College of William and Mary. Gloucester Point, Virginia 23062. ABSTRACT: Hypoxia/anoxia in bottom waters of the Rappahannock River, a tributary estuary of ...
Estuaries Vol. 14, No. 2, p. 113-121
June 1991
Spatial and Temporal Variabilities of Hypoxia in the Rappahannock River, Virginia 1 ALBERT Y. Kuo KYEONG PARK MOHAMED Z. MOUSTAFA 2
Virginia Institute of Marine Science School of Marine Science College of William and Mary Gloucester Point, Virginia 23062 ABSTRACT: H y p o x i a / a n o x i a in bottom waters of the R a p p a h a n n o c k River, a tributary estuary of Chesapeake Bay, was observed to persist t h r o u g h o u t the s u m m e r in the deep basin near the river mouth; periodic reoxygenation o f bottom water occurred o n the shallower sill at the river mouth. T h e reoxygenation events were closely related to s p r i n g tide mixing. T h e dissolved oxygen (DO) in surface waters was always near or at the saturation level, while that o f bottom waters exhibited a characteristic spatial pattern. T h e bottom DO decreased upriver from river mouth, reaching a m i n i m u m u p r i v e r o f the deepest p o i n t of the river and increasing as the water became shallower f u r t h e r upriver. A model was formulated to describe the longitudinal distribution of DO in bottom waters. T h e model is based o n a Lagrangian c o n c e p t - - f o l l o w i n g a water parcel as it travels upriver along the estuarine bottom. T h e model successfully describes the characteristic distribution of DO a n d also explains the shifting of the m i n i m u m DO location in response to spring-neap cycling. A diagnostic study with the model provided insight into relationships between the bottom DO a n d the competing factors that contribute to the DO budget o f bottom waters. T h e study reveals that b o t h oxygen demand, either benthic or water column demand, and vertical m i x i n g have a p r o n o u n c e d effect o n the severity of hypoxia in bottom waters of an estuary. However, it is the vertical m i x i n g which controls the longitudinal location o f the m i n i m u m DO. T h e strength of gravitational circulation is also shown to affect the occurrence o f hypoxia. An estuary with stronger circulation tends to have less chance for hypoxia to occur. T h e initial DO deficit of bottom water e n t e r i n g an estuary has a strong effect o n DO concentration near the river mouth, but its effect diminishes in the upriver direction.
Chesapeake Bay are partially mixed coastal plain estuaries and have a deep basin near the mouth. Hypoxia (DO less than 5 mg 1-1) has been observed frequently in the deep basin of the Patuxent River, Maryland (Laubach and Summers 1987), and the Virginia tributary estuaries (Kuo and Neilson 1987). Kuo and Neilson (1987) reported that hypoxia occurred frequently in the deep waters of the lowest reaches of the Rappahannock and York rivers and rarely occurred in the James River even though it receives the heaviest wastewater loadings among the Virginia estuaries. This difference has been attributed to the relatively strong gravitational circulation in the James River. With slackwater survey data collected since 1971, they calculated that when water temperature exceeded 20~ (typically May through September), DO concentrations below 5 mg 1-1 were observed in about 95% of the 58 surveys in the Rappahannock River and 75% of the 65 surveys in the York River, but only 7% of the 60 surveys in the James River. In this paper, the spatial and temporal variabil-
Introduction Dissolved oxygen (DO), as an index of water quality in estuarine and coastal waters, has received increased attention in recent years. Oxygen deficiency, which alters the aquatic ecosystem in an undesirable direction, is not an uncommon feature in estuarine and coastal waters such as the New York Bight (Falkowski et al. 1980), the New Jersey coast (Swanson and Sindermann 1979), and Chesapeake Bay (Officer et al. 1984). Anoxia (no dissolved oxygen) in Chesapeake Bay, which has been known since the 1930s (Newcombe and Horne 1938), has become more widespread and of longer duration during recent times (Flemer et al. 1983) and appears to have had significant ecological effects (Seliger et al. 1985). All major tributaries on the western shore of l Contribution n u m b e r 1662 of the Virginia Institute of Marine Science, College of William and Mary. 2 Present address: T e t r a "I'ech, Inc., 10306 Eaton Place, Suite 340, Fairfax, Virginia 22030. 9 1991 Estuarine Research Federation
113
0160-8347/91/020113-09501.50/0
114
A . Y . Kuo et al.
