Estuaries Vol. 19, No. 2B, p.

386-407 June 1996

Nutrient Changes in the Mississippi

River and System Responses on the
Adjacent Continental Shelf
NANCY N. [Link] 1 WILLIAM J. WISE~,~, .JR.
Louisiana Universities Marine Consortium Coastal Studies Institute and
8124 Highway 56 Department of Oceanography and Coastal Sciences
Chauvin, Louisiana 70344 Louisiana ,State University
Baton Rouge, Louisiana 70803
R. EUGENE TURNER
DUBRAVKOJUSTI(; BARUN K. SEN GUPTA
Coastal Ecology Institute and Department of Geology
Department of Oceanography and Coastal Sciences Louisiana State University
Louisiana State University Baton Rouge, Louisiana 70803
Baton Rouge, Louisiana 70803

QUAY DORTCH
Louisiana Universities Marine Consortium
8124 Highway 56
Chauvin, Louisiana 70344
ABSTRACT: The Mississippi River system ranks among the world's top 10 rivers in freshwater and sediment inputs to
the coastal ocean. The river contributes 90% of the freshwater loading to the Gulf of Mexico, and terminates amidst
one o f the United States' most productive fisheries regions and the location of the largest zone of hypoxia in the western
Atlantic Ocean. Significant increases in riverine nutrient concentrations and loadings of nitrate and phosphorus and
decreases in silicate have occurred this century, and have accelerated since 1950. Consequently, major alterations have
occurred in the probable nutrient limitation and overall stoichiometric nutrient balance in the adjacent continental shelf
system. Changes in the nutrient balances and reduction in riverine silica loading to the continental shelf appear to have
led to phytoplankton species shifts offshore and to an increase in primary production. The phytoplankton community
response, as indicated by long-term changes in biological uptake o f silicate and accumulation of biologicafiy bound silica
in sediments, has shown how the system has responded to changes in riverine nutrient loadings. Indeed, the accumulation
of biologically bound silica in sediments beneath the Mississippi River plume increased during the past two decades,
presumably in response to increased nitrogen loading. The duration, size, and severity of hypoxia has probably increased
as a consequence of the increased primary production. Management alternatives directed at water pollution issues within
the Mississippi River watershed may have unintended and contrasting impacts on the coastal waters of the northern Gulf
of Mexico.

Introduction (justi~ et al. 1987), a n d s o m e areas o f the Baltic

T h e r e is a general consensus that the eutrophi- Sea (e.g., A n d e r s s o n a n d Rydberg 1988). T h e Mis-
cation of estuaries a n d enclosed coastal seas has sissippi a n d Atchafalaya rivers d e b o u c h o n t o a con-
increased over the last several decades. Evidence tinental shelf that is relatively o p e n c o m p a r e d to
f r o m the Baltic Sea, Kattegat, Skagerrak, a n d the m o r e enclosed water bodies n o t e d above, but
Dutch W a d d e n Sea suggests a long-term increase the biological effects o f increased riverine n u t r i e n t
in the f r e q u e n c y o f p h y t o p l a n k t o n blooms, includ- loads a p p e a r c o m p a r a b l e to those seen of enclosed
ing noxious f o r m s (Smayda 1990). Also, an in- COaStS.
crease in the areal extent a n d / o r severity o f hyp- C h a n g e s in riverine e n d - m e m b e r n u t r i e n t con-
oxia was observed, for e x a m p l e , in C h e s a p e a k e Bay centrations a n d ratios may affect coastal ecosystems
(Officer et aL 1984), the n o r t h e r n Adriatic Sea in several ways. O n e potential effect is to influence
p h y t o p l a n k t o n p r o d u c t i o n rates. Significant in-
1Corresponding author; tele 504/851-2800; fax 504/851- creases in n u t r i e n t c o n c e n t r a t i o n s a n d loadings o f
2874. nitrate a n d p h o s p h o r u s a n d decreases in silicate

9 1996 Estuarine Research Federation 386

 Mississippi River Changes and Consequences 387

have o c c u r r e d in the Mississippi River this century, dicators o f eutrophication and increased oxygen
and have accelerated since 1950 ( T u r n e r and Ra- stress.
balais 1991, 1994a). As a result, alterations have
o c c u r r e d in the relative composition of nutrients The Mississippi River System
and in probable n u t r i e n t limitation in adjacent
T h e Mississippi River watershed, the largest in
continental shelf waters (Justi~ et al. 1993, 1994,
1995a, b). If, indeed, silica is a n u t r i e n t limiting the United States, encompasses 41% of the area of
p h y t o p l a n k t o n growth, a decrease in diatom abun- the c o n t e r m i n o u s 48 states (Fig. 1). T h e Mississip-
dance on the continental shelf could follow declin- pi River ranks a m o n g the world's top 10 rivers in
ing riverine silicate supplies. An increase in nitrate- discharge (580 km 3 yr -1) and sediment yields (210
nitrogen concentration, however, may allow for • 106 t yr -~) to the coastal ocean (Milliman and
m o r e efficient diatom growth at lower silicate con- Meade 1983). Major alterations in the m o r p h o l o g y
centrations. O n the o t h e r hand, while the frequen- o f the main river channel and widespread land-
cy of probable silica limitation has c h a n g e d overall scape alterations in the watershed, along with an-
annually (Justis et al. 1994) or at certain times of t h r o p o g e n i c additions of nitrogen and phospho-
the year (Dortch and Whitledge 1992), the pro- rus, have r e s u l t e d in d r a m a t i c w a t e r quality
duction o f diatoms may have r e m a i n e d u n c h a n g e d changes this c e n t u r y ( T u r n e r and Rabalais 1991).
or increased d u r i n g spring, the critical period for T h e Mississippi and Atchafalaya rivers are the
carbon p r o d u c t i o n and flux related to the forma- primary riverine sources of fresh water to the Lou-
tion of bottom water hypoxia (Qureshi 1995). Even isiana continental shelf (Dinnel and Wiseman
if d i a t o m p r o d u c t i o n r e m a i n s u n c h a n g e d , in- 1986) and to the Gulf o f Mexico (90% o f fresh-
creased nitrogen loading is likely to stimulate non- water inflow to the Gulf; National Oceanic and At-
diatom algal production. A second possible effect m o s p h e r i c Administration 1987). One-third of the
involves a shift in p h y t o p l a n k t o n c o m m u n i t y com- flow of the Mississippi River system enters the Gulf
position. Increased nitrogen and p h o s p h o r u s con- via the Atchafalaya River. O f the remaining dis-
centrations will favor the growth o f flagellates (Of- charge from the Mississippi River delta proper, ap-
ficer and Ryther 1980). In the case o f nitrogen- proximately 53% flows westward o n t o the Louisi-
limited coastal p h y t o p l a n k t o n populations, nui- ana shelf (United States Army Corps o f Engineers
sance blooms might develop as the silica : nitrogen 1974). T h e long-term peak flow occurs in April.
atomic ratio a p p r o a c h e s 1:1 (the Redfield ratio; Although flow is r e d u c e d in summer, large-scale
Redfield 1958), as is in the case o f the lower Mis- circulation patterns often retain the fresh water on
sissippi River system ( T u r n e r and Rabalais 1991, the shelf. Freshwater inflow from the Mississippi
1994a). T h e combination o f primary p r o d u c t i o n and Atchafalaya is a major feature of the Louisiana
changes and species shifts could affect subsequent shelf, even t h o u g h it is a fairly o p e n system, and
c a r b o n utilization, carbon flux, and the areal ex- fresh water attributed to these rivers may be traced
tent or severity o f hypoxia on the continental shelf. as far west as Port Aransas on the south Texas coast
T h e n o r t h e r n Gulf o f Mexico adjacent to the out- (Smith 1980) and as far east as the Atlantic sea-
flows of the Mississippi and Atchafalaya rivers is the board, u n d e r combinations o f major floods and ap-
location o f the largest zone o f hypoxia in the west- propriate o c e a n o g r a p h i c conditions (Atkinson and
ern Atlantic O c e a n (as great as 9,500 km 2 in sum- Wallace 1975; Tester and Atkinson 1994; Walker et
mer; Rabalais et al. 1991, 1994a, b). al. 1994).
T h e effects of changes in riverine nitrogen, Within the n o r t h e r n Gulf of Mexico, the influ-
phosphorus, and silica fluxes to the continental ence of the Mississippi and Atchafalaya outflows is
shelf at the terminus of the Mississippi River is the evident in plots o f surface salinity and Secchi disk
focus of this synthesis. We provide b a c k g r o u n d in- values accumulated from n u m e r o u s hydrographic
f o r m a t i o n on the Mississippi River discharge, the cruises between 85~ (just ea~st o f Cape San Bias,
river's influence on the physical and chemical Florida) to 95~ (just west of Galveston Bay, Texas)
characteristics o f the adjacent coastal waters, a n d (Fig. 2). T h e influence o f the freshwater input is
i m p o r t a n t biological processes o f the study area. m o r e obvious to the west o f the Mississippi delta
We d o c u m e n t changes in Mississippi River n u t r i e n t (approximately 89~ than to the east, with addi-
concentrations, loadings and ratios, and similar tional inputs n e a r 91~ (the Atchafalaya delta).
changes in n u t r i e n t properties o f adjacent conti- Plots of n u t r i e n t values (example given for silicate
nental shelf waters. We then address biological re- in Fig. 2) show a m o r e p r o n o u n c e d decrease in
sponses to these n u t r i e n t alterations, including c o n c e n t r a t i o n over an equal distance from the del-
shifts in probable n u t r i e n t limitations for phyto- tas when c o m p a r e d to the salinity plot. This indi-
plankton growth, shifts in p h y t o p l a n k t o n c o m m u - cates a nonconservative mixing due to biological
nity structure, p h y t o p l a n k t o n production, and in- uptake.
388 N.N. Rabalais et al.

Miss. R.
100' Mississippi 9 Vicksburg
9 River 9 9 Red R.
,-10 = l 00 9
9 ~v espo~~ St. Francisville
::L
9 "din
=. 1 000 9
O m
O. .1
Luling (195)
c-
o New Orleans (168)
.01
10 ...... 100 ..... 1000 Morgan City
(0.5) 9 Venice (18)
N O 3 - N (I~M)
Fig. 1. Drainage of the Mississippi River (from Turner and Rabalais 1991); comparison of annual average N concentration and P
concentration of major world rivers with the Mississippi (modified from Turner et al. 1990; Meybeck1982); and schematic of sampling
stations along the lower Mississippi River, with distance from the terminus (km) in parentheses (from Turner and Rabalais 1991).

