Sponsored Jby
GuJf of Mexico Program
Louisiana Department of Environmental Quality
Lower Mississippi River Conservation Committee
Mississippi Soil and Water Conservation Commission
-------
-------
Points of view expressed in this report do not necessarily reflect the views or policies of the Gulf of
Mexico Program nor of any of the contributors to its publication. Mention of trade names and
commercial products does not constitute endorsement of their use.
.
\
I
United States
Environmental Protection Agency
Gulf of Mexico
Program Office
EPA-55-R-97-001
August 1997
To obtain copies, contact:
Gulf of Mexico Program Office
Public Information Center
Building 1200, Room \ 03
Stennis Space Center, MS 39529-6000
Phone: (601)688-7940
FAX: (601)688-7869
GIN Address: http://pelican.gmpo.gov
-------
-------
'Table of Contents
Science, Policy and Coastal Eutrophication: the Chesapeake Experience
Gulf of Mexico Hypoxia Management Conference Presentation
EPA Committed to Addressing Gulf Hypoxia
Hypoxia in the Gulf—Who's Problem Is It?
Hypoxia in the Northern Gulf of Mexico: Past, Present and Future ....
Responses of Benthonic and Nektonic Organisms, and Communities, to Severe
Hypoxia on the Inner Continental Shelf of Louisiana and Texas
Physical Variability in the Louisiana Inner Shelf Hypoxia Region
Trends in Shrimp Catch in the Hypoxic Area of the Northern Gulf of Mexico
1
12
17
23
25
41
57
64
Distribution, Abundance, Feeding Growth and Mortality of Fish Larvae Associated
with the Mississippi River Discharge Plume, and the Potential Impacts of Hypoxia 76
Potential Impacts of Hypoxia on Fisheries: Louisiana's Fishery-Independent Data 87
Estuarine Hypoxia: The Mobile Bay Perspective 101
Causes and Effects of Coastal Hypoxia Worldwide: Putting the Louisiana
Shelf Events in Perspective 102
Evidence for Nutrient Limitation and Sources Causing Hypoxia on the Louisiana Shelf 106
Estimated Responses of Water Quality on the Louisiana Inner Shelf to Nutrient Load
Reduction in the Mississippi and Atchafalaya Rivers 114
Presentation Discussion by John Day 126
The Regional Transport of Point and Nonpoint-Source Nitrogen to the Gulf of Mexico 127
Spatial Distribution of Nutrients in the Mississippi River System (1991-1992) 133
Effects of Episodic Events on the Transport of Nutrients to the Gulf of Mexico 144
Estimating Background Loads of Sediments, Organic Nitrogen, and Organic
Phosphorous in the Mississippi River Basin 146
Nitrogen Loading in the Upper Mississippi River 159
Estimates of Atmospheric Deposition to the Mississippi River Watershed 160
III
-------
Nutrients in Streamflow of the Mississippi River Basin—Annual Mean Concentrations,
Annual Loads, and Temporal Trends in Concentrations and Loads 174
An Assessment of Watershed Based Projects in the Mississippi River Drainage Basin 176
What is Being Done in the Basin Now to Control Nutrient Loads? 184
What is Being Done in the Basin to Control Nutrient Loads from Agricultural Sources? 185
Louisiana Activities and Programs in Nutrient Control and Management 196
IV
-------
Science,
the
_ _ _
Donald F. Boesch ^rl
University of Maryland
Center for Environmental'and ^tt/jjig
Cambridge, Maryland
-4*--,
•Abstract
and Policy Development
Related to Eutrophication of
• "•« - . • x-x , ' "--
Coastof Ecosystems
In assessing the policy options appropriate
to addressing hypoxia in the northwestern
Gulf of Mexico, it is helpful to learn from
experiences in the use of science elsewhere. The
Chesapeake Bay and other U.S. coastal systems,
and northwestern Europe provide instructive case
studies. Coastal eutrophication commonly results
from nutrient inputs from diffuse sources, thus
effects are not easily seen as connected to causes.
Both science and policy must address the scale
mismatches over space (responsible actions far
removed from effects) and time (present, pre-
existing and future conditions) scales and human
activity sectors (e.g., agriculture and fisheries).
At the first level, science must address the
question of whether such effects as hypoxia or
algal blooms have worsened over time and
whether they have harmful or undesirable con-
sequences. Next, ties need to be made between
nutrient inputs and responses (e.g., in the Chesa-
peake major disputes were waged as to whether N
or P limited algal production). Models have been
very helpful in summarizing complex science in a
way that policy makers could pose "what if "
questions and have, in some cases, become the
central technical tool of management. In the
Chesapeake early models were used to justify the
goal of 40 percent reduction in controllable
nutrients and more sophisticated models are now
used to reassess goals and address subsequent
policy issues, such as the effect of Clean Water
Act ozone reduction on water quality. Finally,
policy makers, and society in general, are unlikely
to undertake ambitious environmental restoration
goals without technological advances that make
them feasible (for example, the application of
biological nutrient removal for sewage treatment
and cost-effective best management practices in
agriculture).
Introduction '„
*• p* X4<1 ''
Eutrophication, the excessive enrichment of
aquatic ecosystems which leads to increased pri-
mary productivity, noxious algal blooms, food-
chain alterations, and depletion of dissolved oxy-
gen, is a widespread phenomenon in coastal
waters of developed nations. Furthermore, con-
trary to many marine pollution problems which
have been ameliorated by waste treatment and
other controls, coastal eutrophication has been in
ascendancy during the latter half of the twentieth
century (National Research Council, 1994). This is
due in large measure to the important contribu-
tions of diffuse sources of nutrient inputs—from
agricultural fertilizers, runoff from developed
land, atmospheric deposition, and soil erosion—
in addition to more controllable point sources,
-------
such as sewage or industrial discharges. The
multitude of sources and the remote nature of
many of the sources make them difficult to
Identify, much less control.
In a number of places in the United States,
Europe and Japan, once the degradation of the
environment due to coastal eutrophication was
identified, extensive efforts are being undertaken
to reduce the point and diffuse sources of nutri-
ent inputs from human activities. Science has
played and is playing a major role in identifying
the causes and consequences of coastal eutrophi-
cation and in finding solutions. Although com-
plete success has been met nowhere, progress is
being made in establishing nutrient reduction
goals, targeting sources and achieving reductions.
Perhaps the most famous, and certainly the most
ambitious, of these efforts is in the Chesapeake
Bay and its watershed.
To help set the stage for discussion of manage-
ment of widespread oxygen depletion, or hypoxia,
attributed to eutrophication in the northern Gulf
of Mexico, I will briefly review the role science
has played in the development of policy and
management solutions for controlling nutrient
over-enrichment in the Chesapeake Bay. My goal
is to examine the role of science in addressing
critical policy questions so that lessons may be
applied to the case of hypoxia in the northern
Gulf of Mexico.
*
'""" ~~~ j
The Challenge of Scale Mismatches
At the heart of the problem of understanding and
managing coastal eutrophication is the challenge
of scale mismatches inherent in the problem. As
Lee (1993) puts it: "when human responsibility
does not match the spatial, temporal or functional
scale of natural phenomena, unsustainable use of
resources is likely, and it will persist until the
mismatch of scales is cured."
In the case of coastal eutrophication, perhaps the
most obvious scale mismatch is on the spatial
scale. In contrast to most coastal environmental
and resource issues, for example waste disposal,
coastal development, habitat destruction, and
resource exploitation, the sources of eutrophica-
tion may not be considered coastal at all. They
may originate far up the watershed, hundreds of
miles away, or even from outside of the watershed
in the case of atmospheric deposition. This poses
significant challenges for science in addressing
these large scales. Even more importantly, it is
hard for people to be aware of or accept that their
activities in Pennsylvania or Iowa may be causing
a problem so far away at the coast. Furthermore,
they may see efforts to reduce nutrient inputs
without benefit to them. The beneficiaries are
those on the coast in Maryland and Virginia or
Louisiana and Texas.
In addition, there are mismatches on a functional
scale, for example the cause of coastal eutrophi-
cation may be related to agriculture, waste dis-
posal, or power generation, but the deleterious
consequences may be felt in marine fisheries.
Even within the coastal and marine environment
it is difficult to connect human responsibility for
fisheries with that for water quality or, in the case
of the Mississippi delta, wetland restoration with
offshore hypoxia.
Finally, there are important temporal scale mis-
matches. In the case of coastal eutrophication
these frequently occur because of a lack of under-
standing of what conditions were like before the
increase in anthropogenic inputs of nutrients and
to what degree and how rapidly the system will
recover.
te. •»-."- " /«"7 ""*"" * " / -~r-v —^ —•—,
IKey Questions '• " ^
There are several key questions which must be at
least partially addressed before public support and
-------
political will can be marshaled sufficiently to
undertake and affect the control of nutrient inputs
into coastal environments in order to reduce
hypoxia. In logical order they are:
1 . Is the hypoxia a natural phenomenon? Has it
worsened?
2. What are the consequences of hypoxia for
resources and environmental quality? Does it
matter?
3. Is hypoxia caused by increased nutrient inputs
from human activities?
4. What are the sources of these excess
nutrients?
5. What will be the effect of reducing these
nutrient inputs?
6. How can the sources of nutrients be feasibly
reduced?
7. What are the incentives for reducing these
sources of nutrients?
AH of these pose significant challenges to science
and technology. In order to effect solutions, it is not
sufficient just to describe hypoxia, put it into his-
torical context, identify causes, pinpoint sources and
predict the consequences of nutrient inputs, as
difficult as these task are. Science and technology
must also help find feasible means for reducing
inputs and contribute to the development of
incentives to accomplish this goal. I will briefly
review how these questions were or are being
addressed in the case of the Chesapeake Bay.
Experience of the Chesapeake
The development of scientific understanding of
the effects of nutrient loading to the Chesapeake
Bay and the impact that science had on policy
development has been thoroughly reviewed by
Malone et al. (1993). Based on their observations
and my own I review how the seven questions
have been addressed for the Chesapeake.
Question 1: Is the hypoxia a natural
phenomenon?
Although anoxic conditions in the deeper waters
of the Chesapeake Bay had been observed since
measurements were first made, it was not until
1984 that Officer et al. argued in a paper in Science
that there was an increase in oxygen depletion
between .1950 and 1980. This interpretation
engendered controversy and Seliger et al. (1985)
countered that when corrected for river inflow,
which increases density stratification and nutrient
mass loading, there was not a statistically signi-
ficant trend (Seliger et al., 1985). About the same
time, results from studies initiated to investigate
the dramatic reduction of submersed aquatic
vegetation in the Bay during the 1970's also
implicated increased nutrient loading (Kemp et
al., 1983). Interestingly, growing public concern
about the health of the Bay, coupled with an
assessment process begun in the late 1970's, set
the stage for policy commitments in 1987 to
reduce controllable nutrients by 40 percent by the
year 2000 even though Question 1 had not been
fully answered.
Subsequently, Cooper and Brush (1991) and
Cooper (1995) were able to demonstrate from
biological and chemical indicators in cores that:
(a) hypoxia was part of the Bay's ecology for a
long time; (b) eutrophication due to human
activities began with extensive land clearing in the
late 18th century; and (c) anoxic conditions have
become more frequent and persistent during the
mid-20th century in association with both rapid
population growth and the advent of widespread
use of artificial fertilizers. From their and other
work, one can sketch a modern environmental
history of the Chesapeake Bay (Table 1) in the
context of which can be placed the more recently
observed deterioration of the environment and
the restoration goals.
-------
Table I.
The modern environmental history of the Chesapeake Bay
Period
Years
Description
Pre-Colonial and Early
Colonial
(I980)
Recognition of problems associated with nutrient
over-enrichment leads to ban of phosphates in
detergents, nutrient removal in sewage
treatment, improved agricultural management
practices, population growth management,
reduction of emissions from vehicles, and
restoration of oyster populations for
environmental rather than commercial reasons
Question 2: What are the consequences of
bypoxia?