(,S .: 1:.[ii::i!ii i
,"
....... t~ ~!!!!!:?i!!~
.o ======================== ?:?i:i:?i:~:i:[:~ : ::::::::::::::::::::::::::::: i :.: :.: :.:. :.: :f~ %. iii::i::i:::i:! ~ k:::::::::::: " ............. i
2.8b . - . . . . " " "8d38
" iiiiiiiiii: i
o~149
8
:::: :: :2:
} ?.i:.:i:.i:i::::i!
~i !:: !:!:i:i. :i: : ,," ( ili ii: ill :iiiiii: :
: : . . . y " 13~i::. :i!':!i............ ii:i:;:i ::..... ..... .... ..... .....:..:. :.::..:... :.:: ,............:...
9
".......
9
~,,20 o I,,
5 , ,I,
, ii
lO j
kilometers
i:ii:!..
3.. ~ : 9 :~- ':~"i616 9 vt:: H
v 9
.. .......
.~_.s~ If
.
V,.~ "..
..
"2~:::::%:~::~:: :::13 ~
iiiiilZiiiiiiiill.................. ililii ii !~ ,~~ii ......................
9 .. (3. k.,~O./i~i~ ........~ii~iii~i::~iz~ .....
9
ff~
-,.. Fig. 1. The lower Rappahannock River and the adjacent Chesapeake Bay. The heavy shaded areas are deeper than 20 m; the depth contours are 13 m and 8 m. 9 marks the slackwater survey station and r designates the current meter as well as slackwater survey stations. ities o f D O in the lowest r e a c h o f the Rappahannock River (Fig. 1) are e x a m i n e d and a model is d e v e l o p e d to describe the longitudinal distribution o f b o t t o m DO. T h e model is then used to p e r f o r m diagnostic analyses to investigate the cause-effect relationship between model p a r a m e t e r s and the D O distribution. Observations
Kuo and Neilson (1987) showed that hypoxic conditions existed only in the lower p o r t i o n o f the R a p p a h a n n o c k River and mostly d u r i n g the summ e r season. A series o f 12 surveys was c o n d u c t e d at slackwater b e t o r e ebb in the lowest 60 km o f the river f r o m J u n e 4 to S e p t e m b e r 14, 1987. All sampling stations were located in the middle o f navigation channels. D u r i n g each survey, temperature, conductivity, and DO m e a s u r e m e n t s were taken at the designated 13 stations along the river, including one in Chesapeake Bay. Station Ioca-
tions, e x c e p t the two most u p r i v e r ones, are shown in Fig. 1. T e m p e r a t u r e and conductivity were m e a s u r e d with an Applied Micro System C o n d u c t i v i t y - T e m p e r a t u r e - D e p t h (CTD) probe. Continuous vertical profiles, top to b o t t o m , for these variables were obtained at each station. Conductivity and temp e r a t u r e m e a s u r e m e n t s were c o n v e r t e d to salinities employing the U N E S C O (1983) algorithm. DO m e a s u r e m e n t s , using a Yellow Springs I n s t r u m e n t probe, were taken every m e t e r f r o m the surface to 15 m depth, then every 2 m until the b o t t o m was reached. T h e c o m p l e t e set o f data was presented as plots ofisopleths in a vertical-longitudinal plane along the river axis (Kuo and Moustafa 1989). T h e r e is no widely accepted quantitative definition for hypoxia. T h e C o m m o n w e a l t h o f Virginia has a d o p t e d water quality standards o f a 5 mg 1-' daily average with no observation below 4 m g 1-I in estuarine waters. Many o t h e r states have c o r n -
Hypoxia, Temporaland Spatial Variabilities
parable standards. DO concentration below 5 mg 1-1 is considered hypoxic for the purpose of the following discussions. All of the DO data from the slackwater survey are daytime values, which are higher than the daily average in general. Thus, 5 mg 1-1 is a reasonable quantitative definition from the water quality management standpoint. Spatial and temporal variations of DO were observed. Figure 2 presents example DO distributions. The surface DO concentration was always high, at or near saturation level, with no persistent longitudinal pattern. The DO concentration in the bottom water, however, was low in all surveys, with a characteristic longitudinal pattern. The bottom DO concentration decreased upriver from the river mouth, reached a minimum about 42 km upriver from the mouth, and then increased to above 5 mg 1-1. Minimum DO was observed within a deep basin but not at the deepest sampling station. A similar longitudinal bottom DO distribution also was observed in the Patuxent River, Maryland (Laubach and Summers 1987), another tributary of Chesapeake Bay. Figure 2 also illustrates some variations in DO distribution among the survey dates. T h e DO distribution on June 4,July 8, and July 28 were highly stratified throughout the study area, while some degree of vertical mixing was evident from the