Adjacent Continental Shelf Hydrography occurs and is best developed during the summer as
Freshwater discharge from the Mississippi and At- a result of both the initial spring flood discharge,
chafalaya rivers rapidly forms the Louisiana Coastal vernal warming, and return flow along the shelf
Current, a highly stratified coastal c u r r e n t that break from the Texas shelf (Cochrane and Kelly
flows, on average, westward along the Louisiana 1986). This secondary, seasonal pycnocline resem-
coast and then southward along the Texas coast. At bles that observed near the Islay front (Hill and
the beginning of the flood season, coastal winds are Simpson 1989). The summer hypoxia (see below)
from the southeast. As spring progresses, the winds f o u n d in this region is capped by this seasonal pyc-
along the south Texas coast b e c o m e favorable for nocline.
upwelling and the local flow reverses and turns to- Strong wind events interact with the waters o f
ward the n o r t h and east (Cochrane and Kelly 1986). the coastal c u r r e n t t h r o u g h o u t the year. Intense
This reversal has been d o c u m e n t e d as far east as wind mixing due to cold air outbreaks and frontal
Cameron, Louisiana (93~ and inferred as far passages is active from as early as late S e p t e m b e r
east as fhe Isles Dernieres (90~ (Kimsey a n d to as late as June. Local squalls a n d thunderstorms,
Temple 1963, 1964). as well as tropical storms a n d hurricanes, are im-
Density stratification in the Louisiana Coastal Cur- p o r t a n t d u r i n g the s u m m e r m o n t h s (DiMego et al.
rent, particularly near the two major freshwater 1976). These wind events often result in the com-
sources, is due primarily to salinity. While the strong- plete h o m o g e n i z a t i o n o f the water c o l u m n (Wise-
est haline stratification is associated with the Louisi- man et al. 1986). At o t h e r times, winds are n o t
ana Coastal Current, a secondary pycnocline often strong e n o u g h to fully break down the local strat-
 Mississippi River Changes and Consequences 389

West Delta I
East West Delta East
12 9 m n I I I I ~ i m i ii mm
160
9 m m m n i i m i

I N U l i n

t-
9 ii I ( m
~'120
.~
t"
8
I
9 mm

I I I 9
mmm

I m
m m J m m m m

m ~ n I
im

I IN II .,t II 9
0 I I I I I I ~ I m I I I I I II

IImI n I m ~ i i m i Illn iI

4 9 II II mE i m i i I Immmi ii

I N . i N
;._ 4(I
9 II

9 I I ~
I
I N
0 0
40 100

~30
E
e.
lO .s
9 d,= =9 lcr'F",'J :.:
'I ,.

=e , 2O 9 l
=m
m

t~ e-
u .1 -l I
o

0 .01
98 94 90 86 82 98 94 90 86 82
Longitude Longitude
Fig. 2. Plots of surface salinity, Secchi disk depth, and silicate concentration from a series of hydrographic cruises in 10-100 m
water depth for 1972-1991, for the months and longitudes indicated (R. E. Turner and N. N. Rabalais unpublished data).

ification, but they result in localized coastal up- described below, to deplete the oxygen c o n t e n t o f
welling, which alters the stratification (Dagg 1988). the near-bottom waters.
T h e d e v e l o p m e n t and persistence o f b o t t o m wa-
ter hypoxia is intimately linked to the temporal Biological P r o c e s s e s
and spatial variability o f the hydrographic proper- High biological productivity in the immediate
ties. T h e strong persistent s u m m e r halocline pre- (320 g C m -2 yr -~, L o h r e n z et al. 1990) and ex-
vents effective reoxygenation o f d e e p waters f r o m t e n d e d plume (290 g C m -2 yr -1) o f the Mississippi
the u p p e r layers. T h e fall destruction o f the halo- River (Sklar and T u r n e r 1981) is mediated by high
cline by wind mixing and t h e r m a l cooling o f sur- n u t r i e n t inputs and regeneration, and favorable
face waters heralds the termination of hypoxic con- light conditions. Small-scale, short-term variability
ditions. T h e m o r p h o l o g y o f the hypoxic region is in productivity is the c o n s e q u e n c e o f various fac-
controlled by the structure o f the secondary near- tors, such as n u t r i e n t concentrations, temperature,
b o t t o m pycnocline (Wiseman et al. u n p u b l i s h e d and salinity ( L o h r e n z et al. 1990, 1994), but on a
data). T h e s u m m e r halocline associated with the seasonal time-scale it is most influenced by Missis-
Louisiana Coastal C u r r e n t isolates near-bottom wa- sippi River flow and n u t r i e n t flux to the system
ters from direct wind forcing. This effect, in con- (Justid et al. 1993). " N e w " nutrients b e c o m e de-
j u n c t i o n with the pressure gradients driving up- pleted along the river-to-ocean mixing gradient
coast flow along the Texas i n n e r shelf, results in t h r o u g h dilution and biological uptake, and regen-
slow-moving b o t t o m waters over the Louisiana in- erated nutrients support primary p r o d u c t i o n for
n e r shelf. T h i s allows b i o l o g i c a l p r o c e s s e s , great distances from the river m o u t h (Dortch et al.
 3@0 N.N. Rabalais e l al.

LOUISIANA

::!25- 50%

TERREBONNE B A Y ~

75-100 %
' ~ ...._.. ~ : . . ! :" i:..,...-....
~50-75 %! :::"": ...-?.':'..~-..:9::
~..:.....~:.:.!~:'!~"

':,":::i::.~. :'~.:::!:.::i!:.~.":."..:~i:.::::" 9

i,:!.!!~:.!!!~i.......
il

GULF OF MEXICO

~i 0 20km
I 9~ ~r . . . . . . . . . .) i 'oo' 8 9~3o' 9 j
Fig. 3. Station locations within the Mississippi River bight for hypoxia monitoring on transects A'-D' (closed circles), mooring
locations (C6A and C6B), LASERstations (closed triangles, those referred to in text are labeled), and "Platform" and "Control"
stations of Fucik (1974) and Ward et al. (1979) (open circles marked with "C" and "P"). Stippled intensity corresponds to frequency
of occurrence of mid-summer hypoxia at monitoring stations (1985-1987, 1990-1993, N. N. Rabalais, R. E. Turner, and w.J. Wiseman,
Jr. unpublished data).

1992a). Particulate organic c a r b o n flux to the low- ability) when m o r e extensive datasets a n d longer
er water c o l u m n is high in the e x t e n d e d p l u m e time periods are e x a m i n e d (e.g., marine-origin
over the i n n e r shelf (approximately 500-600 m g C p h y t o p l a n k t o n p i g m e n t s a n d biogenic silica in sur-
m -z d -] in 15 m water depth; Qureshi 1995; see ficial sediments a n d Pb-210 d a t e d s e d i m e n t cores;
also Redalje et al. 1994). T h e fraction of" p r o d u c - Rabalais et al. 1992a; T u r n e r a n d Rabalais 1994b).
tion e x p o r t e d f r o m the surface waters is highly T h e s e couplings implicate changes in riverine nu-
variable a n d ranges f r o m 10% to 200%, with the trients a n d n u t r i e n t ratios (discussed below) with
h i g h e r p e r c e n t a g e s in the spring. A large p r o p o r - the overall effects on productivity, c a r b o n accu-
tion of the particulate organic c a r b o n flux reaches m u l a t i o n at the seabed, a n d low oxygen conditions
the b o t t o m i n c o r p o r a t e d in z o o p l a n k t o n fecal pel- on this shelf.
lets as well as individual cells or in cell aggregates. O x y g e n - d e p l e t e d b o t t o m waters are seasonally
T h e r e is great daily a n d weekly variability in cur- d o m i n a n t features o f the Louisiana continental
rent flow a n d stratification on the shelf and, there- shelf adjacent to the deltas of the Mississippi a n d
fore, n o simple description o f the couplings be- Atchafalaya rivers (Rabalais et al. 1991, 1992b,
tween c a r b o n p r o d u c t i o n in surface waters a n d de- 1994a, b). T h e areal e x t e n t o f bottom-water hyp-
livery a n d recycling in b o t t o m waters at this time oxia (<-2 m g 1-~ dissolved oxygen) in m i d - s u m m e r
scale. However, there is evidence o f an ecological may cover as m u c h as 9,500 km z, with the spatial
"signal" (couplings) amidst the " n o i s e " (the vari- configuration varying interannually (Fig. 3). Sam-
 Mississippi River Changes and Consequences 391

ples from a transect on the southeastern shelf and pogenic n u t r i e n t loads on a coastal marine ecosys-
c o n t i n u o u s time series data off T e r r e b o n n e Bay tem.
d o c u m e n t hypoxic b o t t o m waters as early as Feb-
ruary and as late as October, with widespread, per- Methods
sistent, and severe hypoxia or anoxia from mid- We present data mostly from published results,
May to mid-September. Spatial and temporal vari- which provide the appropriate details of collection
ability in the distribution o f hypoxia exists and is, and analysis methods. In addition, we present two
at least partially, related to the amplitude and phas- new syntheses: historical changes in phytoplankton
ing of the Mississippi River discharge (Rabalais et species composition and in benthic foraminiferal
al. 1994a; Wiseman et al. unpublished data) and, communities. More detailed m e t h o d s are provided
consequently, to n u t r i e n t flux to the coastal waters for the latter two syntheses.
and subsequent p r o d u c t i o n and flux of carbon While the influence of the Mississippi and At-
from surface waters to the lower water c o l u m n and chafalaya rivers can be seen as far as the south Tex-
seabed. as coast (Smith 1980), the n o r t h e a s t e r n Gulf, the
Seasonal variations in net productivity in the south Florida region, and the United States East
n o r t h e r n Gulf o f Mexico (as exemplified by sta- Coast (Walker et al. 1994 and references therein),
tions C6A and C6B in Fig. 3) are c o h e r e n t with the we limit our discussion of consequences of riverine
dynamics o f freshwater discharge (Justi~ et al. n u t r i e n t changes to the areas influenced by the
1993). T h e surface layer (0 m to 0.5 m) shows an immediate and e x t e n d e d plumes of tile c u r r e n t
oxygen surplus relative to the saturation values birdfoot delta in the Mississippi River bight west to
during February-July; the m a x i m u m occurs during a b o u t 90~ (the e n t r a n c e to T e r r e b o n n e Bay)
April and May and coincides with the m a x i m u m (Fig. 3). T h e o c e a n o g r a p h i c data include the fol-
flow of the Mississippi River. T h e b o t t o m layer (ap- lowing: m o n i t o r i n g cruises c o n d u c t e d during the
proximately 20 m), on the contrary, exhibits an period 1985-1993, across the width of the Louisi-
oxygen deficit t h r o u g h o u t the year; it reaches its ana shelf, primarily in mid-summer (transects A'
greatest deficit value in July. Bottom hypoxia in the through D' in Fig. 3); data from the southeastern
n o r t h e r n Gulf is most p r o n o u n c e d during periods shelf off T e r r e b o n n e Bay on a biweekly to monthly
o f high water-column stability when surface-to-bot- basis in 1985-1986 and 1990-1993 (transect C in
tom density differences are greatest (Rabalais et al. Fig. 3); m o r e f r e q u e n t sampling at an i n s t r u m e n t
1991, Wiseman et al. unpublished data). T h e cor- m o o r i n g (stations C6A and C6B in Fig. 3); six
relation between Mississippi River flow and surface cruises covering 10 m to 80 m water d e p t h within
oxygen surplus is greatest with a time-lag of 1 mo, the Mississippi River bight between 1987 and 1990
and the highest correlation for bottom oxygen def- (i.e., LASER stations in Fig. 3); and miscellaneous
icit is with a time-lag o f 2 m o (Justi~ et al. 1993). o t h e r cruises within the study area.
These findings suggest the oxygen surplus in the Standard water c o l u m n profile data were ob-
surface layer following high flow d e p e n d s on nu- tained from a Hydrolab Surveyor II, Hydrolab Sur-
trients ultimately c o m i n g from the river but regen- veyor 3, or a SeaBird CTD system. All dissolved ox-
erated many times. Annual mass balance calcula- ygen probes were calibrated and quality controlled
tions ( T u r n e r and Rabalais 1991; Dortch et al. with Winkler titrations; conductivity was calibrated
1992a) and N uptake m e a s u r e m e n t s in the fall sug- by discrete m e a s u r e m e n t s analyzed on an AutoSal
gest every N atom is recycled approximately four salinometer. Nutrients were measured on a Tech-
times, although recycling may be less i m p o r t a n t in nicon Auto Analyzer according to m e t h o d s de-
the spring (Dortch et al. 1992a). This is i m p o r t a n t scribed in Parsons et al. (1984).
because a surplus o f oxygen relative to the satura-
Changes in Nutrients
tion value is a g o o d indicator o f net productivity
in the surface waters. An oxygen surplus also LOWER MISSISSIPPI RiVER
means there is an excess o f organic matter derived Water quality data for the lower Mississippi River,
from primary p r o d u c t i o n that can be redistributed p r e v i o u s l y e l a b o r a t e d by T u r n e r a n d Rabalais
within the system; some of this will eventually reach (1991) and updated, were collected from stations
the sediments. T h e d e v e l o p m e n t o f s u m m e r hyp- at St. Francisville, Luling, New Orleans, and Venice
oxia in the n o r t h e r n Gulf o f Mexico (as exempli- (Fig. 1). T h e m e a n annual c o n c e n t r a t i o n . o f nitrate
fied by stations C6A and C6B, Fig. 3) is associated was approximately the same in 1905-1906 and
with the decay o f organic matter accumulated dur- 1933-1934 as in the 1950s, but it has d o u b l e d in
ing spring p h y t o p l a n k t o n blooms (Qureshi 1995). the last 35 yr (Fig. 4). T h e m e a n annual concen-
These findings d e m o n s t r a t e a close coupling be- tration o f silicate was approximately the same in
tween riverborne nutrients, net productivity, and 1905-1906 as in the early 1950s, then it declined
hypoxia, as well as implicate the effects of anthro- by 50%. Concentrations o f nitrate and silicate ap-
392 N . N . Rabalais et al.