Somewhat surprisingly, very little has been done
to quantify the impacts of hypoxia on living
resources, either before the 1987 policy commit-
ment to reduce nutrient loading or subsequently.
Although the effects of seasonal hypoxia on
benthic organisms have been well documented
(Holland et al., 1987) and the physiological
tolerance of motile fishes and invertebrates
indicate that they cannot survive long in near-
anoxic waters (Sea Grant Programs, 1992), the
effects on fisheries and shellfisheries have not
been direcdy assessed. In fact, landings data do
not show an increase or decline in the total
fisheries productivity during the latter half of the
twentieth century, although there has been a shift
in harvested biomass from benthic (e.g., oysters)
to pelagic (e.g., menhaden) species. Rather, the
-------
reductions in important fishery resources that
have occurred are thought to be primarily the
result of over fishing, habitat modification and
barriers to anadromous fish migration. Hypoxia in
the Chesapeake Bay has generally been assumed
to be detrimental to living resources, but its role
relative to these other pressures is not well
understood.
On the other hand, scientific understanding of
exchange of nutrients with bottom sediments
indicates that anoxia has a positive feedback to
eutrophication (Boynton et al., 1995), further
reducing water quality. Hypoxia in bottom waters
results in episodic fluxing of phosphorous from
sediments. Persistent anoxia can also greatly
reduce denitrification rates by shutting down
nitrification of ammonium. The net effect is rapid
recycling of phosphorous and attenuation of the
nitrogen sink, resulting in increased phyto-
plankton production, decreased water clarity, and
increased oxygen demand in bottom waters.
More attention has been given to the other effects
of eutrophication on living resources, such as the
loss of submersed aquatic vegetation, decreases in
water clarity, and changes in food chains. Several
lines of evidence suggest that nutrient enrichment,
at least during this century, has not resulted in
increases in harvestable secondary production.
The Chesapeake Bay is characterized by high
primary productivity relative to nutrient loading
and high secondary productivity relative to its
primary productivity (Nixon et al., 1986). This
suggests that factors other than nutrient availabi-
lity are controlling secondary production in this
system at this point in its eutrophication history.
Question 3: Is hypoxia caused by increased
nutrient inputs?
The Chesapeake Bay restoration effort is well
known for its commitment to reduce controllable
loadings of nutrients by 40 percent, a commit-
ment which is now involving hundreds of waste
discharges, thousands of farmers and citizens, and
massive investments of public and private funds.
It is important to keep in mind that it has been
just since the early 1980's that the notion that the
Bay was suffering ill effects from excess nutrients
began to gain wide acceptance (Malone et al.,
1993). During the late 1960's, massive algal
blooms, oxygen depletion and fish kills in the
upper tidal (freshwater) Potomac River led to
large investments in the early 1970's for sewage
treatment facilities for the metropolitan
Washington area. These treatments included
removal of phosphorus, which was known to be
the culprit in causing eutrophication in freshwater
lakes, as well as BOD. Substantial improvements
in water quality in the tidal freshwater Potomac
resulted (Jaworski, 1990).
It took about a decade more to conclude that the
Bay had undergone widespread eutrophication, to
understand that excess nitrogen as well as phos-
phorus was posing a problem, and to effect policy
for broader control of point sources of nitrogen
and phosphorous. The scientific community first
presented inferential evidence such as data on
trends in nutrient loadings and nutrient ratios
which suggested that nitrogen may be limiting
primary production in more saline parts of the
Bay (Boynton et al., 1982). This evidence
suggested that nitrogen removal was needed in
new sewage treatment facilities then being
planned for the upper Patuxent River estuary in
order to avoid further declines in water quality, a
view that conflicted with the position of federal
and state agencies that only phosphorus removal
was required. The science and uncertainties were
debated in court and in a conflict-resolution
conference which concluded with the agreement
to remove nitrogen as well as phosphorus in these
new facilities. Only later was more direct evidence
of nitrogen limitation of phytoplankton growth
provided from mesocosm studies (D'Elia et al.,
1986). Meanwhile, investigations of the cause of
the widespread decline of submersed aquatic
vegetation contributed more to the understanding
of the consequences of nutrient enrichment than
did studies addressing the relationship of nutri-
5
-------
cnts to hypoxia (Malone et ah, 1993). These find-
ings also brought attention to the importance of
nonpoint sources to the overall eutrophication of
the Bay.
Development of policies which culminated in the
regional federal-state commitment to reduce con-
trollable inputs of nitrogen and phosphorous by
40 percent were initiated with very modest under-
standing of the relationship of these inputs to
hypoxia. As Malone et ah (1993) observed: "In a
qualitative way, these links made sense, but the
scientific evidence needed to make the case
remained weak." Nonetheless, in 1983 the EPA
published a framework for action for the Chesa-
peake Bay which focused on the control and
monitoring of nutrients "to reduce point and
nonpoint source nutrient loadings to attain
nutrient and dissolved oxygen concentrations
necessary to support the living resources of the
Bay." This set the stage for the first multi-juris-
dictional Bay Agreement, a flurry of legislative
actions, the establishment of a monitoring pro-
gram, and a decade of intense scientific studies of
nutrient dynamics. Findings of these studies have
greatly enriched our understanding of the sources
of nutrients, their effects on plankton production
and submersed aquatic vegetation, nutrient cycl-
ing and loss (Boynton et ah, 1995), and their
effects on hypoxia (Sea Grant Programs, 1992).
These have confirmed the wisdom of the policies
developed earlier based on much more limited
scientific evidence.
Question 4: What are the sources of these
excess nutrients?
Although it was long understood that nonpoint
sources of nutrients must be significant for an
estuary with such a large watershed as the Chesa-
peake Bay, early management efforts focused on
controlling point sources, particularly sewage
treatment plants. Concerns about the health of
the Bay led to a five year study begun in 1977
which yielded the first good estimates of total
nutrient load to the Bay. These showed the
quantitative importance of nonpoint sources.
Current best estimates are that agricultural lands,
developed areas, and atmospheric deposition
contribute 2.5 and 1.8 times the amount of nitro-
gen and phosphorus, respectively, that point
sources deliver (Table 2). As a consequence,
extensive efforts are underway to reduce nutrient
losses from agriculture—by applying "best man-
agement practices" and nutrient management pro-
cedures throughout the watershed—and from
developed areas—through sediment erosion
controls, storm water management, and limita-
tions to shoreline development.
Table 2.
Estimated portions of total nutrient loads
that enter the Chesapeake Bay from
land-based sources and activities,
i.e., excluding inputs from the ocean
(Magnien et a/., / 995).
Forests
Agriculture
Development
Atmosphere*
Point Sources
- N J
18%
39%
9%
11%
23%
;i P
3%
49%
8%
6%
34%
* Includes deposition directly on the water;
atmospheric N deposition on land
accounts for an additional 1 6% of total N
load, but is included as part of forest,
agriculture and developed land sources.
The agreement to reduce controllable inputs of
nitrogen and phosphorus into the Chesapeake
Bay includes only a portion of the total nutrient
loadings apportioned in Table 2. Excluded from
the total controllable pool are inputs into the
small parts of the watershed located in Delaware,
New York, and West Virginia, states not
signatories to the Chesapeake Bay Agreement.
Furthermore, there is assumed to be base inputs
from various land uses (i.e. agriculture, developed
land, and forests) which are not amenable to
reduction. Also, atmospheric inputs were not
included at all in the definition of controllable
sources, but these too are receiving increased
-------
attention for control strategies, particularly for
nitrogen. This is a difficult challenge because a
significant fraction of the atmospheric deposition
of nitrogen in the watershed emanates from
sources outside of the watershed, particularly
from the Ohio River basin.
Question 5: What will be the effect of
reducing these nutrient inputs?
The expectations regarding the degree of
improvement of conditions in the Bay which
could be achieved by reduction of controllable
nutrients by 40 percent were general, hopeful and
not very quantitative when the policy was
adopted. Since then, as mentioned earlier, much
has been learned about the dynamics of nutrients
in the Bay and the relationship of nutrient inputs
to both hypoxia and conditions for growth of
submersed aquatic vegetation. This increased
understanding has been rather rapidly
incorporated into various models which allow a
scientific assessment of the effects of the nutrient
reductions targeted and others which may be
achievable.
Particularly prominent has been the three-
dimensional, time-variable numerical model of the
Chesapeake Bay which has evolved over the last
decade or more (Figure 1). This model
incorporates hydrodynamic, geochemical and
biological processes and allows one to predict
dissolved oxygen conditions and nutrient
concentrations (useful in assessing the effects on
submerged vegetation) as a function of point and
nonpoint source inputs. The linked watershed
model includes inputs from atmospheric
deposition and meteorological models and is
adjustable for changes in land use. The watershed
model determines the effects of nutrient loading
changes throughout the watershed on delivery of
nutrients by rivers to the Bay but is, in general,
not founded on the same level of scientific
understanding as is the Bay model.
Large Scale Atmospheric
Deposition Model
Meteorology
Input
1
Land Use, Soil
& Geophysical
Characteristics
, t
Watershed Model
Hydrologic
Submodel
Nonpoint Source
Submodel
Transport
Submodel
Three-dimensional Bay Model
Row
->•
Oce
Bounc
Fore
> '
an
:ary
ng
Surface
Forcing
i
Hydro-
dynamic
Model
! Lagrangian •
Water Quality
Model
Sediment
Submodel
Ocean Boundary
Submodel
Submodel.
Nutrient Loads
Figure /.
Schematic depiction of the linked watershed and
three dimensional Chesapeake Bay numerical models
used to guide assessments of the relationship of
nutrient inputs to hypoxia and other ecosystem
effects in the Chesapeake Bay Program.
These models have been used to predict changes
in the extent of hypoxia in the Bay under future
conditions. This has allowed a re-evaluation of the
original goal of 40 percent reductions of nitrogen
(about 20 percent of total input) and phosphorus
(about 30 percent of the total input). The models
predict a decrease in anoxic volume-days of just
21 percent with a 40 percent reduction in control-
lable nutrients and even at the limits of nutrient
control technology anoxia (<1 mg/1 dissolved
oxygen) would only be reduced by 32 percent.
Also depicted in Figure 2 are the effects on anoxia
of reducing atmospheric nitrogen inputs through
implementation of Clean Air Act Amendment's
goals for ozone attainment via reductions in NOX
emissions. These model predictions inject some
reality for environmental quality objectives: the
Chesapeake Bay will exhibit significant summer
hypoxia even with more rigorous pollution con-
trols as long as much of the watershed is in agri-
cultural production employing artificial fertilizers,
handles the wastes of 15 million people, and
receives the residuum from extensive fossil fuel
combustion.
-------
^Present condition
40% reduction
4O% reduction* Clean Air Act!
JUmitS-QfJechnoloav
0 SO 100 150 200 250 300 350 400
Anoxia billion m^days
Figure 2.
Predicted effects on the extent of anoxia
in the Chesapeake Bay as a result of
achieving various management goals
(Thomann, et at, 1994).
More optimistically, there may be significant
environmental quality improvements beyond the
reduction of anoxia in the Bay as a result of
meeting the 40 percent reduction. These include
significant improvements in dissolved oxygen in
waters presently hypoxic, but not anoxic, and
reduced ambient concentrations of nutrients to
levels which permit growth and survival of
submersed aquatic vegetation (Dennison,
ct ah, 1993).