data of June 18, July 16, and August 17, which were several days after strong spring tides. During the study period, the successive spring tides alternated in strength between strong and weak springs. Tidal mixing during strong spring tides was sufficient to completely mix the water column at the river mouth but insufficient to mix the water column in the deep basin. Thus, the vertical gradient of DO remained strong. This periodic destratification by spring tides has been documented in the Rappahannock River as well as other Virginia tributaries of Chesapeake Bay (Haas 1977; D'Elia et al, 1981; Ruzecki and Evans 1986). The temporal variation of DO may be seen more clearly with DO isopleths in a depth-time plane (Fig. 3). Additional data from the Chesapeake Bay monitoring program of the United States Environmental Protection Agency were also included in preparing Fig. 31 Figure 3a depicts the time history of the vertical DO distribution at the river mouth, in which reoxygenation of bottom water following strong spring tides was evident. DO concentrations became fairly uniform (5 to 6 mg 1-l) throughout the water column at roughly monthly intervals after the strong spring tide. In the period between strong spring tides, bottom DO dropped to as low as 2 mg 1-1, while surface DO increased to 8 mg 1-1. Only a partial reoxygenation of bottom waters by spring tide mixing was observed at stations inside the river mouth (Figs. 3b and 3c). At these deep basin sta-
115
tions, bottom hypoxia persisted from May to September 1987. More severe DO depletion was observed at the upriver shallower station (Fig. 3c) where anoxia was observed on several occasions. In addition to slackwater surveys, current velocities at two locations (Fig. 1) were measured by Inter-Ocean $4 meters; average values were recorded every 30 min. The time-series data were subjected to a low-pass filter with a (36 h) -~ cut-off frequency to remove tidal and higher frequency components. T h e filtered (subtidal) and unfiltered longitudinal velocity components near the bottom at the river mouth are presented in Fig. 4, T h e time-series data contained a semidiurnal component associated with the astronomical tide. T h e subtidal current was much smaller than the tidal current and less than 10 cm s-~ most of the time. Despite being variable in time, the subtidal current showed continuous inflow near the river bottom, a circulation pattern characteristic of partiallymixed estuaries (Pritchard 1967). T h e net bottom velocities averaged over record lengths were 6.2 cm s-1 at the mouth and 3.7 cm s-1 at 16.6 km upriver. These Eulerian residual velocities imply that the water mass at the river bottom originated mainly from Chesapeake Bay. Kuo et al. (1990) calculated the first-order Lagrangian residual velocity (or residual mass transport velocity) from current data measured in the James River and found that the Stokes drift greatly enhanced the estuarine two-layered circulation, and that the Lagrangian residual velocity was about twice the Eulerian residual near the river bottom. Therefore, it may be estimated that a water parcel will travel upriver through the deep basin of the Rappahannock River at an average velocity of 10 cm s-~.
Modeling of Bottom Dissolved Oxygen Since data from the slackwater surveys indicated only bottom water was hypoxic with characteristic longitudinal distribution, a model was developed to describe DO in the bottom water. Kuo and Neilson (1987) proposed a model to investigate the oxygen budget for the deep basin as a whole, that is hereby extended to a one-dimensional model that describes the longitudinal distribution of bottom DO along the axis of the deep basin. If we follow a parcel of water of unit volume moving upriver along the river bottom (Fig. 5), the mass balance of DO in this volume may be expressed as (change of DO in the parcel) = (replenishment by vertical diffusion from surface water) + (horizontal diffusion across the parcel surface) + (source of DO in the parcel) - (DO consumption by biochemical processes within the parcel) - (loss of DO to the river bottom because of benthic demand). Since we are following a water parcel using
116
A.Y. Kuo et al.
1
1 3 5 7
3
.~...------ 5.0 ~'JC,r
,---~\~.~
9
C3
15 17 19 21 23
o
,'o
2'0
3;
20
so
6o
11 13 151719 21 23 0
I
I
I
1
10
20
30
40
1
I
L.(
3 5
3
"~,/
i
60
50
"~"
s.o "-"N
"
3.0
9
a
,,4 _ ; r'-"(
15 171921-
I
'q
23 10
'--'---J
.