200 9 St. F r a n c i s v i l l e 9 1975-85

o New Orleans o 1955-5g
160

o.M E=, ~- _ Z I9
Z 120 120
I
80 Z
J 80
40
~5
~ 40
0
0
200' 250
~ 200
150
13
o I00

2, so
O
I0
O o.
er 4 o .2 10:
L~

O U
2 O

z
O~
0 z I
I g00 1920 1940 1960 I gs0 2000
N
Year
2 4 6 8 10 12 A l l
Fig. 4. Averageannual concentrations of nitrate and silicate Month Months
and Si:N ratio from 1905 through 1994 in lower Mississippi Riv-
er for stations indicated (see Fig. 1) (modified and updated Fig. 5. Monthly average concentrations of nitrate and sili-
from Turner and Rabalais 1991, 1994a). cate, and Si:N ratio, in the lower Mississippi River for periods
indicated (modified and updated from Turner and Rabalais
1991, 1994a).
p e a r to have stabilized, b u t trends are m a s k e d by
increased variability in the data f r o m the 1980s a n d
early 1990s. We f o u n d no substantial records o f to- T h e seasonal p a t t e r n s in nitrate a n d silicate con-
tal p h o s p h o r u s c o n c e n t r a t i o n s in the lower Missis- c e n t r a t i o n have also c h a n g e d d u r i n g this c e n t u r y
sippi River b e f o r e 1972. A l t h o u g h the concentra- (Fig. 5). T h e r e was n o p r o n o u n c e d p e a k in nitrate
tion o f total p h o s p h o r u s a p p e a r s to have increased c o n c e n t r a t i o n earlier this century, whereas there
since 1972, variations a m o n g years are large. were spring peaks f r o m 1975 to 1985, p r e s u m a b l y
T h e rise in nitrate since 1960 was coincidental related to seasonal agricultural activities, t i m e d
with an increase in application of nitrogen fertil- with long-term p e a k river flow. A seasonal s u m m e r -
izer in the watershed sufficient to a c c o u n t for the fall m a x i m u m in silicate c o n c e n t r a t i o n , in contrast,
changes in water quality ( T u r n e r a n d Rabalais is n o longer evident (Fig. 5). C o n s e q u e n t l y the sea-
1991). T h e decrease in silicate since 1960 a p p e a r s sonal signal o f silica : nitrogen atomic ratio has also
to be a c o n s e q u e n c e o f u p s t r e a m p h o s p h o r u s ad- changed. T h e seasonal shifts in n u t r i e n t concen-
ditions that stimulated freshwater d i a t o m p r o d u c - trations a n d ratios b e c o m e increasingly relevant in
tion a n d an eventual burial in freshwater sedi- light of the close t e m p o r a l c o u p l i n g of river flow
m e n t s o f silica in d i a t o m r e m a i n s (e.g., Schelske to surface water net productivity (1-mo lag) a n d
a n d S t o e r m e r 1971; Schelske et al. 1986), thus re- s u b s e q u e n t b o t t o m water oxygen deficiency (2-mo
ducing the a n n u a l supply o f riverine silicate to lag) described above.
coastal waters. T h e p r o p o r t i o n s o f dissolved Si, N, a n d P in the
T h e silicate:nitrate ratios have c h a n g e d as the lower Mississippi River have c h a n g e d historically
c o n c e n t r a t i o n s varied (Figs. 4 a n d 5). T h e silicate : such that they now closely a p p r o x i m a t e the Red-
nitrate atomic ratio was a p p r o x i m a t e l y 4:1 at the field ratio (Si:N:P = 16:16:1; Fig. 6) (Justi~ et al.
b e g i n n i n g o f this century, d r o p p e d to 3:1 in 1950, 1995 a, b). We c o m p a r e d the data for two periods:
a n d t h e n rose to a p p r o x i m a t e l y 4.5:1 d u r i n g the 1960-1962 a n d 1981-1987 (Table 1; data f r o m Tur-
n e x t 10 yr, before p l u m m e t i n g to 1:1 in the 1980s. n e r a n d Rabalais 1991). Substantial increases in N
T h e ratio a p p e a r s stable (little variation) at 1:1 (300%) a n d P (200%) c o n c e n t r a t i o n s o c c u r r e d
t h r o u g h 1994. over several decades, a n d Si d e c r e a s e d (50%). (No
 Mississippi River Changes and Consequences 393

100 100

13.. 13._
~10
oo
84 ~1o

0 1 0
[Link] I 0.01 o ~ 1 lo_
Si/N , vu ~, . Si/N - I uu ~"

60 Redfield
1000[
4O

2O
\ __.1~176
/
o

Si/N 5
6o
"• ,
100

d
10 10
Si/N lOO
Fig. 6. Scatter diagrams of atomic ratios of dissolved inorganic nitrogen (N), reactive phosphorus (P), and reactive silica (Si) in
surface waters of the northern Gulf of Mexico for (A) reconstructed data for 1960 and (B) ambient nuu'ient ratios during 1985-
1991. Vertical bar indicates the Redfield ratio (Si:N:P = 16:16:1). (C) Shift in mean atomic ratio of nitrate-nitrogen (N), total
phosphorus (P), and reactive silica (Si) in the Mississippi River (Mi) from 1960-1962 to 1981-1987. (D) Clustering of N, P, and Si in
the A m ~ o n (Am), Changjiang (Ch), Huanghe (Hu), Mackenzie (Ma), Mississippi (Mi), Po (Po), Rhine (Rh), Seine (Se), Yukon
(Yu), and Zaire (Za) rivers. Data for the Mississippi River is in Table 1. (Modified fi'om Justi6 et al. 1995a).

data on total P c o n c e n t r a t i o n in the Mississippi Riv- twofold between 1960-1962 a n d 1981-1987.) Ac-
er were r e p o r t e d p r i o r to 1973; however, total P in cordingly, the Si:N ratio d e c r e a s e d f r o m 4.2 Io 0.9,
the river showed a m o d e r a t e increase between the Si:P ratio d e c r e a s e d f r o m 39.8 to 14, a n d the
1973 a n d 1987. By applying a linear least-squares N:P ratio increased f r o m 9 to 15. By applying the
regression on the 1973-1987 data, we estimated (p Redfield ratio as a criterion for stoichiometric nu-
< 0.01) that the total P c o n c e n t r a t i o n increased trient balance, o n e can distinguish between P-de-
394 N . N . Rabalais et al.

TABLE 1. Historical changes in concentrations (txM) and coastal n u t r i e n t composition by assuming that the
atomic ratios of nitrogen (N), p h o s p h o r u s (P) and silica (Si) in
relative p r o p o r t i o n o f nutrients in the river-domi-
the lower Mississippi River and the n o r t h e r n Gulf of Mexico;
= mean value, n = n u m b e r of data, S = standard error, p < n a t e d coastal waters reflects the changing compo-
0.001 = highly significant difference in nutrient concentrations sition of riverine nutrients (Table 1). This assump-
between the two periods, based on a two-sample t-test (modified tion was m a d e in view o f the fact that the Missis-
from Justi6 et al. 1995b). sippi River is the most i m p o r t a n t n u t r i e n t source
Mississippi River N o r t h e r n G u l l ot Mexico to the n o r t h e r n Gulf o f Mexico. T h e detailed re-
1960-1962 1981-1987 1960 a 1985-1991 construction p r o c e d u r e is given in Justi~ et al.
Nutrient concentration ([Link]):
(1995b). By calculating the specific rates o f change
N~ 2 36.5 114 2.23 8.13
for Si:N, Si:P, and N:P ratios in the Mississippi Riv-
n 72 200 219 219 er, we obtained a reasonable estimation o f coastal
S 2.9 6.0 0.16 0.60 n u t r i e n t composition 30 yr ago. Comparison o f the
(p < 0.001) reconstructed data with the available historical nu-
pb ~ 3.9 7.7 0.14 0.34 trient data (Thomas and Simmons 1960; T u r n e r
n -- 234 231 231 and Rabalais 1994a) showed a reasonable agree-
S -- 0.4 0.01 0.02
(p < 0.(X)I)
m e n t between the m e a s u r e d and the reconstructed
n u t r i e n t ratios. A similar reconstruction t e c h n i q u e
Sic ~ 155.1 108 8.97 5.34
n 72 71 235 235
for the n o r t h e r n Adriatic Sea p r o d u c e d results that
S 7.5 4.3 0.55 0.33 closely paralleled the real data (Justi~ et al. 1995a,
(p < 0.001) b).
Average atomic ratios: Comparison of m e a s u r e d and r e c o n s t r u c t e d nu-
Si:N 4.2 0.9 4.0 0.7 trient ratios for the n o r t h e r n Gulf adjacent to the
N:P 9 15 16 24 Mississippi River outflow reveals long-term changes
Si:P 39.8 14 64 16 in p r o p o r t i o n s of nutrients in the surface waters
N-NO s for the Mississippi River, dissolved inorganic nitro- (Justi(~ et al. 1995a, b) (Fig. 6). T h e r e c o n s t r u c t e d
gen (DIN = NOs- + NH4 § + NO~ ) for the n o r t h e r n Gulf of n u t r i e n t ratios for 1960, o n average, scatter f u r t h e r
Mexico.
from the Redfield ratio (Si:N:P = 16:16:1; the cen-
b Total P for the Mississippi River, reactive P for the n o r t h e r n
Gulf of Mexico. ter of the grid) than the r e c e n t data. By applying
9 Reactive Si. the Redfield ratio as a criterion for balanced nu-
Reconstructed data. trient composition, it appears that P and N defi-
ciency have decreased while Si deficiency has in-
ficient, N-deficient, and Si-deficient rivers, and creased. Equally important, r e c e n t n u t r i e n t ratios
those having a well-balanced n u t r i e n t composition. scatter very close to the Redfield ratio, suggesting
T h e n u t r i e n t ratios for the Mississippi River (1981- an almost perfectly balanced n u t r i e n t composition.
1987 database) show an almost perfect coinci- Probable n u t r i e n t limitation (Dortch and Whit-
d e n c e with the R e d f i e l d ratio (Fig. 6). T h e pro- ledge 1992) was also assessed by c o m p a r i n g the
portions of Si, N, and P have changed over time ambient n u t r i e n t concentrations with the ks for nu-
in such a way that they now suggest a balanced trient uptake and, in the case o f Si, a threshold
n u t r i e n t composition. value for uptake. Plots of relative frequencies (Jus-
ti~ et al. 1994) (Fig. 7) show that dissolved N con-
ADJACENT CONTINENTAl. SHELF centrations in the surface layer o f the n o r t h e r n
We analyzed extensive n u t r i e n t datasets from the Gulf o f Mexico during the p e r i o d 1985--1992 were
n o r t h e r n Gulf o f Mexico to examine how the lower than 1 txM in a b o u t 13% of the cases. Re-
coastal n u t r i e n t structure may reflect long-term active P was below 0.1 IxM in 17% of the cases, and
changes in the p r o p o r t i o n s of dissolved Si, N, and reactive Si concentrations lower than 2 txM oc-
P in riverine loads Qlusti6 et al. 1994, 1995a, b). c u r r e d in 25% of the cases. In contrast, the cor-
Reliable long-term datasets of the n o r t h e r n Gulf's r e s p o n d i n g frequencies were 39%, 41%, and 10%,
n u t r i e n t composition 30 yr ago were not available, respectively, in 1960. These findings are i m p o r t a n t
however. We reconstructed, therefore, the past because studies of n u t r i e n t uptake kinetics (i.e.,