The watershed and Bay models also allow
scientists and managers to identify the processes
and responses which are poorly known but sig-
nificantly influence the models' predictions so
that research and monitoring may be strategically
focused. Among the critical questions now
receiving attention are how agricultural practices
affect rates of loss of nitrogen into ground water,
residence time in ground water for the wide
variety of geological conditions which exist in the
Bay, nutrient retention as a function of forest age,
and denitrification processes in wetlands and Bay
sediments. Resolution of such issues is important
at this juncture because, although phosphorus
concentrations have been significantly reduced in
the Bay as a result of point and nonpoint source
reductions, riverine fluxes of nitrogen and con-
centrations of nitrogen in the Bay have not yet
decreased (Boynton, etal., 1995).
Question 6: How can the sources of nutrients
be feasibly reduced?
Experience indicates that until feasible solutions
to problems are identified, the will to correct the
problems will not be mobilized no matter how
undesirable the consequences. With regard to
nutrient reductions in the Chesapeake, the dram-
atic improvements in water quality which resulted
from advanced waste treatment at the top of the
Potomac estuary had built some confidence that
feasible, albeit expensive, solutions could be
found. In addition, following this experience and
the experience of the Great Lakes, phosphate
detergents were banned in many jurisdictions in
the Chesapeake watershed. However, the difficul-
ties and formidable costs associated with reducing
sources of nitrogen were, in no small measure,
responsible for the denial by both federal and
state agencies that the Bay was being over-
enriched with nitrogen. The demonstration of the
feasibility of biological nutrient removal (BNR) as
a cost-effective sewage treatment technology
helped overcome this resistance. BNR is now
being implemented or planned for most sewage
treatment works discharging to the tidal waters of
the Bay.
Controlling the agricultural sources of nutrients
has also proven challenging. Upton Sinclair once
observed: "It is difficult to get a man to under-
stand something when his salary depends upon
his not understanding it." Similarly, there was at
first strong denial that agricultural uses of fertiliz-
ers and animal wastes could be contributing to
eutrophication in the Chesapeake Bay, so far
away. To a certain degree, such skepticism still
exists in the agricultural community as it con-
fronts additional costs in implementing BMPs and
the threat of government regulation. Nonetheless,
extensive efforts are underway to implement man-
agement practices such as minimum-till, buffer
strips, and reforestation of riparian areas. How-
ever, there is growing evidence that such conven-
tional methods are not as effective as projected in
controlling nitrogen losses. Many of these
-------
methods were developed to reduce soil losses.
Although they are relatively effective in control-
ling phosphorus losses, they are less effective for
the retention of nitrogen, which is more soluble
and escapes into ground water. Other approaches,
such as winter cover crops, precision agriculture
which minimizes the amount of fertilizer applied
to just that required by the crop, and new animal
food formulations which reduce nutrient losses
via manure, are beginning to be applied, but are
more costly.
The commitment of the states to reduce nutrient
inputs to the Chesapeake Bay has also become a
pervasive organizing principle for many other
aspects of environmental management. The impor-
tance of forests in the watershed in retaining
nutrients is being considered in forestry and
reforestation. Retention of nutrients is also an
important issue in the conservation of nontidal
wetlands. Management of human population
growth is another issue that is being pursued in part
because of the impact of land development and
sprawl on nutrient inputs to the Bay. These growth
management goals coincide with those related to
regional transportation, infrastructure, natural
heritage, and community development. And, finally,
as mentioned above, reduced NOX emissions are
being sought not only because of concerns about
their effects on ground level ozone but also because
of their contributions via atmospheric deposition of
nutrients. In short, commitments to reduce nutrient
inputs are providing impetus to move toward
ecosystem management of the Chesapeake Bay and
its watershed.
Question 7: What are the incentives for reduc-
ing these sources of nutrients?
It is not enough to identify the problem, its causes
and feasible solutions. The solutions must be
mandated or there need to be incentives to imple-
ment the solutions voluntarily. In the Chesapeake,
these incentives included state and federal finan-
cial assistance in construction and upgrading of
sewage treatment plants and implementing storm-
water management programs. In addition tech-
nical assistance and matching funds have been
provided to assist farmers in implementing BMPs.
However, these sources of public financing have
been greatly reduced in recent years as a result of
changes in federal legislation and the competing
fiscal pressures at the federal and state level.
Realizing that one often cannot simply mandate
nutrient reductions from nonpoint sources from
Harrisburg, Annapolis or Richmond, the Chesa-
peake Bay Program is pursuing Tributary Strate-
gies which involve the diverse local interests
within each of some 30 sub-watersheds to find
effective community-based approaches to meet
goals for reducing nutrient inputs. These solu-
tions may include trading off reductions from one
source for another, tax incentives, growth
management, and local ordinances.
Another Activating factor, particularly for com-
munities located well up in the watershed, far-
removed from the Chesapeake Bay, is that there are
many local benefits to be had in reducing nutrient
inputs. These include improving water quality in
lakes, streams and rivers, reducing groundwater
contamination, managing population growth,
forested and agricultural land preservation, and
improving air quality. Also, in the long term there
are frequently cost savings in pollution prevention.
For example, effective nutrient management in
agriculture may actually save costs by reducing the
amount of fertilizer which must be purchased.
Driving Forces k x ": '
There are a number of reasons why the Chesa-
peake Bay region has led the way in terms of
recognition of the effects of large scale eutrophi-
cation and making major commitments to reduce
nutrient inputs in order to restore this ecosystem.
These include at least the following:
• The region has a history of sustained
scientific investigation, including strong basic
research and monitoring.
9
-------
* Many people regularly see the Bay and enjoy
it. Catastrophic events and decimated
resources brought attention to the problem.
• There were very effective political champions
who emerged at the right time.
* Even though the region is large, there is a
sense of community or a sense of "place"
about the Chesapeake. This is evident in
polling which shows that the importance
which the public assigns to restoring the
Chesapeake is as high in headwater regions as
it is on the Bay front.
" The public's attitudes about the Bay, coupled
with the educational and economic status of
the region, creates a condition in which there
is strong public support for Bay restoration.
For example, the Chesapeake Bay Founda-
tion has over 84,000 dues paying members.
* There is a voluminous flow of information
about the Bay, both in the popular media and
via other periodicals such as the widely
distributed 'Bay Journal.
• The Chesapeake has received national atten-
tion, but although there have been some
advantages to proximity to the nation's
capital in garnering federal support, this has
been, in my opinion, less important than
those who envy our progress in the Chesa-
peake think.
* Certainly, these conditions do not always
exist in other areas which are experiencing
eutrophication, such as the northern Gulf or
Mexico. Nonetheless, my observations of
how the seven key questions have been
addressed in the Chesapeake do, hopefully,
suggest some focal points and shortcuts
toward resolution of issues and development
of effective solutions for the deleterious
effects of coastal eutrophication.
References
Boynton, W.R., W.M. Kemp and C.W. Keefe.
1982. A comparative analysis of nutrients and
other factors influencing estuarine phytoplank-
ton production, pp. 69-90. In V.S. Kennedy
(ed.), Estuarine Comparisons. Academic Press,
New York.
Boynton, W.R., J.H. Garber, R. Summers, and
W.M. Kemp. 1995. Inputs, transformations,
and transport of nitrogen and phosphorus in
Chesapeake Bay and selected tributaries.
Estuaries 18:285-314.
Cooper, S.R. 1995. Chesapeake Bay watershed
historical land use: Impact on water quality and
diatom communities. Ecological Applications 5:
703-723.
Cooper, S.R. and G.S. Brush. 1991. Long-term
history of Chesapeake Bay anoxia. Science 254:
992-996.
Dennison, W.C., R.J. Orth, K.A. Moore, J.C.
Stevenson, V. Carter, S. Kollar, P.W. Bergstrom
and R.A. Batuik. 1993. Assessing water quality
with submersed aquatic vegetation. BioScience 43:
86-94.
Holland, A.F., A.T. Shaughnessy and H. Hiegel.
1987. Long-term variation in mesohaline
Chesapeake Bay macrobenthos: spatial and
temporal patterns. Estuaries 10: 227-245.
Jaworski, N.A. 1990. Retrospective of the water
quality issues of the upper Potomac estuary.
Aquatic Science 3:11—40.
Kemp, W.M., R.R. Twilley, J.C. Stevenson, W.R.
Boynton, and J.C. Means. 1983. The decline of
submerged vascular plants in upper Chesapeake
Bay: Summary of results concerning possible
causes. Marine Technology Society Journal'17:
78-89.
10
-------
Lee, K.P. 1993. Greed, scale mismatch, and
learning. Ecological'Applications-3: 560-564.
Magnien, R., D. Howard and S. Bieber (eds.). The
State of Chesapeake 1995. Environmental
Protection Agency, Annapolis, Maryland. 45 p.
Malone, T.C., W. Boynton, T. Horton, and C.
Stevenson. 1993. Nutrient loading to surface
waters: Chesapeake case study, pp. 8-38. In:
M.F. Uman (ed.J Keeping Pace with Science and
'Engineering. National Academy Press,
Washington, D.C.
National Research Council." 1994. Priorities for
Coastal Ecosystem Science. National Academy
Press, Washington, D.C. 92 p.
Nixon, S.W., C.A. Oviatt, J. Frithsen and B.
Sullivan. 1986. Nutrients and the productivity
of estuarine and coastal marine systems. Journal
of the Limnological Society of South Africa 12: 43—71.
Officer, C.B., R.B. Biggs, J.L. Taft, L.E. Cronin,
M.A. Tyler, and W. R. Boynton. .1984.
Chesapeake Bay anoxia: Origin, development
and significance. Science 223: 22-27.
Sea Grant College Programs of Maryland and
Virginia. 1992. Dissolved Oxygen in the Chesapeake
Bay: A Scientific Consensus. Maryland Sea Grant
College UM-SG-TS-92-03. 18 p.
Seliger, H.H., J.A. Boggs, and W.H, Biggley.
1985. Catastrophic anoxia in the Chesapeake
Bay in 1984. Science 228: 70-73.
Thomanh, R.V., J.R. Collier, A. Butt, E. Gasman,
and L.C. Linker. 1994. Response of the Chesapeake
Bay Water Quality Model to leading Scenarios. U.S.
Environmental Protection Agency, Annapolis,
Maryland.
-------
Gulf of Mexico Hyp6T
Conference Presenta||
Melissa Samet ;;:||||j||
Director, Marine Biodiversity Program •:;>!i||| i||
Sierra Oub Legal Defense Fund ... • :jj| j| |||
San Fronc/sco, California 94104 . ''^)Mm
Abstract
On January 24,1995, eighteen
environmental, social justice and
fishermen's organizations, represented
by the Sierra Club Legal Defense Fund,
petitioned EPA and the State of Louisiana to
convene an interstate management conference
of all the states contributing nonpoint source
pollution to the Mississippi River. That confer-
ence is seen as the best hope to develop and
implement enforceable controls to reduce non-
point pollution in the Mississippi River and
clean up the Dead Zone in the Gulf of Mexico.
While refusing to convene a formal interstate
management conference, EPA has convened
this conference to begin a strategic assessment
process in response to the petition.
The Gulfs Dead Zone poses an enormous
threat to the biological integrity and productivity
of the Gulf of Mexico, and exposes the precari-
ous ecological condition of the entire Missis-
sippi River. The Dead Zone is a wake-up call to
EPA and the states to take immediate and con-
certed action to control nonpoint pollution
entering the River and the Gulf. EPA and the
states must commit to and develop a long-term,
written and enforceable strategy to clean up the
Dead Zone. A successful strategy also will stem
the devastation of the Mississippi River eco-
system and improve each state's water and
environmental quality.
Introduction
I thought it would be useful to give you a brief
history of the Petition that prompted EPA and
the Gulf of Mexico Program to convene this
meeting, and to address some of our expecta-
tions for the process that this meeting is
starting.
The Impetus for This Conference
The Dead Zone is a 7,000 square mile swath of
Gulf of Mexico water so devoid of oxygen that
marine life cannot survive. The Dead Zone,
which appears in the Gulf in the summer
months, has grown substantially in size over the
past two years. It now stretches from the mouth
of the Mississippi River to the Texas border.