8 J U L Y 1987 20
30
\ ~)t I /
3:
40
50
\1 /
60
i
~.
0
-
13~
~I ~
I '~
30
40
60
0
5'0
60
, 50
60
r~.o
U
23
50
40
~.-/~
I ,,-
...,. 2
30
o~
1,7-
20
20
,- )
,7-I ~ "~,~I k/ "10
1'0
~-
o ~,4 ~._. "-'/rk Ik....J
0
x..7_! /
10
, 20
, 30
I 40
D I S T A N C E U P S T R E A M F R O M M O U T H (Kilometers) Fig. 2. Examples of dissolved oxygen concentrations (mg 1 ~) during June, July, and August 1987 in the lower Rappahannock
River.
a Lagrangian f r a m e o f r e f e r e n c e , the advective t r a n s p o r t does not c o n t r i b u t e to mass balance in the above expression. T h e r e p l e n i s h m e n t o f bottom D O is a direct result o f a strong vertical DO
c o n c e n t r a t i o n gradient, which is invariably m u c h large than the horizontal gradient. T h e r e f o r e , we may neglect the horizontal diffusion with respect to vertical diffusion. T h e D O in the water c o l u m n
and Spatial Variabilities
Hypoxia, Temporal
0.6 -
(a) River Mouth Station
o,,S
s
S
s
S
s
117
-
') .;
S
s
>- 0 . 0
~
l
'' ''
-0.6 ,
0
'
~
Time
(days)
10
20
30
Fig. 4. Filtered and unfiltered longitudinal velocity comp o n e n t near the b o t t o m o f the river mouth, from August 3 to S e p t e m b e r 2, 1987. Positive is in ebb direction.
(b) Station R16.6
water and surface water, respectively; h is the water depth and k, is the vertically integrated diffusion coefficient; and B combines the last two terms o f the above expression, representing benthic and water column oxygen demands. Since DO in the surface waters has been observed to be at or near saturation values in all surveys, it may be assumed that c, is a constant. Substituting the DO deficit, D = c, - c, into Eq. 1 and integrating it with respect to time, we obtained
t-- -10 '
K ~
q4
0
20
40
60
80
1O0
120
(c) Station R38.8
D(t) -- Do +
-2-
X --
-10-
-14
dt'
(2)
where D Ois the DO deficit of the water parcel when it enters the river at t = 0. If we define a Lagrangian coordinate
-6E3
B - kz
9'
,,..
,, 40,
, J , , 60l,, ~ "-~ 80*"
~o
~2o
Number of days since 14 May 1987
is mainly contributed by photosynthetic process of phytoplankton, thus we may assume that it is negligible in the bottom waters because o f turbidity. T h e n , the above expression may be translated into an equation in terms o f change of DO per unit time, dc q c dt = k,----~
B
(1)
where the derivative with respect to time, t, is the substantial derivative following a water parcel; c and c, are the DO concentrations in the parcel of
(3)
the distance the water parcel has traveled since entering the river, Eq. 2 may be transformed into D(X) = Do +
Fig. 3. Dissolved oxygen isopleths in d e p t h - t i m e plane. S and s on top o f the figure indicate times o f strong and weak spring tides, respectively. 9 at b o t t o m o f the x-axis indicates date when data are available.
u dt'
B -
h / u dX'.
(4)
In the above equation, each of the variables (B, kz, D, h, and u) in the integrand is a function of X' in the most general case. Since X' is a Lagrangian coordinate, each o f these variables follows the water parcel t h r o u g h fully spatially and temporally varying Eulerian fields. A general solution of D(X) is impossible without detailed descriptions o f the spatially and temporally varying biochemical processes as well as physical transport. However, much information may be derived with approximations and reasonable assumptions for a specific estuary. In Eq. 4, u is the Lagrangian residual velocity (i.e., the velocity of the water parcel averaged over the tidal cycle), which is about 10 cm s-' for the bottom waters in the deep basin of the Rappahannock River. Thus, it takes about 5 d for a water parcel to travel from
118
A . Y . Kuo et al.