_..)

Fig. 7. Relative frequency polygons showing temporal changes in surface nutrient structure of the n o r t h e r n Gulf of Mexico,
assuming that the relative rates of change in n u t r i e n t conccntrations over the intervening period were the same as in the Mississippi
River (delta N = +8.0%/yr; delta P = +4.3%/yr; delta Si = - 1 . 4 % / y r ) ; TIN = dissolved inorganic N, RP = reactive P, Si = reactive
Si, N = n u m b e r of data, X = mean value, M = median. Vertical lines indicate threshold values for nutrient uptake based on literature
data from studies of phytoplankton uptake kinetics. (From Justi(. et al. 1994).
 Mississippi River Changes and Consequences 395

1960 1985-1991
A

~o 20 20
ITIN J
>,
O
c 15 15

O"
I,.
~> 10 10
N..-

Q)
>,
" ~ 5

n-
O I 9 I " I " I " 0
0 2 4 6 8 0 4 6 8 10
TIN (}~M) TIN (~M)

40
o~ 40

30 X=0,34 I
X=0.14 I
" 30 M=0.12 I M=0.301
0"
20
& 20 17% < 0.1 I~M

9- - ~o
10

liE: 0 9 01~
0.0 0.5 1 .0 1 .5 2.0 0.0 0.5 1 .0 1 .5 2.0
RP (}~M) RP (~M)

2O 20
o~
>,
0
r
15 "
I
s'2351
X=9.08J
M=6.63 I
15

10
o"
m 10 i
5

~ 0 - ~ , 9 9 o 9 o , 9 . . 9
0
0 5 10 15 20 0 5 10 15 20
Si (~M) Si (~M)
396 N.N. Rabalais et al.

R h e e 1973; H a r r i s o n et al. 1977; G o l d m a n a n d Gli- the earlier studies. In addition, because of uncer-
bert 1983; Nelson a n d Brzezinski 1990) indicate tainties o f taxonomy, differences were only n o t e d
that c o n c e n t r a t i o n s o f 1 ~M, 0.1 p~M, a n d 2 p.M where organisms were identified unambiguously.
may be c o n s i d e r e d as threshold values for N, P, a n d D e m o n s t r a b l e changes have o c c u r r e d in the di-
Si uptake, respectively. Thus, it a p p e a r s that overall a t o m a n d n o n d i a t o m species c o m p o s i t i o n f r o m the
n u t r i e n t limitation has decreased. 1950s a n d 1970s to present. S o m e heavily silicified
Changes in riverine n u t r i e n t loads a n d n u t r i e n t diatom species are either n o t o b s e r v e d at all in
ratios may "affect coastal p h y t o p l a n k t o n c o m m u n i - r e c e n t samples (boldface type in the Appendices)
ties in several ways. T h e Si:N:P ratio o f m a r i n e di- or are m u c h less d o m i n a n t . For e x a m p l e , n o Me-
atoms, which are a b u n d a n t constituents o f coastal losira species, which a p p e a r heavily silicified, were
phytoplankton communities and comprise most of o b s e r v e d in 1990-1993 but were p r e s e n t in b o t h
the biomass, is a p p r o x i m a t e l y 16:16:1 w h e n nutri- 1 9 5 5 - 1 9 5 7 a n d 1972-1973. Asterionella japonica
ent levels are sufficient. Deviations f r o m this ratio (=Asterionellopsis glacialis, R o u n d et al. 1990) was
in nutrients available in the water c o l u m n may be o b s e r v e d recently at low salinities in the spring b u t
a limiting factor for diatoms, as well as for o t h e r is n o t a d o m i n a n t species as it was f r o m 1955-1957.
p h y t o p l a n k t o n g r o u p s (Hecky a n d Kilham 1988; Similarly, m o r e lightly silicified diatoms are doc-
Dortch a n d Whitledge 1992). Also, a decreasing Si: u m e n t e d for the 1970s a n d present, especially at
N ratio may e x a c e r b a t e e u t r o p h i c a t i o n by r e d u c i n g higher salinities. Rhizosolenia fragilissima a n d Cera-
the potential for d i a t o m growth in favor of noxious tulina pelagica, which are so lightly silicified they
flagellates (Officer a n d Ryther 1980). A n o t h e r rea- are s o m e t i m e s difficult to see, were n o t r e p o r t e d
sonable hypothesis that tbllows a m o r e favorable, fbr 1955-1957. Two species o f I#ptocylind~icus spp.
b a l a n c e d n u t r i e n t composition, as evidenced in were frequently d o m i n a n t in 1990-1993 but were
the Mississippi River a n d in the coastal waters as a m i n o r constituent in 1955-1957. During 1972-
well, is that surface p r i m a r y productivity has in- 1973, lightly silicified diatoms were r e p o r t e d , in-
creased u n d e r these conditions. cluding Rhizosolenia delicatula, Leptocylindricus dani-
cus, a n d Ceratulina pelagica, but a m o r e quantitative
Consequences of Shifts in analysis would be r e q u i r e d to d e t e r m i n e if their
Nutrient Composition a b u n d a n c e was less than at present.
T h e data suggest that the shift in d o m i n a n t di-
PHYTOPIANKTON SPECIES COMPOSITION
a t o m composition, toward m o r e lightly silicified
T h e changes in riverine a n d coastal n u t r i e n t species, o c c u r r e d between 1955-1957 a n d 1972-
c o n c e n t r a t i o n s a n d ratios over time suggest that 1973, but m e t h o d o l o g i c a l differences p r e c l u d e
there should be observable changes in phytoplank- conclusions a b o u t changes in n o n d i a t o m s . Since Si
ton species composition. T h e increasing N avail- availability c o n t i n u e d to decrease after the early
ability a n d decreasing Si:N ratios should lead to 1970s ( T u r n e r a n d Rabalais 1991), a c o n t i n u e d
increases in d o m i n a n c e of lightly-silicified diatoms shift in species c o m p o s i t i o n to n o n d i a t o m s would
a n d n o n d i a t o m s . Published reports o f phytoplank- be expected. T h e p h y t o p l a n k t o n at stations C6A
ton species c o m p o s i t i o n for 1955-1957 n e a r the a n d C6B in 1990-1993 often were d o m i n a t e d nu-
delta ( S i m m o n s a n d T h o m a s 1962) a n d for 1972- merically by small flagellates a n d cyanobacteria
1973 a p p r o x i m a t e l y 80 km west of the delta (Fucik (not shown in A p p e n d i x 3). T h e y were n o t consid-
1974; Ward et al. 1979) were c o m p a r e d with data e r e d in this c o m p a r i s o n , because it is n o t clear
for 1990-1993 f r o m n e a r the delta a n d f r o m sta- w h e t h e r they would have b e e n o b s e r v e d in the
tions C6A a n d C6B in 20 m water d e p t h off Terre- 1972-1973 study, even if p r e s e n t in large n u m b e r s .
b o n n e Bay (Dortch et al. u n p u b l i s h e d data) (Ap- It is also t e m p t i n g to hypothesize that the p r e s e n c e
p e n d i c e s 1-3, Fig. 3). This c o m p a r i s o n is qualita- o f Trichodesmium sp. in 1990-1993, but not in the
tive because o f differences in locations, seasons earlier studies, indicates d e c r e a s e d Si availability.
sampled, and methodology. The methodology Trichodesmium sp., however, is a b l o o m - f o r m i n g spe-
used in earlier r e p o r t s may have missed c o m m o n cies that has b e e n r e p o r t e d previously for this re-
n o n d i a t o m s , such as small coccoid cyanobacteria gion (Eleuterius et al. 1981).
a n d phytoflagellates, which now often d o m i n a t e . Several s p e c i e s with i m p o r t a n c e to h u m a n
Consequently, it was not possible to d e t e r m i n e health are now p r e s e n t b u t were either absent be-
w h e t h e r the d o m i n a n c e of these g r o u p s has in- fore or have increased in d o m i n a n c e . T h e domi-
creased. A conservative a p p r o a c h was taken in this n a n c e of Nitzschia pungens on the Louisiana shelf
c o m p a r a t i v e analysis. D a t a w e r e c o m p i l e d by a p p e a r s to have increased dramatically since the
m a t c h i n g season a n d location as closely as possible 1950s, a n d c o n c e n t r a t i o n s now frequently exceed
a n d including in the r e c e n t data only the phyto- 1 • 106 cells 1-l. S o m e f o r m s of this species have
p l a n k t o n types that would have b e e n o b s e r v e d in been associated with amnesiac shellfish p o i s o n i n g
 Mississippi River Changes and Consequences 397