The existence of the Dead Zone, and its impli-
cations for the health of the entire Gulf region
and the Mississippi River watershed, prompted
concerned groups to ask the U.S. Environ-
mental Protection Agency (EPA) to take action
to clean it up.
On January 24, 1995, 18 environmental, social
justice and fishermen's organizations, repre-
sented by the Sierra Club Legal Defense Fund,
petitioned EPA and the State of Louisiana to
convene an interstate management conference
of all the states contributing nonpoint source
pollution to the Mississippi River. The petition-
12
-------
ers were not limited to groups from Louisiana
and Texas—the states suffering the most direct
impacts of the Dead Zone—but included
groups representing individuals from Minnesota
to Louisiana. That call for action has been
joined by the Gulf Restoration Network, a
coalition of over 30 local, regional and national
groups dedicated to protecting and restoring the
health of the Gulf of Mexico.
The petition was prompted by scientific
research showing that the Gulfs Dead Zone is
caused in large part by the nutrient loads enter-
ing the Gulf from the Mississippi River. I am
sure that many of you are aware of Dr. Rabalais'
research on the Dead Zone. While you will be
hearing from Dr. Rabalais directly at this meet-
ing, I would like to highlight one very significant
conclusion reached by her. At a Gulf of Mexico
Program Symposium held in March 1995, Dr.
Rabalais stated that the Dead Zone could not be
cleaned up without reducing nonpoint pollution
entering the Mississippi River, and ultimately
the Gulf, from the up-river states.
The beauty of the requested interstate manage-
ment conference is that it would bring together
those very states identified by Dr. Rabalais—all
the states within the Mississippi River water-
shed—-that have the actual authority to control
the nutrient runoff causing the problem. More-
over, those states would come together for just
one purpose: to create real in-the-water reduc-
tions of nutrient loading into the Mississippi.
EPA refused to convene the interstate manage-
ment conference. Instead it elected to attempt
to address the Dead Zone through the Gulf of
Mexico Program, and other programs already
addressing nonpoint source pollution. The State
of Louisiana also informed the Petitioners that
it would like to use mechanisms already in place
to address the Dead Zone problem, despite the
fact that in June of this year, the Louisiana
Legislature passed a resolution calling for our
requested interstate management conference.
Louisiana officials have said, however, that the
State is prepared to request an interstate
management conference if those mechanisms
prove insufficient to address the problem.
This meeting begins EPA's attempts to respond
to our petition without utilizing an interstate
management conference.
i f < - *
»*-,;-
The Dead Zone is a Wake up Call for
Action x " , -,
< ' / '
. ~°~ *vv ^ / w= ~f ,
The Dead Zone must be seen as a wake up call
for immediate action to begin the clean up
process. And that alarm must be heeded.
The magnitude of the Dead Zone problem
cannot be overstated. When last measured, the
Dead Zone covered more than 7,000 square
miles—ran area larger than the states of Con-
necticut and Rhode Island combined. Over the
past few years, the Dead Zone has more than
doubled in size. Indeed, it is now larger than
many bodies of water in EPA's watershed
protection program.
The Dead Zone poses a serious threat to the
biological integrity and productivity of the Gulf
of Mexico. Its impact is akin to taking Saran
Wrap and placing it over an area the size of
Connecticut and Rhode Island, slowly pulling it
down and suffocating everything that cannot
escape out the sides. While the area appears to
undergo recolonization beginning each fall
when the Dead Zone dissipates, the long term
implications of a yearly die off remain unclear.
By causing such devastation, the Dead Zone
also poses a very real threat to the economy of
the Gulf region. Already, officials at one sea-
food processing plant that closed down in
13
-------
Louisiana, blamed the closure in part on the
Dead Zone. As a result of that one plant
closure, Louisiana lost 176 jobs. Forty six jobs
were lost altogether, and 130 others were
relocated to Texas.
Because the Dead Zone is caused by excess
nutrients entering the Mississippi River, and
ultimately the Gulf, it is a manifestation of land
use practices throughout the entire Mississippi
River watershed. As such, the Dead Zone
exposes the precarious ecological condition of
that enure watershed, and should raise alarm
bells in each watershed state. We are not alone
in this analysis. Many biologists with the Upper
Mississippi River Conservation Committee
believe that a sudden collapse of the Upper
Mississippi River System "is likely to occur."
Upper Mississippi River Conservation Commit-
tee, Fating the Threat: .An Ecosystem Management
Strategy for the Upper Mississippi River (P&c. 1995)
at 8. Indeed, the Committee is meeting this
week to develop an ecosystem-wide protection
strategy.
Efforts to clean up the Dead Zone will of
necessity help stem the devastation of the Missi-
ssippi River ecosystem. Those efforts will
improve the environmental and water quality in
all the states in the Mississippi River watershed.
Decisive Action and StrongLeadership
Are Needed to Clean upthe Dead Zone
The Petitioners, the Gulf Restoration Network
and the Sierra Club Legal Defense Fund fully
recognize that cleaning up the Dead Zone will
not be an easy task. We also understand the
importance of basing policy decisions to control
nutrient enrichment on sound science. As such,
we applaud the efforts of the Gulf of Mexico
Program in convening this meeting. It is an
important first step.
However, we would not have filed the Petition
requesting an interstate management confer-
ence, if the existing science did not already make
clear that actions must be taken to clean up the
Gulf, and that those actions must begin now.
While there may be a need to fill in data gaps,
that need cannot be used as an open-ended
excuse for not taking action. Additional studies
will not make the Dead Zone go away. Only
appropriate controls will accomplish that task.
Existing scientific knowledge shows that
controls to reduce nonpoint pollution entering
the Mississippi River must be implemented
quickly. It also shows where at least some of
those controls should be. Methods for reducing
nitrogen loading (the primary culprit in the
Dead Zone) are well recognized and have been
implemented successfully in many places. Thus,
site specific controls could be implemented
immediately in areas of direct nitrogen applica-
tion and runoff. All that is missing is the
appropriate leadership and political will.
Additional innovative control measures also
have been suggested. These include reestablish-
ing a natural vegetative corridor along the main
stem of the Mississippi River. This would help
reduce nitrogen (and other) runoff, and would
have the added benefit of returning some of the
natural processes of this great floodplain river.
This also is an action that could be funded by
EPA as a best management practice.
If we are to have any hope of succeeding in
cleaning up the Dead Zone, some basic ground
rules must be in place:
1. The appropriate parties must be at the
table to develop—and then
implement—viable controls. It is
estimated that 80 percent of nutrients
are in the Mississippi by the time it
14
-------
passes Cairo, Illinois, and the vast
majority of nitrogen entering the system
is coming from the up-River states.
Unfortunately, the Gulf of Mexico
Program does not have the authority or
the mandate to pull those states into the
process being started by this meeting.
Thus, it will be up to EPA to show
strong leadership and bring into the
process all the states in the Mississippi
River watershed.
2. An aggressive clean up strategy cannot
wait until all scientific data gaps are
filled. We must begin immediately to
develop an aggressive timetable for
action. It is essential that we quickly
develop a written strategy that prescribes
specific solutions to be implemented
within a set time frame. The strategy also
must set a realistic timetable for this
process to show concrete results. One
such concrete measurement would be a
commitment by the up-River states to
reduce their proportionate share of
nutrient loading in the River.
3. EPA must show strong leadership, and
provide a long term commitment of
resources if we are to have any hope of
seeing real in-the-water improvements in
the Gulf.
The Petitioners, the Gulf Restoration Network,
and the Sierra Club Legal Defense Fund have
that long term commitment to solving the
problem before us, and will do everything
necessary to ensure that the process I just
outlined is implemented, and continues, until
the Dead Zone is cleaned up.
Presentation Discussion
Melissa Samet (Sierra Club Legal Defense Fund—San
Francisco, CA.)
Eugene Buglewicz (Corps of Engineers—Vicksburg,
MS) asked Melissa Samet to better define the
Coalition's expectation of terms "clean up" and
"get results."
Melissa
-------
Melissa Samet suggested that private land-
owners would have to change their practices in
order to achieve consistent nonpoint source
pollution reduction, and that it was in their own
best interests to do so.
Ron Kucera (hiissouri Department of Natural
Resources—Jefferson City, MO) raised two issues.
« The State of Missouri has implemented a
self-imposed sales tax for nonpoint
source controls through a Soil and
Water Conservation program.
* The Sierra Club often supports positions
that are counterproductive to solving
water resource issues in the Missouri
River Basin. For example, the MNI-
SOSE Intertribal Water Coalition, Inc., a
tribal corporation interested in water
marketing, is claiming 20 percent of the
Mississippi River flow above Cairo,
Illinois, and 40 percent of the Missouri
River flow above St. Louis, Missouri.
The MNI-SOSE Coalition asserts that
they should be able to market those
water rights outside of the Missouri
River Basin. Since the Department of
the Interior is supporting the initiative to
achieve the best use of the resources,
and since marketing those rights in the
upper river areas would not reflect the
interests of the Gulf Coast or the State
of Missouri, he asked the Sierra Club to
discuss potential reduction in available
freshwater quantity with him.
Melissa Samet agreed to discuss the issue with
him during the conference, but pointed out that
the Sierra Club Legal Defense Fund was a
separate entity from the Sierra Club.
16
-------
EPA CommittellJlp
Robert H. Wayland
Director, Office of Wetlands,
U. S. Environmental Protection
Washington, D.C 20460'^rljlp^**
Abstract _ *-; ' - y ^ 'f _
This is an important conference and the
"Dead Zone" in the Gulf of Mexico is an
important issue. We're going to be com-
mitted to addressing it from the national level at
EPA. There are a number of efforts underway at
the national level that can play a role in clarifying
the severity of the problem, developing strategies
to counter it, and implementing those strategies.
Just to cite a couple of examples in areas where I
have been directly involved recently: Yesterday I
was kicking off a two-day National Nutrient
Assessment Workshop that EPA is sponsoring.
We have convened 22 experts from EPA,
17 academic experts, 10 state officials with
particular expertise in nutrient management,
7 people from the consulting community, and
3 experts from local and interjurisdictional
governments to help formulate a better tool box
for addressing nutrient problems nationally. Last
month I addressed the Inter-governmental
Conference to Adopt a Global Program of Action
for the Protection of the Marine Environmental
from Land-Based Activities. More than a hundred
countries participated in this meeting which grew
out of the UNEP-Rio Conference in which there
was an international commitment to better
address marine resources and the impacts to them
from land-based sources.
EPA Funding Will Affect Our Ability tp
„ ~ jL.*. ..-„-», ,: . *S •'
Address Hypoxia
' "" ->/
At the same time, however, our ability to deal
with this and other environmental problems is
closely related to the availability of resources with
which to work. The President has stated that he
will veto the Appropriations bill for EPA because
the Senate and House have cut funding below the
Administration's request and below the FY 1995
level. This is the first trip I've made by airplane
paid for by EPA since the beginning of the fiscal
year on October 1. I'm also gratified to say that
the Section 319 Program to address run-off has
been fully funded by both the House and the
Senate at the $100 million level the Administra-
tion requested.
The-ProBlem arid Its Causes
First let me talk about the problem and its
apparent causes: Hypoxia and other effects of
nutrient over-enrichment are not just limited to
the Gulf of Mexico, or even to our coastal waters.
Nutrient over-enrichment is a pervasive problem
which reduces the quality and productivity of the
Nation's1 Waters. It has been the primary focus of
efforts to restore the productivity of the
Chesapeake Bay. There has been a dead zone for
17
-------
many years in Long Island Sound—not caused by
toxic chemicals but by lack of oxygen.