7
--~-
h(x) - ~
I
Cs RIVER~-~BAY Diffusion k cs-c r"-~',~])~-"--~ z
h ~.,fi~"~'-I
,,
B
, I
Fig. 5. A conceptual model of the dissolved oxygen budget for a water parcel travelling upriver along an estuarine bottom. c and c, are DO concentrations of the water parcel and surface water, respectively; k, is vertical diffusion coefficient; B is oxygen demand including bottom and water column demand; u is residual mass transport velocity.
the river mouth to the upriver end of the deep basin, 45 km fi'om the mouth. This time period is short compared with the time scales of changes of other parameters. T h e time scale of the occurrence of mixing events was observed to be a month. The variation of oxygen demand depends primarily on temperature, which changes seasonally. Therefore, for all the water parcels residing at the bottom of the deep basin at a given time, each of them has experienced the same fate as it successively travels from river mouth to the location where it resides. Then, the spatial distribution of DO deficit at a given time may be expressed as D(x) = Do" +
dx' B(x') - k~(x') D(x')] h(x')J ~
(5)
where x is the Eulerian coordinate, the distance upriver from the river mouth. Equation 5 would provide the spatial distribution of bottom DO at a given time if the spatial distributions of oxygen demand (B), the vertical diffusion coefficient (k,), the residual velocity (u), and water depth (h) were known. T h e vertically integrated diffusion coefficient takes the general form (Pritchard 1960), k z = constant.Uthr
(6)
where Ut is the vertically averaged tidal velocity, r is the stability function and Ri is the Richardson number. Since a / h , where a is the tidal amplitude, is on the order of 10 -~ in the Rappahannock River, a small amplitude wave is assumed. Then, Ut may be related to the tidal amplitude as Ut = a/h(gh) v' = a(g/h) v', where g is the gravitational acceleration. Using the constant and q~(Ri) proposed by Pritchard (1960), Eq. 6 may be expressed as k~ = 2 x 10-4a(g/h) I/, h(1 + 0.276Ri) -2
(7)
where the Richardson number is approximated by
TABLE 1. Slack water survey data and oxygen demands calculated with Eq. 8.
Date
Dm (mg 1-')
x~a (km)
a (m)
h (m)
~S b (ppt)
ATb (~1)
Ri
$(Ri)
B= (grn -2 d-')
6/4 6/18 7/2 7/8 7/16 7/20 c 7/28 8/4 8/10 8/17 8/24 9/14
7.0 6.3 6.5 6.7 4.4 7.8 6.8 6.9 5.5 4.0 5.4 3.4
43 33 39 39 25 25 39 29 25 25 39 29
0.27 0.25 0.26 0.26 0.23 0.23 0.26 0.24 0.23 0.23 0.26 0.24
12 14 15 15 14 14 15 12 14 14 15 12
3.5 2.9 3.6 3.0 1.2 1.9 1.6 2.0 0.9 0.7 1.0 2.2
4.5 2.5 2.1 2.4 0.5 1.2 1.3 0.4 0.8 0.0 0.0 0.0
15.0 17.7 21.6 19.1 7.6 12.9 10.2 8.0 6.5 3.9 5.0 8.3
0.04 0.03 0.02 0.03 0.10 0.05 0.07 0.10 0.13 0.23 0.18 0.09
1.1 0.66 0.49 0.62 1.5 1.3 1.7 2.5 2.3 3.1 3.5 1.2
9x= is the distance from river mouth to the minimum DO location. b AS and AT are the vertical difference in salinity and temperature, respectively. c DO in surface waters were supersaturated.
Ri = g Ap/h p (0.7U,/h) 2 with p and AO being, respectively, water density and density difference between surface and bottom waters. There was no measurement of oxygen demand, including water column and benthic demands, in the deep basin of the Rappahannock River. It may be estimated from the balance between the oxygen sources and sinks at the location of observed minimum DO. The minimum DO, or maximum DO deficit (Din), occurs at the location where d D / d x = 0, that is, where Dm
B = B., = k z - ~ -
(8)
Using the observed value of Dm and its location extracted from the slackwater survey data, the value of Bm was calculated with equations 7 and 8 for each of the surveys. T h e calculated results and the values of parameters are listed in Table 1. The calculated oxygen demands range from 0.49 to 3.5 g 02 m -2 d -x. These values are in good agreement with observed values in coastal plain estuaries. T h e oxygen demands, including both benthic and biochemical oxygen demands, have been shown to be on the order of several g m -~ d -~ based on field data (Kuo and Neilson 1987). T h e calculated results in Table 1 also show that highest oxygen demands occur in August when the bottom waters have the highest temperature. T h e values of q~(Ri) and Bm in Table 1 are the values at the locations of minimum DO. For lack of more detailed information, these values, as well as value of u measured at a fixed location, were
Hypoxia, Temporal and Spatial Variabilities 0
01
(a)
"~
......