(reviewed in Shumway 1990; Hallegraeff 1993). .~ 30

9 B e f o r e 1979
Samples from a r e c e n t bloom at station C6B have
Q o A f t e r 1979 ~ ,
b e e n identified (G. Fryxell, personal communica-
tion) as a sometimes toxic f o r m o f Nitzschia pun- 20. All Months
gens, now called Pseudonitzschia pseudodelicatissima ~: .
(Martin et al. 1990). Since this species is heavily
silicified for a marine diatom (Conley et al. 1989),
it is difficult to attribute its increase to decreasing
Si:N ratios, although it could be r e s p o n d i n g to in- z o
creasing N availability. Dinophysis caudata, a dino- ~. 30"
flagellate which may be associated with diarrhetic 9 B e f o r e 1979
e~
shellfish poisoning (Dickey et al. 1992), was n o t
r e c o r d e d in the earlier studies. Recently, it was of-
ten present at low to m o d e r a t e concentrations;
P3
~A
~aE
2O Months
6-C~
o A f t e r 1979

sometimes it was d o m i n a n t and r e a c h e d concen- -~ ::k

trations as high as 1 • 105 cells 1-1. .g lo
T h e indicated shift from heavily to lightly silici- 0
fled forms may have altered carbon flux via directly
z 00
sinking p h y t o p l a n k t o n cells, if silica has b e c o m e . . . . . . . . . ! oo . . . . . . . . . 2bo
limiting. Subsequently, organic loading to the sea- River Concentration (pM Sil
bed and, possibly, oxygen depletion may have been Fig. 8. The estimated n e t uptake o f silicate above dilution
affected. O n the o t h e r hand, an increase in non- losses at 30%0, for all data (top), and for s u m m e r data (months
diatom forms that are less preferentially grazed 6--8). (From T u r n e r and Rabalais 1994a.)
than diatoms may have increased the flux o f these
cells into the lower water c o l u m n and altered ox- s u m m e r m o n t h s d u r i n g peak hypoxia develop-
ygen c o n s u m p t i o n in a different manner. m e n t (June-August) as well. T h e results from this
analysis suggest that net silicate uptake in the di-
SILICATE-BASED PHYrOPLANKTON
lution gradient from river to sea has r e m a i n e d the
COMMUNITY RESPONSE same, or even increased, as the c o n c e n t r a t i o n de-
Bien et al. (1958) first d o c u m e n t e d the dilution creased. Two causal factors explaining this result
and nonconservative uptake o f silicate in the Mis- are related to the effects o f o t h e r nutrients and
sissippi River plume by sampling from the river changes in the light regime this century.
m o u t h seaward in 1953 and 1955. A notable char- T h e 20-30%0 range o f the mixing zone is where
acteristic o f the mixing diagram is that the concen- light conditions improve rapidly in the seaward di-
tration o f silicate often falls below the conservative rection and where the highest primary p r o d u c t i o n
mixing line, thus indicating biological uptake. Up- rates are usually observed (Sklar and T u r n e r 1981;
take can be statistically m o d e l e d as a deviation L o h r e n z et al. 1990). T h e light conditions in this
f r o m this mixing line, which we did for 31 ade- region may have c h a n g e d because o f dams trap-
q u a t e l y s a m p l e datasets ( T u r n e r a n d Rabalais ping sediments, especially in the Mississippi River
1994a). We f o u n d that the c o n c e n t r a t i o n o f silicate watershed. Phytoplankton p r o d u c t i o n should have
at the 20%0 mixing point declined in the last sev- increased s o o n e r in the mixing gradient and with
eral decades d u r i n g the winter-spring (January- greater strength if turbidity decreased d u r i n g this
April) and s u m m e r m o n t h s (June-August); how- period o f watershed manipulation. Also, Justi~ et
ever, there was no discernible change during the al. (1995a, b) analyzed the n u t r i e n t ratios within
fall-winter m o n t h s ( O c t o b e r - D e c e m b e r ) . We nor- the Mississippi River and offshore waters. T h e y ob-
malized for the effects o f varying concentrations in served that the ratios o f N, P, and Si, although dif-
the riverine e n d - m e m b e r (e.g., L o d e r and Rei- f e r e n t in the last several decades, were now closer
chard 1981) and c o m p a r e d the estimated net sili- to the Redfield ratio, an indicator o f improved nu-
cate uptake at 30%0 as a function o f silicate river- trient conditions for p h y t o p l a n k t o n growth. T h e y
ine e n d - m e m b e r c o n c e n t r a t i o n (Fig. 8). Noncon- suggested that coastal p h y t o p l a n k t o n p r o d u c t i o n
servative uptake of silicate was indicated in all data would be stimulated by the favorable balance o f N,
collections. T h e net uptake (at 30%0) above dilu- P, and Si, r a t h e r than an increase in any one par-
tion r a n g e d f r o m 1% to 19% o f the intercept con- ticular nutrient.
centration, and the data groups before and after BIOLOGICALLY B O U N D [Link] AND
1979 were remarkably similar. T h e net silicate up- CARBON A C C U M U l a T I O N
take appears h i g h e r after 1979 than before 1979 As n o t e d earlier, nearly coincidental trends in
(Fig. 8); these differences were a p p a r e n t in the silicate (decreasing) and nitrogen (increasing)
398 N.N. Rabalais et al.

concentrations in the Mississippi River caused the general, accumulation o f BSi in sediments mimics
average annual dissolved silicate:nitrate-N atomic overlying water c o l u m n productivity, and that, the
ratio (Si:N) to decline from about 4:1 earlier this m o r e diatoms that are p r o d u c e d by nutrient-en-
century to 1:1 this decade. T h e present Si:N ratio h a n c e d growth, the m o r e BSi will be deposited.
(1:1, Fig. 4) is t h o u g h t to be a significant limiting Additional information is in T u r n e r and Rabalais
threshold for diatom growth, intraspecific compe- (1994b).
tition, and p r o d u c t i o n (Officer and Ryther 1980; If the assumption is m a d e that the BSi:C ratio at
Smayda 1990; Dortch and Whitledge 1992; T u r n e r the time of deposition r e m a i n e d constant this cen-
and Rabalais 1994a). Thus, two contrasting hypoth- tury, then the increased BSi deposition represents
eses predict changes in phytoplankton on this a signifcant change in carbon deposition rates (up
coast since the 1950s. T h e first is that the coastal to 43% higher in cores dated after 1980 than those
phytoplankton are nitrogen, not silica, limited, and dated between 1900 to 1960). T h e pattern in %
higher nitrogen loading will result in p r o p o r t i o n - BSi changes parallels the d o c u m e n t e d increases in
ally h i g h e r phytoplankton p r o d u c t i o n rates. A nitrogen loading in the lower Mississippi River,
c o m p e t i n g hypothesis is that the combination o f over the same period d u r i n g which the silicate con-
lower silica tluxes and a Si:N ratio n e a r 1:1 will centrations have been decreasing (Fig. 10). We
result in lower p h y t o p l a n k t o n p r o d u c t i o n rates conclude from our analyses that the flux of dia-
t h r o u g h limits on diatom p r o d u c t i o n along with toms from surface to b o t t o m waters, beneath the
species composition shills. This is a nontrivial issue Mississippi River plume, increased this century.
for managers, because diatoms are an ecologically These changes were coincidental with changes in
i m p o r t a n t constituent of phytoplankton and con- riverine nitrogen loadings and resulted in h i g h e r
tribute significantly to the organic loading o f bot- organic sedimentation to b o t t o m water layers. T h e
tom waters and sediments and the subsequent ox- depletion o f bottom water oxygen, its persistence
ygen depletion. and areal coverage on this shelf, is thus indicated
We d o c u m e n t e d that surficial sediments, directly to have b e e n altered this century.
downstream and b e n e a t h the surface riverine-es- Conley et al. (1993) predicted that coastal ma-
tuarine dilution plume, reilected the in situ pri- rine systems would not see long-term depletion of
mary p r o d u c t i o n and subsequent transport of or- dissolved silicate with eutrophication as in en-
ganic carbon from surface to b o t t o m waters within closed lakes, because regeneration within the sys-
the Mississippi River bight (Rabalais et al. 1992a; tem would maintain sufficient dissolved silicate lev-
T u r n e r and Rabalais 1994b) (Fig. 9). We f u r t h e r els to prevent silica limitation. O u r results which
quantified the silica in the skeletal remains of di- show c o n t i n u e d a c c u m u l a t i o n o f b i o l o g i c a l l y
atoms sequestered as biologically b o u n d silica b o u n d Si in sediments b e n e a t h the plume and sim-
(BSi) in dated sediment cores from the same re- ilar or h i g h e r net silicate uptake by phytoplankton
gion. T h e highest concentrations of BSi were in in the plume surface waters, in spite of lower con-
sediments deposited in 25-50 m water d e p t h in the ccntrations of Si delivered by the Mississippi River,
middle o f the sampling area. T h e % BSi in sedi- s u p p o r t this hypothesis.
ments from d e e p e r waters (110 m and 200 m) were
generally stable t h r o u g h time but rose in the shal- CONSEQUENCES TO HYPOXIC BO'IVFOMWATER
lower stations (10 m and 20 m) a r o u n d the begin- FORMATION AND SEVERITY
ning o f this century. At the intermediate depths Long-term changes in the severity and extent of
(27 m to 50 m), where both the % BSi concentra- hypoxia c a n n o t be assessed directly, because sys-
tion and accumulation rates were highest, coinci- tematic sampling o f dissolved oxygen concentra-
dental changes in the % BSi with time were evi- tions in b o t t o m water did not begin until 1985.
dent, especially in the 1955 to 1965 period (a rise Prior records o f hypoxia, dating to 1973, were ob-
and tiall) and a post 1975 (1980?) rise that was sus- tained sporadically as c o m p o n e n t s of o t h e r studies;
tained to the sampling date (1989) ( T u r n e r and previous events were drawn from anecdotal rela-
Rabalais 1994b) (Fig. 10). T h e general pattern that tionships with shrimp trawl yields. T h e r e f o r e , bio-
emerges is an equilibrium accumulation of BSi logical, mineral, or chemical indicators of eutro-
from 1800 to 1900, then a slow rise, tollowed by a phication a n d / o r hypoxia preserved in sediments,
m o r e dramatic rise in the past two decades. Dia- w h e r e a c c u m u l a t i o n rates r e c o r d h i s t o r i c a l
genesis o f the BSi u n d o u b t e d l y occurs in these changes, provide clues to prior hydrographic and
cores, but will be a relatively low because the sed- biological conditions.
imentation rate is high (>0.5 cm yr-l). Further- We used d o m i n a n c e trends of benthic forami-
more, others have f o u n d the r e c o r d o f BSi to be a nifera to d e t e r m i n e their use as indicators o f re-
good indicator o f in situ production. Conley et al. d u c e d oxygen levels a n d / o r carbon-enriched sed-
(1993) summarized for freshwater lakes that, in iments (Sen Gupta et al. 1981; Sen Gupta and Ma-
 MississippiRiverChangesandConsequences 399

25 ~ ,'~-~ 9
9

9 -
9 9 9 ~75

/25"
.7/ 5 o ' ~
- 28"30'N 90.00, w 89o50, w
I i I t

SEDIMENT TOTAl PIGMENTS (~g/g dry weight) SEDIMENT BIOGENIC SILICA (weight %)

Vr F
MISSISSIPPI RIVER
M = MARINE DELTA
T = TERRESTRIAL M / \ 0 0 ~ o ~ ' ~ j O/o
I 1 i E i

$,3C SEDIMENT TOTAL ORGANIC CARBON

Fig. 9. Tile spatial distribution in April 1989 of phytoplankton pigments (b~gg-1 dry weight sediment), biogenic silica (% BSi by
weight), de113Csignature, and total organic carbon (% by dry weight of sediments), in sediments beneath the Mississippi Riverplume
(from Turner and Rabalais 1994b).

chain-C&stillo 1993). T h e same series o f Pb-210 preted as foraminiferai responses to increasing ox-
dated sediment cores used for BSi analyses (Tur- ygen stress (Sen Gupta et al. 1996). Benthic fora-
n e r and Rabalais 1994b; Fig. 3) were used for de- miniferal density and diversity are generally low in
termination o f benthic foraminifera. Some down- this environment, but a c o m p a r i s o n o f assemblages
core shifts in species a b u n d a n c e s at station G27 in surficial sediments from areas differentially af-
(Fig. 3) in the Mississippi River bight were inter- fected by oxygen depletion indicates that the dom-
400 N.N. Rabalais et al.