My first illustration (Figure 3)shows that indeed
nutrients are a problem upstream in the
Mississippi-Missouri Watershed as well as
downstream in the Gulf of Mexico. The states
shown in green are those states that, in their 1994
Water Quality Assessment Reports to EPA,
identified nutrient enrichment as the primary
cause of water quality impairment in their waters.
River and streams have a natural flushing action
which often makes the effects of nutrient
enrichment less apparent and does, in fact,
transport the nitrogen and phosphorus in
fertilizers, animal waste, and domestic sewage to
downstream areas, either lakes or estuaries, where
there is limited or no flushing action for it to sink.
Cross-hatched states are states which identify
nutrient enrichment as the primary cause of
impairments for their lakes and reservoirs.
Thus, control of nutrients in the upper watershed
of the Mississippi and Missouri Basins will
potentially benefit not only the Gulf of Mexico
and downstream states, but also have at-home
benefits for many of those states further up in the
watershed. Although states shown here in blue
did not list nutrient enrichment as a primary cause
of impairment, it may be a primary cause in some
waters or a secondary cause in many waters. In
the aggregate, nutrients are the leading source of
water quality impairment in the United States.
Sources and Distribution of Nutrient
Discharges
Let me turn to the amounts and distribution of
the activities that contribute to nutrient loadings.
There are about 11.5 million tons of nitrogen
fertilizer currently applied to croplands, 6.5 mil-
lion tons of manure generated by 11 billion farm
animals, 3.2 million tons of nitrogen entering our
waterways as a result of atmospheric emissions.
About .8 million tons are discharged from public
owned wastewater treatment works.
Figure 4 depicts potash fertilizer use in tons per
square mile, on a county-wide basis. The heaviest
use rates are shown as yellows and reds. (The
greatest number of these is in the upper Missis-
sippi basin.) Information on the prevalence of
livestock which, of course, correlates with
manure, shows a highly similar picture (Figure 5).
Figure 6 shows nitrogen fertilizer use in 1991 on a
county basis.
"Nutrient Reduction Experience
EPA and partner states have substantial
experience in developing strategies to address
nutrient over enrichment in coastal areas. Of the
28 estuaries enrolled in the National Estuary
Program, Galveston Bay, Tampa Bay, Sarasota
Bay, Corpus Christi, and Barataria-Terrebonne are
all on the Gulf Coast. All of those estuaries, and
in fact all of the estuaries in the National Estuary
Program, identify nutrient enrichment as a
primary environmental problem that they want to
deal with.
Of course, the consequences of hypoxia aren't
generally public health problem, rallying public
concern and public interest, is more of a challenge
than if we were confronting a drinking water
contaminant or a toxic cloud. Ground water
contamination by elevated levels of nitrate is a
public health concern in some instances, however.
18
-------
For the most part, we are talking about both an
economic problem—lost economic returns in
terms of fisheries productivity and catch, and
aesthetic problems in terms of lakes, estuaries,
and other water bodies which are unable to fully
support recreation due to algae blooms and other
problems.
While the relative contribution from point
sources and run-off varies from watershed to
watershed, nationally, only about 6 percent of the
nitrogen loadings come from point-source
discharges. It is generally possible to remove a
pound of nitrogen from non-point sources in a
far more cost effective manner than is true for
removing that same pound of nitrogen from a
point source.
I think some of the most encouraging news I can
share with you is that there are simple, practical
and affordable control measures for reducing
nutrient :run-off from non-point sources. A
number of these take the form of prevention
approaches, meaning that the environmental
benefit is realized by reducing use of the potential
pollutant, in this case fertilizer, without reducing
the benefit to food production or the producer.
I'm sure John Burt will talk in a few moments and
perhaps later in the Conference about some of
the progress being made in agronomic practices.
Just a few factoids I've collected are:
States reporting nutrients as a
leading cause of impairments to
lakes, ponds, and reservoirs
States reporting nutrients as a
leading cause of impairments to
rivers and streams
Figure 3.
19
-------
POTASH USE,
IN TONS PER
SQUARE MILE
i ; No dala or 0
j 0.01 -0.31
[0.32-1.05
11.06-3.90
3,91-11.44
More than 11.44
Coverage
name: nlt91
Attribute
name: K2O91.USE
Data from agricultural chemical
use studies of W.A. Battaglin
and D.A. Goolsby
Figure 4.
35
NUMBER OF
CATTLE AND
CALVES, 1987
No data or 0
-9,027
i 9.028-21,292
I 21.293-42,680
142,861-115,820
More than 115,820
Coverage
name: ag_stock
Attribute
name: CA0930
Data from Agricultural Chemical
Use studies of W.A, Battaglin and
D. A. Goolsby
Figure 5.
20
-------
45,
NITROGEN -
FERTILIZER USE, IN
TONS PER SQUARE MILE
No data or 0
j 0.01-1.00
1.01-3.32
; 3.33-8.08
I 8.09 - 20,22
More than 20.22
Coverage
name: nit91
Attribute
name: NTOT91.USE
D«ti from « jrleuttuta! chemical
u» studlw oT W-A. Battaglln and
D.A.Gooltby
Figure 6.
In the Big Spring Basin Demonstration Project
where they've been working with new farming
techniques since 1981, farmers have reduced their
nitrogen fertilizer use by 34 percent from about
115 Ibs/acre in 1993 compared to 174 Ibs/acre in
1981, realizing a cost savings of $360,000 or about
$1800 per producer with" overall improvements in
yield.
Participating farmers in Maryland's state-wide
nutrient management program have achieved an
average reduction of 35 Ibs. of nitrogen per acre
and 41 Ibs of phosphates per acre; in Nebraska,
nitrogen applications to com have been reduced
by 30 Ibs./acre with no decrease in yield and at a
cost savings of about $900,000 annually for
participating farmers.
Of course, there are other controls, as opposed to
prevention methods, like the use of buffer strips
to physically and biologically intercept run-off.
There are also mitigation strategies, like the restora-
tion of wetlands, which can sequester some of
the nutrients, before they reach lakes or estuaries
or the Gulf. The work we are doing nationally to
demonstrate and evaluate a variety of control
methods or prevention methods and their
cost-effectiveness will be highly beneficial to the
work that's undertaken to address problems in
the Gulf. Strategies to reduce nutrient loadings to
the Gulf are also likely to return significant water
quality improvements and benefits for those who
live in upstream states.
I want to quickly mention a couple of national
!
trends that I think are going to affect the nutrient
21
-------
management picture and therefore, affect what's
taking place with this Conference and its
follow-up work. The first is the changing
paradigm for addressing water quality problems in
the nation. We're increasingly moving, along with
states and our federal colleagues, away from a
source-by-source or a pollutant-by-pollutant
approach to a whole watershed approach Don
Boesch illustrated in his discussion of the
Chesapeake Bay experience. I think the Bay
experience, along with our experience in the Clean
Lakes Program and National Estuary Program,
demonstrates that watershed management is the
approach we need to use to engage the
stakeholders who can be a part of the problem
and can become a part of the solution. The Clean
Water Act re-authorization debate has been
underway for some time now. I don't think we're
expecting significant changes in federal law in this
Congress. Not withstanding that, we've moved
forward with the states to realign programs on a
watershed basis.
Louisiana has been a participant in a related effort
of redesigning and revitalizing the National
Non-Point Source Program. I alluded earlier to
the fact that EPA's Clean Water Act Section 319
grants program is fully funded at the level
requested by the Agency and equal to last
ycar'slcvel. We're also very close to articulating a
new policy for pollution trading. This market-
oriented mechanism will encourage point sources
to meet their pollution reduction requirements by
financing or undertaking non-point source
control practices which may save them significant
resources.
All of the Gulf states, along with 20 or so other
states, have prepared and submitted to EPA,
coastal non-point source control programs
developed as a result of legislation enacted in
1990. We're expecting to see some significant
water quality improvements as a result of these
programs. We are evaluating the State programs
now to determine which can be approved, which
would be conditionally approved, and which
disapproved.
I look forward to the expert presentations which
have been arranged for the balance of this
Conference and outlining, at the conclusion, some
thoughts on how we proceed from here.
,,,.. ,.,- f „ s
Presentation Discussion
Robert Wayland (U.S. EPA, Washington, D.C.)
Don Boesch (University of Maryland,
Cambridge, MD) commented that to gain local
support, it is imperative to communicate the local
water quality benefits as a result of solving the
nonpoint problem instead of solely focusing on
the benefit to the hypoxia area. For example,
Pennsylvania began participating in the Chesa-
peake Bay Program because it benefitted their
local water quality.
22
-------
: , " -'^*::='-j:s::-::;--^
Hypoxia in the
* -_::--: _ ""J^^TS-
William A. Kucharski
Secretory ';. - : sr^^^^f
-- / • .-...--. f ,••..! \S?5ph:=i::;:L:;:;=~£
Baton Rouge, Louisiana 70884
Abstract ^ "';_. -1. 5 _A
The Mississippi River basin drains over
40 percent of the land area of the United
States. The Gulf of Mexico, into which the
Mississippi flows, also supports approximately 40
percent of the fishery landings for our country. An
area of low oxygen concentration, termed a
hypoxia region, has been documented in the Gulf.
This conference is being held to present and
review the technical realities of the problem and to
bring together all interested parties for discussions
and education.
This presentation will remind the participants that
we actually know very little about a potentially big
problem. Further, those states that believe that
they have no responsibilities toward the Gulf
because they are far away, are misinformed. The
problems that manifest themselves off the coast
of Louisiana are national problems that will
require study and change in many upriver states.
The forum under which this study and resultant
work will be done is equally important to the
future of the environmental protection business.
The non-enforcement, non-threatening aspects of
this study/program are being looked upon as a
measure of the environmental maturity in this
country. Those that believe the change can occur
without threats must be active in this program.
The technical issues related to the depleted oxygen
area in the Gulf of Mexico have recently been
catalogued to present a more detailed explanation of
this phenomenon. Recommendations have
followed which express the need for more detailed
studies about cause and effect and the development
of appropriate control measures. Since this seasonal
problem, however big and however serious it finally
turns out to be, exists off of the coast of Louisiana,
people might assume that this is Louisiana's
problem alone. Nothing could be further from the
truth. I will attempt to outline in general detail, why
the problem is of national concern rather than
simply a single state issue. I will also provide
support of a multi-state, federal and citizen
partnership as the tool for solving, or at least
mitigating this problem.
The problem of having a very large area of low
oxygen in a normally productive area of the Gulf
of Mexico has both environmental and economic
implications. What we do not know about this
situation is greater than what we do know
however. Speaking from a policy perspective, it is
sometimes very difficult not to 'do something
before all of the facts are known. We, as policy
managers, are often subjected to varying amounts
of outside pressure to take steps that are designed
to solve simple problems. The fact that most of
the issues we have to deal with are not simple, but
rather, are quite complex does not seem to deter
the "you must do something now" voices. That is
one of the issues that this conference must
address over these next two days. How do we
design short term, intermediate and long term
remedies of the problem? First we must be able to
describe what the problem actually is. We can
describe the manifestations of the hypoxia
23
-------
problem. In fact, the name adequately describes
the impact. Low oxygenated waters. This does not
however, describe the cause or causes of the
hypoxic area. We are all certain that nutrient
loadings are one of the primary contributors to
the hypoxia problem, but we also know that some
level of nutrient loading is necessary to sustain the
productivity of the Gulf ecosystem. What we do
not know is how much nutrient loading is enough
or how much is too much. We do not know when
we reached the point of over supply of these
nutrients nor are we certain of how long the
hypoxic region has existed. In short, we know
there is a problem, but beyond that we are
guessing about the relative contributions of each
factor. I make these observations as a political
appointee, not as a biologist or a fisheries expert.
