of impact.
Currently proposed management strategies are not directed at changing these intrusion
frequencies or duration. The improvement in DO conditions to greater depths in the water
column, however, would quickly impact their severity. Achieving the TP+TN scenario is
projected to result, on the average, in virtually no extremely hypoxic water (< 0.5 mg/1)
available for intrusion onto the Bay's shelf regions.
CAVEATS TO THESE PROJECTIONS
Models simplify reality in order to deal with the complexity that confronts us when we try to
interpret the natural world. This simplification process is inherently hazardous, as workers
found when the Potomac Estuary Model was unable to reproduce observed algal concentra-
tions during a major 1983 bloom (Mutman and Masse, 1985). Subsequent research suggest-
ed that a previously unrecognized feedback mechanism between the sediments and water
column had operated to mediate the bloom (Seitzinger, 1985). Such mechanisms could easi-
ly be encountered again in Chesapeake Bay.
Some workers (Magnien, personal communication) expect that reductions in nutrient
loadings would not necessarily result in incremental improvements of DO with depth, but
rather in a gradual overall rise in average sub-pycnocline oxygen. We have no way of
directly evaluating this possibility but note that the model calculates oxygen distribution as
we have analyzed it.
64
-------
Newell, R.E.I. in G. Mackiernan, Ed. Dissolved Oxygen in the Chesapeake Bay:
Processes and Effects. Univ. MD. Sea Grant, College Park, MD. 177 p.; 1987
Sale, J.W.; Skinner, W.W. The vertical distribution of dissolved oxygen and the
precipitation by salt water in certain tidal areas. J. Franklin Inst. 148 (1104-58): 837-848p.;
1917
Sanford, L.; Sellner, K.; Bundy, M. Moored measurements of Dissolved Oxygen in the
Chesapeake Bay during the Summer of 1987. AGU Ocean Sciences Meeting. New
Orleans, LA. 1987
Seliger, H.; Boggs, J.; Biggley, W. Catastrophic Anoxia in the Chesapeake Bay in 1984.
Science. 228:70-73p.; 1985
Seitzinger, S. P. The effect of pH on the Release of Phosphorus from Potomac River
Sediments. Report 86-8F Acad. Nat. Sci. Phila. Publ. by U.S. Envir. Prot. Agency.
Chesapeake Bay Prog., Annapolis, MD. 21403; 50 p.; 1985
USEPA. Chesapeake Bay: A Profile of Environmental Change. Appendices Sec. 5 Trends
in Dissolved Oxygen. U.S. Environ. Prot. Agency. Region EL Phila., PA. 19106; 1983a
USEPA. Chesapeake Bay: A Profile of Environmental Change. U.S. Environ. Prot.
Agency. Region in. Philadelphia, PA. 19106; 200 p.; 1983b
65
------- |