E
o~
v
5-
~
~
10-
o E)
04 June 18 June ---
o
0 +
~6
0
io
O
4b
~o
so
(# "1o 150 a
@(Ri) = 0.05
20 ......
0-
(b)
E
0
0
v
4-
0
[3
0
I
§ 16 July 28 J u l y - - -
4+
8
v
+
r
+ . . . §. . . . §
~o
~ ++
n
&
50
~ ++
O00~176
+++
B =
1.5
B =
3.0
mooooool ~+++ +++4++
a
- ~ - - - : ; - - ~". . . .
~o
46
3b
+
0
0
2b
1~)
o
+
70
+++4+§
r
so
++~
6
4-
a
0 E3
-6
aa a
E 9,-,
8 = 3.0
t+..++§
++++++++++§
'10
e-
u~
+++4
E v
:02 "6
0 r
(o)
I
03
.o_
119
o -14
aAa&a ~anaa
1'0 2'0 ,3b 4'0 Distance upstream from mouth (kilometers)
50
Fig. 7. Effectsof varyingoxygendemand on DO.
s -22 0
10
20
30
40
50
Distance upstream from mouth (kilometers) Fig. 6. Comparison of model predictions with observations. applied uniformly in space, and Eq. 5 was used to calculate the longitudinal distribution of DO deficit for each survey. Figures 6a and 6b compare the calculated results with field data. Figure 6c shows the simplified bottom profile used for calculation. T h e agreements between the model results and field data are satisfactory for some surveys (e.g., June 4 and July 28), but very poor for others (e.g., June 18). O f the total 12 slackwater surveys, the model failed to provide satisfactory results for half of the surveys. This may be due to the assumption of spatially uniform 4~(Ri), B, and u. Spatially varying values are definitely required to improve the predictive capability of the model. Diagnostic Studies Although the model did not result in accurate prediction without spatially varying model parameters, it did permit us to carry out some diagnostic studies. T h e model was used to investigate the effect of the parameters' values on the magnitude as well as the distribution of DO in bottom waters. Figure 7 shows the effect of varying oxygen demand (B) on bottom DO deficit. An increase in oxygen demand will increase the absolute value of the DO deficit, but not the shape of its distribution, particularly the location of maximum DO deficit, or minimum DO. Since B is assumed uniform over the deep basin, the relative DO concentration would not be altered by changing the magnitude of B.
T h e effect of the varying vertical diffusion coefficient was studied by changing the values of 4~(Ri). The vertical diffusion may vary as a result of changing stratification or tidal amplitude. Figure 8 shows that varying r changes the DO distribution as well as its absolute values. A decrease in 4~(Ri) (i.e., a decrease in vertical diffusion) results in decreased DO concentrations and the landward migration of the minimum DO location. Conditions in the Rappahannock River are such that the DO sink term is greater than the source term near the river mouth. Therefore, the DO concentration starts to decrease once the bottom water enters the river, and keeps decreasing while it travels upriver into the deep basin. As the bottom water travels beyond the deepest basin into shallower depth, vertical diffusion and thus the DO replenishment rate increase. Since water depth decreases in the upriver 0T. 0 4--" i
I
~0o.
~(R,~ = 0.3
g o'.+
~,~~
++
~
~
,+,+++
9~ x ~ ~ 0 O o _ 4.
_ooOO
o 2 I
+++ o ~176
,~ x x *ac, •
g *" 9
O
8- B = O
~ a
10
2]0
upstream
[
"'***,, 0.05 ,'*'**'***t
2.C g m -2 d o y - :
Distance
0.075
from
310
40
mouth
(kilometers)
eO
Fig. 8. Effectsof varying vertical diffusioncoefficienton DO.
120
A . Y . Kuo et al.