What are the implications of this problem? We
often speak of sustainable development and using
our natural resources wisely. Killing a large area of
the Gulf ever)' year can not be considered
responsible stewardship by even the most
indifferent polluter. The question is not whether
we must address this issue, it is how we must
address it. As the Secretary of the Louisiana
Department of Environmental Quality, I am
obligated by state statutes to protect human health
and the envkonment in my state. How do I
protect the environment of my state when the
problem is not created in or by my state? How do
we as states, who by the way are demanding that
the federal government get off of our backs and
let us regulate our states, handle interstate
transport problems? These are the questions that
this conference must also address.
From a single state regulator's perspective this
problem can be viewed as a measure of our
nation's environmental protection system maturity.
Our challenge is how to contact, work with, and
when necessary, modify existing practices carried
out in multiple states. The question is whether we
can accomplish such a task in a voluntary or
prescriptive manner. I do not know the answer to
this question. I do know that if we states continue
to demand autonomy, or the U.S. Congress in its
wisdom gives us that autonomy and we fail to use
this freedom of action wisely, we will have failed in
our charge to effect positive change in the current
command and control system we say we dislike.
The problems we face in the Gulf are real. The
solutions that we may have to implement will
require a level of communication, study and action
that we have heretofor only thought about. Can we
cooperate when a state may have to change how its
citizens work, farm or handle waste, and not see
any appreciable environmental improvement in
that state? This is the real question that we all
must face. This will be hard, this will create
political challenges to all of the regulatory agencies
involved. But let us return to the basic issue. Some-
thing is wrong. We have to find out just what we
have to fix and then we have to fix it. This will take
time, lots of money and a positive attitude that the
"fixes" we propose will work. This brings to mind
the famous quote from the American revolution,
"Gentlemen, we will all hang together, or most
assuredly we will hang separately." This is what we
face. A multi-state cause, a local effect and a
national impact. How we as states handle this issue
may frame how environmental controls and
regulatory compliance are managed into the
twenty-first century. We must, I believe, face this
reality of our plight with open eyes, with a willing-
ness to change and a clear recognition that some
very tough decisions will most likely be part of any
control solution we impose.
^resentaUoj\DisciissJQn *
William Kucharksi—Louisiana Department of
Environmental Quality, Baton Rouge, LA
No questions/discussion after Secretary
Kucharksi.
24
-------
Abstract
The inner to mid-continental shelf from
the Mississippi River westward to the
upper Texas coast, is the site of the
largest zone of hypoxic bottom water in the
western Atlantic Ocean. The areal extent during
mid-summer 1993-1995 (16,800 km2, 16,300
km , and 18,200 km2, respectively, of near
bottom waters < 2 mg/1) rivals the largest
hypoxic areas elsewhere in the world. Spatial
and temporal variability in the distribution of
hypoxia is, at least partially, related to the
amplitude and phasing of the Mississippi River
discharge. Freshwater fluxes dictate, along with
climate, a strong seasonal pycnocline. Nutrients
delivered by the Mississippi and Atchafalaya
support high primary production, of which
approximately 50 percent fluxes to the bottom.
The high particulate organic carbon flux fuels
hypoxia in the bottom waters below the
seasonal pycnocline. Significant increases in
riverine nutrient concentrations and loadings of
nitrate and phosphorus and decreases in silicate
have occurred this century, and accelerated
since 1950. As a result of the nutrient
alterations, the overall productivity of the
ecosystem appears to have increased since the
1950's along with an increase in oxygen
deficiency stress this century. Variable changes
in nutrients and/or changes in freshwater fluxes
will result in differing scenarios for distribution
of hypoxia in the future.
Hypoxia is operationally defined as dissolved
oxygen levels below 2 mg I'1, or ppm, for the
northern Gulf of Mexico, because that is the level
below which trawlers usually do not capture any
shrimp or demersal fish in their nets (Leming and
Stuntz,1984; Renaud, 1986). In this presentation,
we outline the distribution and dynamics of
hypoxia in the northern Gulf of Mexico, including
present and historical conditions. We further
detail historical conditions evident in the sedi-
mentary record and use these retrospective
analyses to predict future scenarios, including that
of climate change. A more complete review of the
subject is provided in Rabalais et al. (in press and
in review).
---?•*
% Present Distribution \ ,
The inner to mid continental shelf of the
northern Gulf of Mexico, from the Mississippi
River birdfoot delta westward to the upper
25
-------
Texas coast, is the site of the largest zone of
hypoxic bottom water in the western Atlantic
Ocean. The areal extent of this zone during
mid-summer surveys of 1993—1995 (approx.
16,000 km2 to 18,000 km2; Rabalais et al., in
review; Figure 7) rivals the largest hypoxic areas
elsewhere in the world's coastal waters, namely
die Baltic Sea and the northwestern shelf of the
Black Sea.
Conditions during the-Great Summer Flood of
1993 point to the importance of the river in the
formation and persistence of hypoxia (see
references in Dowgiallo, 1994). As a result of
higher streamflow, especially in mid to late
summer, there were:
* Lower than normal surface salinities
• Higher surface temperatures
• Increased stability in the coastal waters
* Increased overall loading of nutrients
• An order-of-magnitude higher than normal
total phytoplankton counts
• A predicted greater flux of carbon to the
seabed
* A significantly lower oxygen content of the
lower water column
* An approximately two-fold increase in the
areal extent of hypoxia with respect to the
1985—1992 mid summer averages, over an
extensive area (Rabalais et al., 1994a).
^Previous Years
Prior to 1993, the average areal extent of bottom
water hypoxia in mid-summer was 8,000 to 9,000
km2 (Rabalais et al., 1991). Distribution maps of
mid-summer bottom water hypoxia since 1985
often show disjunct areas of low oxygen down-
field of each of the river deltas (see 1992 in
Figure 7). Other distributions are continuous
from the Mississippi River delta to the upper
Texas coast. When the 2 mg I'1 isopleths are not
continuous along the shelf, however, the areas
between are still undersaturated in oxygen with
values usually below 4 mg I'1 and mostly below
3mgK
f ~ «~ ^ -°/° "'"•*/^
«"- • " ,' i
^Annual Cycle
Critically depressed dissolved oxygen concen-
trations occur below the pycnocline (Figure 8)
from as early as late February through early
October and nearly continuously from mid May
through mid September. The importance of
stratification and the physical structure of the
water column in defining the distribution of
hypoxia was discussed by Wiseman et al. (this
volume). Hypoxic waters are distributed from
shallow depths near shore (4 to 5 m) to as deep as
60-m water depth. The more typical depth
distribution of hypoxic bottom waters, however,
is between 5 and 30 m.
In March, April and May, hypoxia tends to be
patchy and ephemeral; it is most widespread,
persistent, and severe in June, July and August
(Figure 9). The persistence of extensive and
severe hypoxia into September and October
depends primarily on the breakdown of the
stratification structure by winds from either
tropical storm activity or passage of cold fronts.
Hypoxia is not just a bottom-hugging lens of
water. It occurs well up into the water column;
depending on depth of the water column, hypoxia
may encompass from 10 percent to over
80 percent of the total water column.
Continuous time series show long periods of
hypoxia and anoxia, a draw-down of hypoxia in
the spring in response to respiration in the lower
water column and at the seabed and sediment
oxygen demand, vertical mixing and loss of strati-
26
-------
fication, response to winds (e.g., upwelling of
deeper oxygenated waters), and, in other parts of
the shelf, the influence of tidal advection
(Rabalais et al., 1992; Rabalais et al., 1994b).
While the 1995 areal extent of bottom water
hypoxia was the largest ever recorded for the
Louisiana shelf, its permanence over such an
extent is not known. Based on monthly
monitoring transects off Terrebonne Bay from
5-m to 30-m water depth and a continuously
recording oxygen meter in 20-m water depth, the
following can be said about the 1995 hypoxia
season: Extensive low oxygen occurred as early
as late May. During June and July oxygen levels
were extremely low over large areas. Tropical
storm activity re-aerated the water column during
August, but low oxygen conditions again
developed and low and extremely low values
persisted into late September.
/ \ ^ /
Proximal Causes
The relative magnitude in changes of freshwater
discharge and nutrient flux from the Mississippi
and Atchafalaya Rivers to the coastal ocean
affects water column stability, surface water
productivity, carbon flux, and oxygen cycling in
the northern Gulf of Mexico.
High biological productivity in the immediate
(320 gC m-2 yr1) and extended plume (290 gC
nr2 yr1) of the Mississippi River is mediated by
high nutrient inputs and regeneration,
temperature and favorable light conditions (Sklar
and Turner, 1981; Lohrenz eral., 1990). Spatial
variation in primary production in a given period
is related to salinity and the associated
environmental and biological gradients (Lohrenz
et al., 1990, 1994). There is a worldwide
relationship between flux of dissolved inorganic
nitrogen and primary production in coastal
waters, and a similar relationship has been
observed for the plume of the Mississippi River
(Lohrenz et al., in review). The availability of
dissolved silicate and its ratio to total inorganic
nitrogen are also important in controlling the
extent of diatom production and the composition
of the diatom community with implications to
carbon flux and control of oxygen depletion
(Dortch and Whitledge, 1992; Nelson and
Dortch' in press).
Particulate organic carbon flux to the lower water
column is high in the extended plume over the
inner shelf (Qureshi, 1995). The fraction of
production exported is highly variable but
averages about half of the estimated integrated
primary productivity with statistically higher
fluxes in the spring. A large proportion of the
paniculate organic carbon flux reaches the
bottom incorporated in zooplankton fecal pellets
(55 percent), but also as individual cells or in cell
aggregates. Although Qureshi's data are limited to
a single station in 20-m water depth off Terre-
bonne Bay for a 2-yr period, it appears that the
amount of carbon fluxed is greater when the
spring freshet of the river is higher. Also, the
carbon ifluxed via fecal pellets (during period of
high flux) is sufficient to deplete the bottom water
oxygen reserve in spring, thus creating hypoxic
conditions that then prevail through the stratified
summer period. When fluxes are lower (e.g., in
lower flow years, or in summer) the oxygen
depletion rates for these fluxes are close to the
calculated oxygen depletion rate.
On a seasonal time scale, productivity is most
influenced by Mississippi River flow and nutrient
flux to the system. Long-term mean seasonal
variations in net productivity at a station in 20-m
water depth off Terrebonne Bay are coherent
with the dynamics of freshwater discharge (fustic
et al., 1993). The surface layer shows an oxygen
surplus i during February-July, the maximum
occurs in April and May and coincides with the
maximum flow of the Mississippi River. The
bottom layer exhibits an oxygen deficit through
the year, but reaches its highest value in July
(coincident with maximum pycnocline strength).
The correlation between the Mississippi River
27
-------
flow and surface oxygen surplus peaks at a time-
lag of one month, and the strongest correlation
for bottom water oxygen deficit is for a time-lag
of two months. The oxygen surplus is also a good
indicator of excess organic matter derived from
primary production which can be redistributed
within the system [which follows Qureshi's (1995)
results).
A similar 2-month lag of bottom water hypoxia
following peak Atchafalaya River flow was
observed by Pokryfki and Randall (1987) for the
southwestern Louisiana shelf for data from the
early 1980's. Low surface salinities lagged one
month behind peak river flow. Their model did
not incorporate any biological processes, which
with additional lags would increase the accuracy
of their predicted low oxygen periods. The phys-
ics, geological setting and important biological
parameters, such as light fields-and nutrient flux,
differ on the southwestern Louisiana shelf from
that of the southeastern shelf. It is also not clear
how the effluents of the Mississippi River and the
Atchafalaya River merge to produce the physical
structure of the area. Wiseman and Kelly (1994)
demonstrated that salinity signals from both river
discharges were detectable off the Calcasieu
estuary. However, similar biological processes
occur on the southwestern shelf so that a large
area of hypoxic bottom waters to the west of the
Atchafalaya River appears to form each season.
Historical Data
1
Our systematic surveys began in 1985.