9 % BSi g. Core G27

Ammonla/(Amrnonia + Elphidlurn) X 100
._ I 0 z
(D

'!1
~e
a0
m
E50
o
g25 Ig50 I g75 2000
Fig. 10. Average % BSi c o n c e n t r a t i o n o f s e d i m e n t s in each
section o f Pb-210 dated s e d i m e n t cores f r o m station E50. A 3- <
70'
yr r u n n i n g average for each s a m p l i n g date is shown. T h e % BSi
data are s u p e r i m p o s e d with a 3-yr r u n n i n g average o f the nitro- <
g e n loading f r o m the Mississippi River t h r o u g h the delta passes
(from T u r n e r a n d Rabalais 1994b).

50 , ,
inance of Ammonia parkinsoniana over Elphidium 1700 1800 1900 2 O0
spp. is m u c h m o r e p r o n o u n c e d u n d e r h y p o x i a
than in well-oxygenated waters. T h e a b u n d a n c e o f
A. parkinsoniana is also c o r r e l a t e d with p h y t o p l a n k - Core G27
ton biomass in surface waters a n d % BSi (food 5o Fursenkolna j
source indicator) in sediments. In the c o n t e x t o f
m o d e r n hypoxia, species distribution in d a t e d sed- 01
i m e n t cores reveals stratigraphic trends in the Am- t~
monia:Elphidium ratio that indicate an overall in- G) 30-
crease in oxygen stress (in intensity or d u r a t i o n ) 0z-
in the last 100 yr (Fig. 11). In particular, the stress O. 20-
seems especially severe since the 1950s. For this
time period, b o t h Ammonia a n d Elphidium b e c o m e 10
less i m p o r t a n t c o m p o n e n t s of the assemblage,
while Fursenkoina (known to be hypoxia tolerant Ol 9 , 9 ~ ,

elsewhere) shows an unusual d o m i n a n c e . Quin- 1700 1800 1900 2000

queloculina (a significant c o m p o n e n t o f the m o d -
e r n assemblage only in well-oxygenated waters) has 20
b e e n absent f r o m the r e c o r d of this core since the Core G27
1870s but was a c o n s p i c u o u s e l e m e n t o f the f a u n a Quinqueloculina
in the previous 100 yr. Thus, there are indications
that oxygen deficiency stress increased as n u t r i e n t ~)

loads a n d c a r b o n flux to the seabed increased. tO

*~ lO
Summary of Historical Trends 0
L..

Mississippi River n u t r i e n t c o n c e n t r a t i o n s a n d
loadings to the adjacent c o n t i n e n t a l shelf have
c h a n g e d dramatically this century, with an accel-
eration of these changes since the 1950s. T h e con-
centrations o f dissolved N a n d P d o u b l e d a n d Si
d e c r e a s e d by half, the dissolved Si:N ratio d r o p p e d 1700 1800 1900 2000
f r o m 4:1 to 1:1, a n d seasonal trends have c h a n g e d . Year
T h e resulting n u t r i e n t c o m p o s i t i o n in the receiv-
Fig. 11. C h a n g e s in b e n t h i c foraminifera species with strati-
ing G u l f waters shifted toward stoichiometric nu- graphic d e p t h in Pb-210 dated s e d i m e n t core f r o m station G27
trient ratios closer to the Redfield ratio a n d m o r e in the Mississippi River bight.
b a l a n c e d than previously. N a n d P are indicated to
be less limiting now for p h y t o p l a n k t o n growth,
while s o m e increase in Si limitation is probable. In p l a n k t o n c o m m u n i t y in the mixing zone c o m p a r e d
spite of a p r o b a b l e decrease in Si availability, the to the 1950s, a n d g r e a t e r a c c u m u l a t i o n rates o f BSi
overall productivity o f the ecosystem a p p e a r s to in sediments b e n e a t h the p l u m e , but n o t f u r t h e r
have increased. This is evidenced by equal or great- away, which is in a g r e e m e n t with results f o u n d in
er n e t u p t a k e o f silica by the silicate-based phyto- freshwater systems. T h e increased % BSi in Missis-
 Mississippi River Changes and Consequences 401

~. I Hist~ I [ Predictions I

0
~9 ...?

1oo~ '''? = "- --

, , l , ,, 1 i , ,, , i l ,
:= t950 1990
z
Yes N Limitation Some N Limitation Some N Limitation No N Limitation Yes
No Si Limitation Some Si Limitation No Si Limitation No Si Limitation No
4:1 Si:NRatio -1:1 Si:N Ratio >1:1 Si:N Ratio -1:1 Si:N Ratio >>1:1

I DOCUMENTED
RESPONSE I PREDICTED RESPONSE I

BSi Accumulation BSi Accumulation BSi Accumulation ~ BSi Decrease t

C Accumulation C Accumulation C Accumulation ~i~ C Decrease
Hypoxia Increase Hypoxia Increase Hypoxia Severe ~ Hypoxia Decrease t
Fig. 12. A schematic of documented historical changes in riverine nutrient concentrations, nutrient ratios, and biological responses,
and a series of predicted responses depending on a constant increase of silica and varying changes in nitrogen loadings. A stronger
response is indicated by double arrows. (See text for discussion.)

sippi River bight sediments that parallels increased Predictions

N loading to the system is direct evidence for the
N u t r i e n t levels in the Mississippi River may be
effects of e u t r o p h i c a t i o n on the shelf adjacent to
influenced by a variety of m a n a g e m e n t a n d eco-
the Mississippi River. Individual p h y t o p l a n k t o n spe- n o m i c decisions, with c o n s e q u e n c e s to the Missis-
cies c o m p o s i t i o n shifts (heavily silicified diatoms to sippi River delta coastal ecosystem. Agricultural
lightly silicified diatoms; d i a t o m to n o n d i a t o m ) practices, e c o n o m i c incentives a n d disincentives,
would indicate s o m e population-level responses to water quality regulations, water t r e a t m e n t practic-
r e d u c e d Si supplies a n d / o r changes in n u t r i e n t ra- es, a n d c o n s u m e r p r e f e r e n c e s may all c o n t r i b u t e
tios. Finally, analysis o f benthic f o r a m i n i f e r a indi- to the a m o u n t o f dissolved N, P, a n d Si in the Mis-
cates an increase in oxygen deficiency stress this sissippi River. I f freshwater p h o s p h o r u s loading is
century, with a d r a m a t i c increase since the 1950s. r e d u c e d , we predict that the rate o f BSi burial in
I n c r e a s e d bottom-water hypoxia could result f r o m freshwater sediments will decrease because fresh-
increased organic loading to the seabed a n d / o r water d i a t o m p r o d u c t i o n will be lower (similar to
shifts in material flux (quantity a n d quality) to the the d e c r e a s e d % BSi in m o r e r e c e n t Lake Michi-
lower water column. gan sediments; shown in Conley et al. 1993). T h e
T h e i m p o r t a n c e o f the water c o l u m n physical release o f dissolved Si f r o m the sediments in the
structure to the d e v e l o p m e n t a n d persistence o f watershed will e x c e e d the u p t a k e o f dissolved Si in
hypoxia is clear, a n d the discharge of the Mississip- the water column. T h e result will be an increase
pi River (i.e., a m o u n t of flow) since the 1950s has in dissolved Si within the Mississippi River a n d in
b e e n relatively constant aside f r o m n o r m a l decadal the adjacent coastal waters (a r e t u r n to the 1950s
scale variations in runoff. T h e allocation of flow a n d turn-of-the-century levels). At the same time,
between the Mississippi River p r o p e r a n d the At- controls on N use a n d loadings within the water-
chafalaya River has b e e n m a i n t a i n e d by the United shed may or may n o t be affected.
States A r m y Corps of Engineers a c c o r d i n g to Con- Based on the biological c o n s e q u e n c e s of docu-
gressional m a n d a t e . Thus, the o b s e r v e d changes in m e n t e d historical riverine changes, we could sug-
b i o l o g i c a l r e s p o n s e s a r e p r o b a b l y n o t d u e to gest three possible scenarios for future ~ s t e m re-
changes in a m o u n t or distribution o f freshwater sponses (Fig. 12). First, if N c o n c e n t r a t i o n r e m a i n s
r u n o f f a n d resultant stratification. the same a n d Si c o n c e n t r a t i o n increases, overall N
 402 N . N . Rabalais et al.