013
B = 2.0 g m -~ c c y -~ @(Ri) = 0 . 1 5
~176 +
~+ -f
o =c
+
oe
[]
c
3
~
-acoo_
*+,
v ,'=_
u = 0 16 m s "
"~176176
~
.2
I
O
|
J :
1:3 ~
0.04 m s
~
*§247 +§
i
m3~ ++-+ t,
0
direction, the DO source term eventually becomes equal to the oxygen demand, at which point the minimum DO occurs. This location depends on the magnitude of vertical diffusion. An increase in vertical diffusion increases the oxygen replenishment rate at a given water depth. Therefore, the balance between the source and sink terms will occur earlier in deeper water, resulting in downriver movement of the minimum DO location. This downriver movement was observed periodically around the times of stronger spring tides. T h e data in Table 1 show that minimum DO was located closer to the mouth at the middle of the months when stronger spring tides occurred. Figure 9 demonstrates the effect of varying residual mass transport velocity. Faster water movement allows less time for DO to be consumed as a water parcel travels upriver. Therefore, the bottom DO concentration increases with residual mass transport velocity. This is the reason offered by Kuo and Neilson (1987) to explain the systematic variability among three Virginia tributary estuaries of Chesapeake Bay. T h e degree to which the initial DO deficit at river mouth affects bottom DO inside the river also was investigated with the model. Figure 10 shows that the DO concentration in the water entering the river mouth has a strong effect on the DO distribution inside the river. It not only affects the severity but also the spatial extent of hypoxic conditions. T h e figure also shows that values of DO concentrations converge as the bottom water moves upriver even though the initial DO concentrations of the incoming bottom water at the mouth might be greatly different. This indicates that irrespective of the DO concentration of the incoming bay water, the bottom water in the Rappahannock River would end up being hypoxic. Discussion T h e model presented in this paper is based on a simplistic budgeting of dissolved oxygen in a par-
Fig. 10. on DO.
~
ooO!
~
~++++++ +- + ++++
~. . . . . . . . . . . . . . . . . . . . . . . xxxx•215215
Distance u p s t r e a m from mouth (kilometers) Effects o f varying residual transport velocity on DO.
0 iI
~ ~
6t . . . . .
i
=
OiOB rn s -~
oc
~
4
2.0 g m -2 day -I
co *+
so
o! i
Fig. 9.
m
cc
-o
tm
} ( R I)
2- ,+ uc
+ ++
.2
c
I._ nc
.
8 =
o
]~
++
++an & • •
:t~*
~xx• xx•215215
~'0 2'0 Distance u p s t r e a m f r o m
3'0 mouth
aS (kilometers)
50
Effects o f varying initial DO deficit at river m o u t h
cel of water as it travels landward along the estuary bottom. T h e DO budget consists of only one source and one sink term. Although the predictive application presented here is not always satisfactory, it is not because of the model itself but because of the lack of complete information for input parameters. If there were data for the spatial distributions of the oxygen demand and the vertical diffusion coefficient, the model should be able to predict the bottom DO distribution accurately. T h e variations of input parameters with time, be it due to seasonal temperature change or fortnightly spring-neap cycling, may be easily transferred to the temporal variation of bottom DO. T h e diagnostic study enabled us to investigate cause-effect relationships. The field observation that the minimum DO concentration occurs at the shallow location upriver of the deepest basin, can be reasonably explained by the model. DO concentration in a parcel of bottom water decreases with time, or equivalently the distance from river mouth, as long as the sink term is greater than source term. It will not reach a minimum until the water arrives at a location where the source term due to vertical mixing becomes equal to the sink term due to oxygen demand. T h a t location is not necessarily the location of maximum water depth but somewhere upriver on shallower sloping bottom. T h e model further explains the observed periodic downriver movement of the minimum DO location due to the increase of vertical diffusion, and thus the source term, around the time of stronger spring tides. T h e model study on the effect of residual mass transport velocity also confirms the argument of Kuo and Neilson (1987) that variability in the strength of the gravitational circulation is responsible for some of the difference in frequency, duration, and severity of hypoxic conditions among three Virginia tributary estuaries of Chesapeake Bay. It has been suggested that anoxia in the main-
Hypoxia, Temporal and Spatial Variabilities
stem of Chesapeake Bay might be a primary contributing factor to the hypoxic condition in the Rappahannock River. T h e diagnostic study, however, shows that the effect of initial DO concentration in the incoming bay water at the river mouth is dampened out as the bottom water travels upriver in the Rappahannock River. In addition, bottom DO concentrations above 5 mg 1-1 were often observed at the station in the river mouth, while hypoxia persisted in the deep basin of the lower reach of the river throughout the summer. Therefore, the hypoxic condition in the deep basin in the Rappahannock River seems to be more locally driven than merely an extension of that in Chesapeake Bay. It could be that the Rappahannock River would end up having hypoxic problems regardless of the quality of the bottom water coming in from Chesapeake Bay. ACKNOWLEDGMENTS The field surveys from which this paper was derived were supported by the Virginia State Water Control Board and the Virginia Institute of Marine Science under the Cooperative State Agencies Program. The development of the model was funded by the Commonwealth of Virginia as part of Chesapeake Bay Initiatives. The authors wish to express their appreciation to S. Snyder, K. Dydak, and T. Shannon for their participation in field surveys, to N. Wilson for her assistance in data analyses, and to F. Hoffman of the Virginia State Water Control Board for providing data from the EPA Chesapeake Bay monitoring program. We also thank B. Neilson for his constructive criticism in the preparation of this manuscript. LITERATURE CITED
D'ELIA, C. F., K. L. Wr~B, AND R. L. WETZEL. 1981. Time varying hydrodynamics and water quality in an estuary, p. 597-606. In B.J. Neilson and L. E. Cronin (eds.), Estuaries and Nutrients. The Humana Press, New Jersey. FALKOWSKI, P. G., T. S. HOPKINS, ANDJ. J. WALSH. 1980. An analysis of factors affecting oxygen depletion in the New York Bight. Journal of Marine Research 38:479-506. FLEMER, D. A., G. B. MACKIERMAN, W. NEHLSEN, AND V. K.