Extensive shelf-wide distributions of hypoxia
occurred each summer of those years, with the
exception of 1988 (a record low flow year for
mid-summer; hypoxia developed as usual in the
spring, but not maintained in the summer).
Prior to 1985, the data are mostly ancillary to
other studies, thus do not form a complete
survey, either temporally or spatially. There
were some directed studies in the early 1970's
(Ragan et al., 1978; Turner and Allen, 1982).
The mention of low oxygen conditions from the
mid 1930's in the Conseil Permanent
International pour FExploration de la Mer
Bulletin Hydrographique for 1935 were
identified in Hedgpeth's Treatise on Marine
Ecology and Paleoecology (Brongersma-
Sanders, 1957; Richards, 1957) as records from
the oxygen minimum zone of deeper waters
(e.g., 400—500 m deep) and several authors have
shown that there is no continuation of the
oxygen minimum zone with the hypoxia on the
inner to mid continental shelf (Pokryfki and
Randall, 1987; Rabalais et al., 1991).
Hypoxia was first recorded in the early 1970's
off Barataria and Terrebonne/Timbalier Bays as
part of environmental assessments of oil
production and transportation development
studies (the Offshore Ecology Investigation,
OEI, and the Louisiana Offshore Oil Port,
LOOP). Ragan et al. (1978) and Turner and
Allen (1982) followed up with studies in 1975
and 1976 along the Louisiana coast and
documented low oxygen conditions over most
of the areas they studied in the warmer months.
Environmental assessments for brine disposal
areas and further studies of oil and gas
production areas revealed low oxygen
conditions in most inner shelf areas studied in
mid-summer.
Hypoxia along the upper Texas coast is usually
an extension of the larger hypoxic zone off the
Louisiana coast, although isolated areas may be
found (e.g., Big Hill area and Bryan Mound
areas, but may be an artifact of the sampling).
Most instances of hypoxia along the Texas coast
are infrequent, short-lived, and limited in extent.
There are no records of hypoxia below the
Freeport, Texas area (with the exception of one
record at SEADOCK off the Brazos River)
28
-------
(Rabalais, 1992). There are very few systematic
surveys for this area.
There are reports of hypoxia off Mississippi
Sound during high stages of the Mississippi
River; also reports off Mobile Bay in
bathymetric low areas. There are usually more
reports in flood years (especially 1993, related to
the high flow of the river late in summer and
movement of the waters to the east of the delta
by the persistent southerly and southwesterly
winds).
Prior to the 1970's, there is some anecdotal
information from shrimp trawlers in the
1950's-1960's of low or no catches, of "dead"
or "red" water, but no systematic analysis of
these records.
s " " *'
Changes in Nutrient Loadings ;'*
These results are oudined in papers by Turner
and Rabalais (1991, 1994a,b), Justic' et al. (1994,
1995a,b) and Turner et al. (this volume):
• Nutrient concentrations and loadings have
changed dramatically this century and
accelerated since the 1950s.
• Concentrations of dissolved N and P have
doubled, and Si have decreased by 50 percent.
• Nutrient composition in river and adjacent
Gulf waters has shifted towards ratios closer to
the Redfield ratio and more balanced than
previously.
• These changes are closely related to N and
P fertilizer applications in the watershed.
• Offshore nutrient compositions shifted along
with potential and probable nutrient
limitations.
• Water quality changes are specific to changes
in nutrients. Freshwater inflow has remained
fairly stable, although there is a slight increase
in total flow in the last two decades due to an
increase in Atchafalaya River flow (Bratkovich
et al., 1994).
Ecosystem Changes ^
Long-term changes in the severity and extent of
hypoxia cannot be assessed directly, because
systematic sampling of bottom water dissolved
oxygen concentrations did not begin until 1985.
Therefore, biological, mineral or chemical
indicators of eutrophication and/or hypoxia
preserved in sediments, where accumulation rates
record historical changes, provide clues to prior
hydrographic and biological conditions. Similar
analyses have proven useful in the Great Lakes
and Chesapeake Bay and were done for the
Mississippi River bight.
An analysis of long-term data sets and diatom,
foraminifera, and carbon accumulation in
sediments supports the inference of increased
eutrophication and hypoxia in the Mississippi
River delta bight primarily because of changes in
nitrogen loadings. These results are outlined in
Rabalais et al. (in press).
The work of Eadie et al. (1994) demonstrated
from two cores in the Mississippi River delta
bight an increased accumulation of marine-origin
carbon in the last 100 years, consistent with
changes' in productivity beginning in the mid-
1950's when benthic foraminiferans rapidly
became isotopically lighter. Beginning in the mid-
1960's, the accumulation of organic matter,
organic d13C and d15N showed large changes in a
direction consistent with increased
productivity. The latter period coincided with a
doubling of the load of nitrates in the
Mississippi River outflow which leveled off in
the 1980's. Increased carbon accumulation was
also calculated from BSi (a surrogate for diatom
29
-------
production) accumulation rates in Turner and
Rabalais (1994a).
Diatom-based productivity and BSi accumula-
tion provide other lines of evidence of increased
productivity. In spite of a probable decrease in
Si availability, the overall productivity of the
ecosystem appears to have increased this
century. This is evidence by:
• Equal or greater net silicate-based
phytoplankton community uptake of silica in
the mixing zone, compared to the 1950s
(Turner and Rabalais, 1994b)
• Greater accumulation rates of biogenic silica
(BSi) in sediments beneath the plume, but
not further away, and in agreement with
results found in freshwater systems (Turner
and Rabalais , 1994a). The increased BSi in
Mississippi River bight sediments parallels
increased N loading to the system and is
direct evidence for the effects of
eutrophication on the shelf adjacent to the
Mississippi River.
Finally, an analysis of benthic foraminiferans in
offshore sediments indicates an increase in
oxygen deficiency stress this century, with a
dramatic increase since the 1940's-1950's.
Several cores from areas of varying levels of
frequency of hypoxia (in the Mississippi River
delta bight) were examined by Barun Sen Gupta
and colleagues (in Rabalais et al. in press, Sen
Gupta et al. in press). They documented a
progressive overall rise in oxygen stress (in
duration or intensity) with these indicators: (1)
an increase in the ratio otAtnmonium to
E.lplndinm, (2) a decrease in species not tolerant
of oxygen stress, and (3) an increase in species
tolerant of low oxygen stress. These changes
were coincident with the rise in river-borne
nutrients and accumulation of biogenic silica.
Increased bottom-water hypoxia could result
from increased organic loading to the seabed
and/or shifts in material flux (quantity and
quality) to the lower water column. Oxygen-
depleted bottom waters in the coastal ocean are
found worldwide. The incidence and extent of
such areas in coastal waters is apparently
increasing and related to anthropogenic nutrient
loadings in rivers (Diaz and Rosenberg in press).
The patterns of worsening water quality in
coastal waters adjacent to the terminus of major
rivers undergoing nutrient flux or water quality
alterations are consistent with the conditions
identified for the Mississippi River.
i Future Scenarios
The enormity of effecting environmental
change on the continental shelf at the terminus
of the Mississippi River might seem insur-
mountable for a watershed that includes
41 percent of the conterminous U.S.,
encompasses parts of many states and
innumerable other regulatory or legislative
boundaries, and integrates centuries of
landscape changes within the watershed and
alterations of the Mississippi River proper. Still,
effective policy for managing and restoring
ecosystems can be accomplished, especially if
the results of scientific inquiry are integrated
into the process.
Rabalais et al. (in press) made several
predictions of ecosystem response given certain
changes in nutrients (Figure 10):
If Si increases, and N remains the same; overall
N limitations would be similar to present, but Si
will no longer be limiting. This would result in
increased BSi and carbon accumulation in
sediments, and an increase in the extent and
severity of hypoxia.
-------
If Si increases, N increases, and Si and N remain
in balance; no N or Si limitations. The result
would be greatly increased BSi and carbon
accumulation and substantial increase in
severity and extent of hypoxia. This is the result
demonstrated in the 1993 flood and in a
doubled CO2 climate scenario.
If Si increases, N decreases to 1950s values; N
would return to the limiting nutrient status, and
although Si would be in abundant supply, the
system would be restricted by N supplies and
hypoxia would decrease.
To reach the 1950s levels of dissolved N would
require a 40-50 percent reduction in the current
loadings that exit the Mississippi River delta.
Identification of sources of nutrients within the
Mississippi River watershed that eventually reach
the Gulf of Mexico should lead to avenues of
management. While the results of changes in
nutrient delivery to the northern Gulf of Mexico
are clear, the delineation of the sources and their
fate and transformation as they are delivered to
the Gulf is not complete (however see Alexander
et al., Antweiler, and Goolsby in this volume). It
is important to understand which agricultural
practices, water treatment practices, water quality
regulations, consumer preferences, and economic
incentives and disincentives result in the amount
of dissolved N, P and Si in the Mississippi river.
Management alternatives directed at water issues
within the Mississippi River watershed may have
unintended and contrasting impacts on the coastal
waters of the northern Gulf of Mexico.
Howarth et al. (in review) modeled N loadings to
the North Atlantic Ocean and treated the
Mississippi River watershed as a unit. Sewage
input of N is 9 percent of the total inputs from
the Mississippi River to the North Atlantic
Ocean. Of four anthropogenic sources,
application of fertilizer contributes the greatest
input of N (54 percent), followed by fixation by
leguminous crops (31 percent) and atmospheric
deposition of NOy (15 percent). The ratio of
current, river N export to "pristine" river N export
for the Mississippi River ranges from a 4.8 to 7.4
fold increase. The Howarth et al. model points to
a reduction in nonpoint sources, agricultural,
direct or indirect, as the key to nutrient pollution
control.
There is a direct connection between river
nutrient loading and the hypoxic zones on the
Louisiana shelf. River diversions aimed at wetland
restoration might be considered a possible
management tool to decrease nutrient loading to
the offshore waters and thereby raise oxygen
concentrations in offshore bottom waters.
However, the amounts of river water to be
diverted are so small relative to the size of the
total discharge that river diversions will have an
insignificant effect on the size, frequency and
duration of oxygen depletion within bottom
waters offshore. For example, river diversions are
currently operating, or being planned, to bring
large quantities of water from the Mississippi
River into adjacent estuaries of Barataria Bay,
Breton Sound, and Lake Pontchartrain. U.S.
Army Corps of Engineers estimated maximum
flow for:
Davis Pond
10,650 cfs (likely to move west from delta)
Caernarvon
8,000 cfs (to east)
Bonnet Carre
30,000 cfs (to east)
Total i
48,650 cfs (equal 10 percent of average flow
4.6 xlO5 cfs, 1954-1988).
However, these diversions will function perhaps
one or two months of the year on variable
schedules for either fresh water or sediment
delivery, and the flows listed above are maximal.
The total diverted discharge will be significantly
less than 10 percent of the total discharge volume
for the year. Further, diversions of river waters to
-------
die east of the Mississippi River delta may
aggravate die limited and ephemeral conditions of
hypoxia there or further flow dirough die
Atchafalaya delta might increase die duration,
severity or extent of hypoxia on the southwestern
Louisiana shelf or die upper Texas coast.
Any discussion of nutrient management scenarios
must be undertaken within another overall
context—that of global climate change. A general
circulation model by Miller and Russell (1992)
predicted that runoff in freshwater would likely
increase for 25 of the world's 33 largest rivers.
Precipitation in the Mississippi River watershed is
likely to increase 20 percent with a doubled CO2
climate, and runoff is expected to increase in most
mondis, particularly May through August. This
response is likely to affect water column stability,
surface water productivity, and oxygen cycling in
die northern Gulf of Mexico.