limitations to phytoplankton p r o d u c t i v i t y will b e I,ITERATURE CITED

s i m i l a r t o p r e s e n t , b u t Si will n o l o n g e r b e l i m i t i n g . ANDER~SSON,L. AND I,. RYr)BI~RG. 1988. Trends in nutrient and
T h e r e s u l t w o u l d b e i n c r e a s e d BSi a n d c a r b o n ac- oxygen conditions within the Kattegat: Effects on local nutri-
cumulation in the sediments, and an increase in ent supply. Estuarine, Coastal and Shelf Science 26:559-579.
the extent and severity of hypoxia. Second, if both ATKINSON, L. E AND D. WAI,I,XC~. 1975. The source of unusually
low surface salinities in the Gulf Stream off Georgia. Deep-Sea
N a n d Si c o n c e n t r a t i o n s i n c r e a s e , t h e r e will b e n o Research 23:91 '3-916.
N o r Si l i m i t a t i o n s a n d t h e S i : N r a t i o will b e bal- BIEN, G. S., D. E. CoN'roIs, AND W. II. Ttlo~cs. 1958. The re-
a n c e d . T h e r e s u l t w o u l d b e g r e a t l y i n c r e a s e d BSi moval of soluble silica from fresh water entering the sea. Geo-
and carbon accumulations and substantial increase chimiea et Cosmochimica Acta 14:35-54.
COOlie,AXE, J. D. a,XD E J. K~:LIX. 1986. l,ow-frequency circula-
in the severity of hypoxia. Third, if N concentra-
tion on the Texas-Louisiana shelf. Journal of Geophysical Re,-
t i o n s a r e r e d u c e d (e.g., t o 1 9 5 0 s v a l u e s ) w h i l e Si search 91 : 1064"5-10659.
concentrations increase, then N would again re- CONLEY, D.J., s. s. KOLHAM,A,XDE. THEm(Yr. 1989. Differences
turn to the limiting nutrient status, and although in silica content between marine and freshwater diatoms. Lim-
Si w o u l d b e i n a b u n d a n t s u p p l y , t h e s y s t e m w o u l d nolo~ and Ocean,ography 34:205-213.
CONLEY, D.J., C. L. [Link],[Link]. E S'I'OE~'~ER. 1993. Mod-
be restricted by N supplies, and hypoxia would de- ification of the biogeochemical cycle of silica with eutrophi-
crease. cation. Marine Ecology. Progress Series 101 : 179-192.
Because of the close coupling between riverine DA(;C;, M.J. 1988. Physical and biological responses to the pas-
nutrient loading and phytoplankton production, sage of a winter storm in the coastal and inner shelf waters
of the northern Gulf of Mexico. Continental Shelf Research 8:
reversal of the current effects of nutrient increases
167-178.
a n d n u t r i e n t r a t i o c h a n g e s is p o s s i b l e t o t h e d e g r e e DICKEY, R. W., G. A. FRYXFI,I,, II. R. GRANADE, AND D. ROEI,K~:.
that water quality trends are altered. However, the 1992. Detection of the marine toxins okadaic acid and do-
management o f o n e n u t r i e n t (Si o r N ) m a y n o t b e moic acid in shellfish and phytoplankton in the Gulf of Mex-
sufficient to reduce eutrophication to an accept- ico. Toxicon 30:355-359.
DIMEGO, G. J., I,. F. BOSART, AND G. W. ENI)ERSEN. 1976. An
able level if the compensatory qualitative adapta- examination of the frequency and mean conditions surround-
tions of species lead to new phytoplankton com- ing frontal incursions into the Gulf of Mexico and Caribbean
munities, including those with noxious or toxic Sea. Monthly Weattwr Review 104:709-718.
species. DINNEL, S. P. ANt) W. J. WISFMAN,JR. 1986. Fresh water on the
I,ouisiana and Texas shelf. Continental Shelf Research 6:765-784.
The preceding discussion has been based on the DORTClI, Q. AND T. E. WHITLEDG~:. 1992. Does nitrogen or sil-
a s s u m p t i o n t h a t r i v e r d i s c h a r g e will r e m a i n statis- icon limit phytoplankton production in the Mississippi River
tically similar to past conditions. This may not be plume and nearby regions? Continental Shelf Research 12:1293-
the case. Possible diversions of fresh water (and 1309.
DORTCH, Q., A. BODE, AND R. R. TWILLEY. 1992a. Nitrogen up-
associated sediments and nutrients) through Lou-
take and regeneration in surface waters of the Louisiana con-
i s i a n a m a r s h e s as a m e a n s o f r e d u c i n g c o a s t a l l a n d tinental shelf influenced by the Mississippi River, p. 52-56. In
loss a n d o r a s s i s t i n g n a v i g a t i o n a r e b e i n g d i s c u s s e d . Proceedings, Nutrient Enhanced Coastal Ocean Productivity
I t is a l s o n o t c l e a r h o w m u c h l o n g e r t h e U n i t e d Workshop. Publ. no. TAMU-SG-92-109, Texas Sea Grant Col-
S t a t e s A r m y C o r p s o f E n g i n e e r s will h e a b l e t o p r e - lege Program, Texas A&M University, College Station, Texas.
DORTCH, Q., D. MILS'FED, N. N. RABAIAIS,S. E. LOHRENZ, D. G.
vent the Atchafalaya River from capturing the ma- REDALJE, M. J. DAGG, R. E. TU~\FR, AND T. E. [Link].
jor part of the Mississippi River discharge. Either 1992b. Role of silicate availability in phytoplankton species
scenario could severely alter the spatial distribu- composition and the fate of carbon, p. 76-83. In. Proceedings,
tion of fresh water and imtrients onto the shelf. Nutrient Enhanced Coastal Ocean Productivity WorLshop.
Publ. no. TAMU-SG-92-109, Texas Sea Grant College Pro-
gram, Texas A&M University, College Station, Texas.
ACKNOWLEDGMENTS [Link], I.., H. PERRY, C. ELEUTERIUS,J. WARRFN, AND J.
(2~mDWELL. 1981. Causative analysis on a nearshore bloom of
Data were collected during programs funded by the National Oscillatoria erythraea (Trichodesmium) in the n o r t h e r n Gulf of
Oceanic and Aunospheric Administration Ocean Assessments Di- Mexico. Northeast Gulf Science 5:1-11.
vision, the I,ouisiana Board of Regents [Link] award 86-LUM(1 )- FUOK, K.W. 1974. The effect of petroleum operations on the
()8~13 and LEQSF award (1987-90)-RD-A-15, the Louisiana Sea phytoplankton ecology of the Louisiana coastal waters. M.S.
(;rant College Program, the National Oceanic and Atmospheric thesis, Texas A&M University, College Station, Texas.
Adminisu'ation National Undersea Research Center, the Louisi- GOLD.V~N,J. ('. ~XD E M. GLmERT. 1983. Kinetics of inorganic
ana State University, the Louisiana Universities Marine Consor- nitrogen uptake by phytoplankton, p. 22'3-274. In E. J. Car-
tium, and the National Oceanic and Atmost)heric Administration penter and D. G. Capone (eds.), Nitrogen in Marine Envi-
Coastal Ocean Program Office, Nutrient Enhanced Coastal ronment. Academic Press, New York.
Ocean Productivity (NECOP) study grant no. NAg0AA-D-SG691 t-tALLEGRAEFF, G. M. 1993. A review of harmful algal blooms
to the I,ouisiana Sea Grant College Program, awards no. MAR31, and their apparent global increase. Phycolog~a 32:79-99.
MAR24, and MAR924)2. We thank the many re~arch a~sistants 11ARRISON, P.J., H. I.. CONWAY, R. W. HOLMES, AND C. 0 . DAVIS.
and associates who helped with the collection and analyses of 1977. Marine diatoms in chemostats under silicate or am-
samples as well as captains and crew of the R/V K J. [Link], R / V monium limitation. III. Celhflar chemical composition and
Ac~diana, and R/V Pelican. Funding for this synthesis and manu- morphology of three diatoms. Marine Biology 43:19-31.
script preparation was provided by the National Oceanic and tlECKEY, R. E. AND P. KILIIAM. 1988. Nutrient limitation o f phy-
Atmospheric Administration NECOP program. toplankton in freshwater and marine environments: A review
 Mississippi River Changes and Consequences 4D3

o f recent evidence on the effects o f enrichment. Limnology QURFSIn, N. A. 1995. The role of fecal pellets in tile flux of
and Oceanography 33:796-822. carhon to the sea floor on a river-influenced continental shelf
IIII.I., A. E. AND J. II. SIMPSON. 1989. O n the interaction of subject to hypoxia, Ph.D. dissertation, D e p a r t m e n t of Ocean-
thermal and haline fronts: Tile lslay fi'ont revisited. Estuarine, ography and Coa_stal Sciences, Louisiana State University, Bat-
Coastal and ,Shelf &ience 28:495-505. on Rouge, I,ouisiana.
JUSTI("., D., T LEGOVId, AND L. ROTTINI-SANDRINI. 1987. Trends RABALAIS, N. N., R. E. TURNER, AND Q. DOR'rCH. 1992a. l,oui-
in the oxygen c o n t e n t 1911-1984 and occurrence of benthic siana continental shelf sedinmnts: Indicators o f riverine influ-
mortality in the n o r t h e r n Adriatic Sea. Fstuarine, Coastal and ence, p. 131-135, hz Proceedings, Nutrient E n h a n c e d Coastal
Shelf Science 25:43.5--445. Ocean Producfi,aty Workshop. Publ. no. TAMU-SG-92-109,
Jt:sTI(:, 1)., N. N. RA~AI.~aS,A.'~DR. E. Tt;PC~ER. 1994. Riverborne Texas Sea Grant College Program, Texas A&M University, Col-
nutrients, hypoxia, and coastal ecosystem evohuian: Riologi- lege Station, "lexas.
cal responses to long-term changes in nutrient loads carried RABAI,Ats, N. N,, R, E, TURNFa, AND W. J. [Link],JR, 1992b.
by the Po and Mis,sissippi rivers, p. 161--67. In BL R. Dyer and Disuibution and characteristics of hypoxia on the [Link]
R.J. Orlh (eds.), Changes in Fluxes of Estuaries: Implications shelf in 1990 and 1991, p. 15-20. [n l'roceedings, Nutrient
from Science to Management. Proceedings of Estuau'ine and E n h a n c e d Coastal Ocean Productivity Workshop. Puhl. no.
Coastal Sciences Associafinn/Fstuarine Resem'ch Federation TAMU-SG-92-109, Texas Sea Grant College Program. Texas
Symposium. International Symposium Series, Olsen & Olsen, A&M University, College Station, Texas.
Fredensborg, Denmark. RABALAIS, N. N., R, E, TURNER, W. J. WISEMAN,JR., AND D, F.
JusrIc, D., N. N. RABAL~IS, AND R. E. TURNER. 1995a. Stoichi- BOESCH. 1991. A brief summary of hypoxia on the n o r t h e r n
ometric nutrient balance and origin o f coastal eutrophication. Gulf of Mexico c o n f n e n t a l s h e l f 1985-1988, p. 35-46. In R.
Marine Pollution Bulletin 30:41-46. V. Tyson and T. H. Pearson (eds.), M o d e r n and Ancient Con-
Jtrs'rI(:, D., N. N. bb~BAI~US, R. E. TURNER, AND Q. DORTCH. tinental ShelfAnoxia. Geological Society Special Pnbl. No. 58.
1995b. Changes in nutrient su'ucture o f river-dominated The Geological Society, London.
coastal waters: Stoichiometric nutrient balance and its con- RABALAIS, N. N., W. J. WISEMAN,.JR., ANt) R. E. TURNER. 1994a.
sequences. Esmarine, Coastal and Shelf Science 40:339-356. Hypoxic conditions in bottom waters on the Louisiana-Texas
JusTt(:;, D., N. N, RABAL~S, R. E. TURNER, AND W. J. W1SEMA.\', shelf, p. 50--54. In M. J. Dowgiallo ted.), Coastal Oceano-
JR, 1993. Seasonal coupling between riverborne nutrients, gTaphic Effects of S u m m e r 1993 Mississippi Rivet" Flooding,
net productivit$; and hypoxia. Marine Pollution Bulletin 26:184- Special National Oceanic a n d Atmospheric Administration
189. Report, United States D e p a r t m e n t of Coinmerce, National
KIMSEY, J. B. AND R. E TEMPLE. 1963. Currents on the conti- Oceanic and Atmospheric Administration, Coastal Ocean
nental shelf of the northwestern Gulf o f Mexico. United Program, Silver Spring, Maryland.
States Burean of Commercial Fisheries Circular 161:23-27. RABALAIS, N. N., W. J, WISEMAN,,JR., AND R. E. TURNER, 1994b.
KIMSE',, J. B. AND R. E TEMI't.F. 1964. Currents on the conti- Comparison o f continuous records of near-bottom dissolved
nental shelf o f tile northwestern Gulf of Mexico. United oxygen from the hypoxia zone of l,ouisiana. Estuaries 17:850-
States Bnreau o f Commercial Fisheries Circular 183:25-27. 861.
I,ODER, T. C. AND R. E REmHARD. 1981. The dynamics of con- REDALJE, D. G., S. E. LOItRENZ, AND G. L. FAIINENSTIEL. 1994.
servative mixing in estnaries. Estuaries 4:64-69. The relationship between primary production and the verti-
LOHRFNZ, S. E,, M . J . DA(;(;, AND T. E. WHITII:DGE. 1990. En- cal export of particulate organic matter in a river-impacted
h a n c e d primary production at the p l n m e / o c e a n i c interface coastal ecosystem. Estuaries 17:829-838.
of the Mississippi River. CoutiTwntal Shelf Research 10:639-664. R~Dr'tELD, A. C, 1958. The biological control o f chemical factors
LOIIRENZ, S. E.. G, L, FAItNENSTIEL, ANt) D, G. REDAIJI-'. 1994. in the environment. American Scientist 46:205-222.
Spatial and temporal variations in photosynthesis parameters RHEE, G. Y 1973. A continuous culture study o f ptmsphate tt D-
in relation to environmental conditions in coastal waters o f take, growth rate, and polyphosphate in &enedesmus sp. Jour-
rim n o r t h e r n Gnlf o f Mexico. Estual~s 17:779-795. nal of PhyroloKv 9:495-506.
MARTIN, J., K HAVA, K'<D L. E. BI:mUDGE. 1990 N~tzschia pseu- ROUND, E E., R. M. CRAWFORD, AND D. G. MANN. 1990. The
dodelicatiss,ma--A source o f domoic acid in the Bay of Fundy, Diatoms. Biology and Morphology o f the Genera. Cambridge
eastern Canada. Marine Ecology Progress S~'/es 67:177-182. University Press, Cambridge.
ME'mEt:K, M. 1982. Carbon, nitrogen, and p h o s p h o r u s trans- SCHELSKE, C. L., D.J. CONI,EY, E. E STOFRMER,T. I,. NEWaEm'~V,
port by world rivers. American Journal of Science 282:401-450. AND C. D. C~dVIPBELI,. 1986. Biogenic silica and p h o s p h o r u s
[Link],J. D. AND R. H. MF~kDE. 1983. World-wide delivery o f accumulation in sediments as indices o f eutrophication in the
river sediment to the oceans. TheJournal of Geology 91:1-21. Laurentian Great Lakes. Hydrobiologia 143:79-86.
NATIONAL OCEANIC AND ATMOSPHI.~RIC ADMINISTRATION. 1987. SCtlEI,SKE, C. L. AND E. 17. STOERMER. 1971. Eutrophication, sil-
National Estuarine Inventory. Data Arias. National Oceanic ica depletion, and predicted changes in algal quality in l~ake
and Atmospheric Administration, National Ocean Service, Michigan. Science 173:423-424.
Strategic Asses,sment Branch, Ocean Asse~ments Division, Of- SEN GUPTA, B. K., R- E LEE, AND M. S. MAY. 1981. Upwelling
fice of Oceanography and Marine AsseSsment, Rockville, and an unusual assemblage o f benthic foraminifera on the
Maryland. n o r t h e r n Florida continental slope. Journal of Paleontology 55:
NEi~So.x, D. M. ~'4D M. A. B~tZE~INSga. 1990. Kinetics o f silicic 853-857.
acid uptake by natnral diatom assemblages in two Gulf Stream SVN GUFTA, B. K. AND M. L. MACHAIN-([Link]. 1993. Benthic
warm-core rings. Marine EcoloD Progress Series 62:283-292, foraminifera in oxygen-poor habitats. Marine Mieropaleouotolo-
Owlet:a, C, B AXnJ. H. RYrtlER. 1980. The possible importance gy 20:183-201.
of silicon in marine eutrophication. Marine Ecology Progress Se- SFN GUPTA, B. K., R. E TURNER, A:~D N. N. RABALmS. 1996. Sea-
7{es 3:83-91. sonal oxygen depletion in continental shelf waters of Lonisi-
OFFICER, C. B., R. B. BIGGS,J. L. TAFT, 1.. E. CRONIN, M. TYLER, ana: Historical record of benthic forantinifers. Geology24:227-
ANt) W. R. BOYNTON. 1984. Chesapeake Bay anoxia: Origin, 230.
development, and significance. Science 223:22-27. SHUMWA~, S. E. 1990. A review o f the effects of algal blooms
PARSONS, "I: R., Y. MALTA, AND M. LALLI. 1984. A manual o f on shellfish and aquacuhure. Journal of the. Wtn'M Aquaculture
chemical and biological m e t h o d s tbr seawater analysis. Per- Society 21:65-104.
g a m o n Press, New York. SIMMONS, E. G. AND W. H. THOMAS. 1962. Phytoplankton o f the
404 N . N . Rabalais et al.