TIPPIE. 1983. Chesapeake Bay: A profile of environmental change. U.S. Environmental Protection Agency, Washington, D.C. 199 p. HAAS, L. W. 1977. The effect of the spring-neap tidal cycle
121
on the vertical salinity structure of the James, York and Rappahannock rivers, Virginia, U.S.A. Estuarine and Coastal Marine Science 5:485-496. Kvo, A. Y.,J. M. HAMRmK, AND G. M. SISSON. 1990. Persistence of residual currents in the James River estuary and its implication to mass transport, p. 389-401. In R. Cheng (ed.), Residual Currents and Long T e r m Transport, Coastal and Estuarine Studies, Vol. 38. Springer-Verlag, Berlin. Kuo, A. Y. AND M. Z. MOUSTAFA. 1989. Hypoxia in the lower Rappahannock estuary. Virginia Institute of Marine Science Special Rep. 302. V1MS, Gloucester Point, Virginia. 75 p. Koo, A. Y. ANn B.J. NEIt.SON. 1987. Hypoxia and salinity in Virginia estuaries. Estuaries 10:277-283. LAUBACH, E. B. AND R. M. SUMMERS. 1987. Patuxent estuary water quality survey 1986 data summary. Rep. No. 67. Modeling and Analysis Division, Water Management Administration, Maryland Dept. of the Environment. 70 p. NEWCOMnE, C. L. ANn W. A. HORNE. 1983. Oxygen poor waters of the Chesapeake Bay. Science 88:80-81. OFFICER, C. B., R. B. BIGGS,J. L. TAFT, L. E. CRONIN, M. A. TYLER, AND W. R. BOYTON. 1984. Chesapeake Bay anoxia: Origin, development and significance. Science 223:22-27. PRITCHARD, D. W. 1960. The movement and mixing of contaminants in tidal estuaries, p. 23-36. In Proceedings of the 1st International Conference on Waste Disposal in the Marine Environment, at the Univ. of California, Berkeley, 1959. Pergamon Press, New York. PRITCHARD, D. W. 1967. Observations of circulation in coastal plain estuaries, p. 37-44. In G. H. Lauff(ed.), Estuaries. Publication No. 83. American Association for the Advancement of Science, Washington, D.C. RUZECKI, E. P. ANn D. A. EVANS. 1986. Temporal and spatial sequencing of destratification in a coastal plain estuary, p. 368-389. In J. Bowman, M. Yentsch, and W. T. Peterson (eds.), Tidal Mixing and Plankton Dynamics, Lecture Notes on Coastal and Estuarine Studies, Vol. 17. Springer-Verlag, New York. SELIGER,H. H.,J. A. BOGGS, ANDJ. A. BIGGLEY. 1985. Catastrophic anoxia in the Chesapeake Bay in 1984. Science 228: 70-73. SWANSON, R. L. AND C. J. SINDERMANN. 1979. Oxygen depletion and associated benthic mortalities in New York Bight, 1976. NOAA Professional Paper 11, United States Department of Commerce. 345 p. United Nations Educational, Scientific and Cultural Organization. 1983. Algorithms for Computation of Fundamental Properties of Seawater. UNESCO Technical Papers in Marine Science, 44. Paris. 53 p. Received for consideration, July 16, 1990 Accepted for publication, December 14, 1990