eastern Mississippi delta. Publications of the Institute of Marine WARD, (i. H., M. E. BENDER,AND D.J. REISH (eds.). 1979. The
Science, University of Texas 8:269-298. Offshore Ecology Investigation. Effects of oil drilling and pro-
SKLAR, E H. AND R. E. TURNER. 1981. Characteristics ofphyto- ducdon in a coastal environment. R/ce University Studies 65:1-
plankton production off Barataria Bay in an area influenced 589.
by the Mississippi River. Contributions in Marine Science 24:93- WISEMA~N,W. J., JR., R. E. Tt:RNFR, K IC KELLY,L.J. RousE, JR.,
106. AND R. F. SHAW. 1986. Analysis of biological and chemical
SMAVDA,T.J. 1990. Novel and nuisance phytoplankton blooms associations near a turbid coastal front during winter 1982.
in the sea: Evidence for global epidemic, p. 29--40. In E. Gra- Contributions in Marine Science 29:141-151.
neli, B. Sundstrom, R. Edler, and D. M. Anderson (eds.), Tox-
ic Marine Phytoplankton. Elsevier Science Publishing Co., Received for consideration, February 21, 1994
New York. Accepted for publication, August 15, 1995
SMITH, N. P. 1980. On the hydrography of shelf waters off the
central Texas Gulf Coast. Journal of Physical Oceanography 10:
806-813. APPENDICES
TESTER, E A. A.'<DL. E A'rgJNSON. 1994. [Link] salinity water in Data from Simmons and Thomas (1962) included four areas
the Gulf Stream off North Carolina, p. 72-76. In M.J. Dow- east of the Mississippi River delta (two with salinity below 18%o
giallo (ed.), Coastal Oceanographic Effects of Summer 1993 combined under "River" in Appendix 1, and two with salinity
Mississippi River Flooding, Special National Oceanic and At- of 18-32%o combined under "Plume" in Appendix 2). Most of
mospheric Administration Report, United States Department the 1990-1992 data are from west of the delta but are matched
of Commerce, National Oceanic and Atmospheric Adminis- for salinity and distance from the river (no stations west of
tration, Coastal Ocean Program, Silver Spring, Maryland. 89~ The July 1990 data included stations east of the delta,
THOMAS, W. H. AND E. G. SIMMONS. 1960. Phytoplankton pro- but there were no substantial differences in species composition
duction in the Mississippi River Delta, p. 103-116. In F. P. from Simmons and Thomas (1962), and the data were exclud-
Shepard (ed.), Recent Sediments, Northwest Gulf of Mexico. ed. Methods used by Simmons and Thomas (1962) were reliable
American Association of Petroleum Geologists, Tulsa, Okla- for diatoms >0.6 btm, thus Appendices 1 and 2 include only
homa. diatoms.
TURNER, R. E. AND N. N. RABALAIS. 1991. Changes in Mississippi The data of Fucik (1974) and Ward et al. (1979) were ob-
River water quality this century. Implications for coastal food tained from a study of petroleum operations approximately 80
webs. BioScience 41:140-147. km west of the Mississippi delta. Their data ("Control" and
TUmXER, R. E. AND N. N. [Link]. 1994a. Changes in the Mis- "Platform") are comparable to monthly data in 1990-1993
sissippi River nutrient supply and offshore silicate-based phy- from nearby stations C6A and C6B (Appendix 3 and Fig. 3).
toplankton community responses, p. 147-150. In K. R. Dyer High river flow in spring 1973 and spring-summer 1990 and
and R.J. Orth (eds.), Changes in Fluxes in Estuaries: Impli- 1993 makes these periods most comparable. Neither Fucik
cations from Science to Management. Proceedings of Estua- (1974) nor Ward et al. (1979) cited enumeration methodolo-
rine and Coastal Sciences Association/Estuarine Research gies, but their species lists indicate they used the standard Uter-
mohl method. Their listings do not mention any other groups,
Federation Symposium. International Symposium Series, Ol-
besides diatoms and dinoflagellates, which should have been
sen & Olsen, Fredensborg, Denmark.
observed by this method if they were present. For example,
TURNER, R. E. AND N. N. RABALAIS. 1994b. Coastal eutrophica-
cryptomonads are now often dominant, especially in the fall
tion near the Mississippi river delta. Nature 368:619-621. through early spring, but were not listed by Fucik (1974) or
TUIL"~ER, R. E., N. N. RABAIAIS,AND Z.-N. ZHANG. 1990. Phyto- Ward et al. (1979). Consequently, Appendix 3 includes only di-
plankton biomass, production, and growth limitations on'the atoms, dinoflagellates, and large filamentous cyanobacteria,
Huanghe (Yellow River) continental shelf. Continental Shelf Re- which they would surely have observed if they had been present.
search 10:545-571. Water samples for 1990-1993 were preserved with glutaral-
UNITED STATES ARMYCORPS OF ENGINEERS. 1974. Deep draft dehyde, and filtered onto 0.2-p,m, %~m, and 8-p~m polycarbon-
access to the ports of New Orleans and Baton Rouge. Draft ate filters (Dortch et al. 1992b). Organisms on the ~p.m and 8-
Environmental Statement, United States Army Corps of En- p~m filters were stained with proflavin. The 0.2-p~m and 3-p~m
gineers, New Orleans Disu'ict, New Orleans, Louisiana. filters were counted immediately by epifluorescence microsco-
WALKER,N. D., G. S. FARGION, L. J. RousE, AND D. C. BIGGS. py, and the 8-~zm filters were frozen and counted later. Domi-
1994. The Great Flood of Summer 1993: Mississippi River nant species for each area and time were picked by ranking the
discharge studied. Eos, Transactions, American C-eophysical Union phytoplankton by concentration, summing the ranks, and pick-
75:409, 414-415. ing the top-ranked groups.
 Mississippi River Changes and Consequences 405

9 ~,

~v:E
~,~v ~'E

o
E

~f3

-., ~.~ ~

0
v "~
~.~ .~
~, ~.~,

V ~

oo
V

e~ ~ .~ ~o~
E
o

8
z o
~ m :z
m
N ~
406 N.N. Rabalais et al.

E
o

V ~

.~ m ~6~
~-~'~

~ ~.~ Z.~ --.~

v
i
~'~ ,~ r~e I
,~ ~ ~~ ~ ' ~ ~
'~ ~ '~ ~ 1
I
_~ ~ ~ ~ ~ . ~ -~l
o
~e ~r-. e ~ ~ ~ ~eI

~I .~ ~ 9
g

" ~ v ~
~ ' ~ . ~ ~ ~L
E
o

v
.-

e~ F_
~ ~, .~ ,.~ ~.~ ~'~
c ~
~ ~.;
N ~
x
r,
~.~ 8
z ~ t-.- o
Z
<
 Mississippi River Changes and Consequences 407

oh, ~. .e. v
& ~,
dh
Ob

~ ~ ~ ~' ~ ~-~ ',-

~~ , ~.~ -~ d~.~{ ~ ~,
9 ~ ~

~z ~ ~z ~ e, r:. .~,..
I2".-

d.,
e,~
v

!}i
..m

I
~~-~ .~ ~

~ { ~
,.0
2- e'~ ~ ~ ~ ~ ~ e'~
0

~_g " .~ ~ ~ .~.

,0

g~

=1. ,.- 5

e',

5
r

<

~i ~ iii i ~i