Justic' ct al. (in press) applied a 20 percent
increase in freshwater flux, primarily during the
May-August period, to calculate an estimated
average monthly runoff of the Mississippi River at
Tarbert Landing compared to 1985-1992. The
integrated doubled CO2 runoff at Tarbert Landing
will be around 0.5 x!012m3 yr1. Assuming that the
highest increase in runoff will occur during May,
die maximum monthly runoff value for a doubled
CO2 climate will be approx. 4 x 104 m3 s'1. This
result is substantially higher than die mondily
maximum for the Great Flood of 1993 (3.2 x 104
m5 s"1). Surface salinity in the Gulf is likely to
decrease substantially, and water column stability
will increase. Manipulations of a physical-
biological two-box model (Justic' et al. in press)
indicated that there will be a 30-60 percent
decrease in summertime subpycnoclinal oxygen
content, relative to the 1985-1992 average. Under
those conditions, the hypoxic zone in the
nordiern Gulf of Mexico will probably expand
and encompass an area greater than that of die
summer of 1993.
f Effects on Living Resources
Hypoxia may affect fisheries resources by direct
mortality, altered migration, reduction in suitable
habitat, increased susceptibility to predation
(including by humans), changes in food resources
and susceptibility of early life stages. Studies of
benthic communities and demersal communities
show distinct responses of various members of
the communities to decreases in dissolved oxygen
concentration (Rabalais and Harper in prep.).
Oxygen deficiency stressed benthic communities
are characterized by limited taxa (none with direct
development, e.g., amphipods), characteristic
resistant infauna (e.g., a few polychaetes and
sipunculans), reduced species richness, severely
reduced abundances (but never azoic), low
biomass, and limited recovery following
abatement of oxygen stress (Rabalais et al, 1993;
1995).
^Summary
Hypoxia is a severe and dominant feature of the
northern Gulf of Mexico, that is linked to die
freshwater fluxes and nutrient loads of the
Mississippi and Atchafalaya Rivers. River water
quality has changed since the turn of the century
and accelerated since the 1950s. The adjacent
continental shelf ecosystem has responded by
increased productivity, eutrophication, and
oxygen stress. Solutions are warranted that are
directed at nutrient reductions dirough
management practices in the watershed.
fS r\ "x* ~~ ,' t
^Acknowledgments
Data were collected during programs funded by
NOAA Ocean Assessments Division, Louisiana
Board of Regents LaSER award 86-LUM(l)-083-
13 and LEQSF award (1987-90)-RD-A-l 5,
Louisiana Sea Grant College Program, NOAA
National Undersea Research Center, Louisiana
32
-------
State University, Louisiana Universities Marine
Consortium, Minerals Management Service,
Cooperative Agreement 14-35-0001-30470, and
the NOAA Coastal Ocean Program Office,
Nutrient Enhanced Coastal Ocean Productivity
(NECOP) study grant no. NA90AA-D-SG691 to
the Louisiana Sea Grant College Program, awards
no. MAR31, MAR24 and MAR92-02. Funding
for preparation of this manuscript was provided
by the NOAA NECOP program.
Refer prices ^--jf-^ ? T - x --- •
Bratkovich, A, S. P. Dinnel and D. A. Goolsby.
1994. Variability and prediction of
freshwater and nitrate fluxes for the
Louisiana-Texas shelf: Mississippi and
Atchafakya River source functions.
Estuaries 17(4): 766-778.
Brongersma-Sanders, M. 1957. Mass mortality
in the sea. Pages 941-1010 in}. W.
Hedgpeth (ed.), Treatise on Marine Ecology
and Paleoecology, Vol. I. Geol. Soc. Amer.
Mem. 67.
Conseil Permanent International pour
1'Exploration de k Mer. 1936. Bulletin
hydrographique pour Fannee 1935 (Series
Bl), 105 pp.
Diaz, R. J. and R. Rosenberg. 1995. Marine
benthic hypoxia: a review of its ecological
effects and the behavioral responses of
benthic macrofauna. Oceanogr. Mar. Biol.
Annual Rev. 33: in press.
Dortch, Q. andT. E. Whitledge. 1992. Does
nitrogen or silicon limit phytoplankton
production in the Mississippi River plume
and nearby regions? Continental Shelf
Research 12: 1293-1309.
Dowgiallo, M. J. (ed). 1994. Coastal Oceano-
graphic Effects of Summer 1993 Mississippi
River Flooding. Special NOAA Report,
NOAA Coastal Ocean Office/ National
Weather Service, Silver Spring, Maryland.
Eadie, B. J., B. A. McKee, M. B. Lasing, J. A.
Rpbbins, S. Metz and J. H. Trefry. 1994.
Records of nutrient-enhanced coastal ocean
productivity in sediments from the
Louisiana continental shelf. Estuaries 17(4):
754-765.
Howarth, R. W., G. Billen, D. Swaney, A.
Townsend, N. Jaworski, K. Lajtha, J. A.
Downing, R. Elmgren, N. Caraco, T.
Jordan, F. Berendse, F. Freney, V.
Kudeyarov, P. Murdoch and Z. Z-L. in
review. Riverine inputs of nitrogen to the
North Atlantic Ocean: Fluxes and human
influences. Biogeochemistry Special Issue,
Report of the Workshop on the Nitrogen
Budget for the North Atlantic Ocean,
International SCOPE Nitrogen Project,
Block Island, Rhode Island, May 1994.
Justic', D., N. N. Rabalais, R. E. Turner, and
W. J. Wiseman, Jr. 1993. Seasonal coupling
between riverborne nutrients, net
productivity and hypoxia. Mar. Pollution
Bull. 26(4): 184-189.
Justic', D., N. N. Rabalais and R. E. Turner.
1994. Riverborne nutrients, hypoxia and
coastal ecosystem evolution: Biological
responses to long-term changes in nutrient
loads carried by the Po and Mississippi
Rivers. Pages 161-67 in K. R. Dyer and R. J.
Orth (eds.), Changes in Fluxes in Estuaries:
Implications from Science to Management.
Proceedings of ECSA22/ERF Symposium,
International Symposium Series, Olsen &
Olsen, Fredensborg, Denmark.
Justic', D., N. N. Rabalais, R. E. Turner and Q.
Dortch. 1995a. Changes in nutrient
structure of river-dominated coastal waters:
Stoichiometric nutrient balance and its
consequences. Estuar. Coastal Shelf Sci. 40:
-------
339-356.
Justic', D., N. N. Rabalais and R. E. Turner.
1995b. Stoichiometric nutrient balance and
origin of coastal eutrophication. Mar.
Pollution Bull. 30(1): 41-46.
Justic', D., N. N. Rabalais and R. E. Turner, in
press. Effects of climate change on hypoxia
in coastal waters: a doubled CO2 scenario
for the northern Gulf of Mexico. Limnol.
Oceanogr.
Leming, T. D. and W. E. Stuntz. 1984. Zones
of coastal hypoxia revealed by satellite
scanning have implications for strategic
fishing. Nature 310: 136-138.
Lohrenz, S. E., M. J. Dagg and T. E.
Whitledge. 1990. Enhanced primary
production at the plume/oceanic interface
of the Mississippi River. Continental Shelf
Res. 10: 639-664.
Lohrenz, S. E., G. L. Fahnenstiel and D. G.
Redalje. 1994. Spatial and temporal
variations in photosynthesis parameters in
relation to environmental conditions in
coastal waters of the northern Gulf of
Mexico. Estuaries 17(4): 779-795.
Lohrenz, S. E., G. L. Fahnenstiel, D. G.
Redalje and G. A. Lang, in review.
Seasonal variability in primary production
and nutrient processes in the Mississippi
River plume. Continental Shelf Res.
Miller, J. R. and G. L. Russell. 1992. The
impact of global warming on river runoff.
J. Geophys. Res. 97: 2757-2764.
Nelson, D. M. and Q. Dortch. in press.
Silicic acid depletion and silicon limitation
in the plume of the Mississippi River:
Evidence from kinetic studies in spring
and summer. Continental Shelf Res.
Pokryfki, L. and R. E. Randall. 1987.
Nearshore hypoxia in the bottom water of
the northwestern Gulf of Mexico from
1981 to 1984. Mar. Environmental Res.
22: 75-90.
Qureshi, N. A. 1995. The role of fecal pellets
in the flux of carbon to the sea floor on a
river-influenced continental shelf subject
to hypoxia. Ph.D. Dissertation,
Department of Oceanography & Coastal
Sciences, Louisiana State University, Baton
Rouge, 255 pp.
Rabalais, N. N. 1992. An Updated Summary
of Status and Trends in Indicators of
Nutrient Enrichment in the Gulf of
Mexico. Report to Gulf of Mexico
Program, Nutrient Enrichment
Subcommittee. Publ. No. EPA/800-R-92-
004, U.S. Environmental Protection
Agency, Office of Water, Gulf of Mexico
Program Office, Stennis Space Center,
Mississippi, 421 pp.
Rabalais, N. N., R. E. Turner, W. J.
Wiseman, Jr. and D. F. Boesch. 1991. A
brief summary of hypoxia on the northern
Gulf of Mexico continental shelf: 1985—
1988. Pages 35-46 in R. V. Tyson and T.
H. Pearson (eds.), Modern and Ancient
Continental Shelf Anoxia. Geological
Society Special Publ. No. 58. The
Geological Society, London.
Rabalais, N. N., R. E. Turner and W. J.
Wiseman, Jr. 1992. Distribution and
characteristics of hypoxia on the Louisiana
shelf in 1990 and 1991 Pages 15-20 in
Proceedings, Nutrient Enhanced Coastal
Ocean Productivity Workshop. Publ. No.
TAMU-SG-92-109, Texas Sea Grant
College Program, Texas A&M University,
College Station, Texas.
34
-------
Rabalais, N. N., L. E. Smith, E. B. Overton
and A. L. Zoeller. 1993. Influence of
Hypoxia on the Interpretation of Effects
of Petroleum Production Activities. OCS
Study/MMS 93-0022. U.S. Dept. of the
Interior, Minerals Management Service,
Gulf of Mexico OCS Region, New
Orleans, Louisiana, 158 pp.
Rabalais, N. N., W. J. Wiseman, Jr., and R. E.
Turner. 1994a. Hypoxic conditions in
bottom waters on the Louisiana-Texas
shelf. Pages 50-54 in M. J. Dowgiallo (ed.)
Coastal Oceanographic Effects of
Summer 1993 Mississippi River Flooding,
Special NOAA Report, U. S. Dept. of
Commerce, National Oceanic and
Atmospheric Administration, Coastal
Ocean Program, Silver Spring, Maryland.
Rabalais, N. N., W. J. Wiseman, Jr. and R. E.
Turner. 1994b. Comparison of continuous
records of near-bottom dissolved oxygen
from the hypoxia zone of Louisiana.
Estuaries 17(4): 850-861.
Rabalais, N. N., L. E. Smith, D. E. Harper,
Jr. and D. Justic'. 1995. Effects of Bottom
Water Hypoxia on Benthic Communities
of the Southeastern Louisiana Continental
Shelf. OCS Study MMS 94-0054. U.S.
Dept. of the Interior, Minerals
Management Service, Gulf of Mexico
OCS Region, New Orleans, Louisiana, 105
pp.
Rabalais, N. N., R. E. Turner, D. Justic', Q.
Dortch, W. J. Wiseman, Jr. and B. K. Sen
Gupta. 1996. Nutrient changes in the
Mississippi River and system responses on
the adjacent continental shelf. Estuaries
19(2B): in press.
Rabalais, N. N., R. E. Turner and W. J.
Wiseman, Jr. in review. Hypoxia in the
northern Gulf of Mexico: Linkages with
the Mississippi River. In Proceedings, Gulf
of Mexico Large Marine Ecosystem
Symposium, August 1995.
Ragan, J. G., A. H. Harris and J. H. Green.
1978. Temperature, salinity and oxygen
measurements of surface and bottom
i
waters on the continental shelf off
Louisiana during portions of 1975 and
1976. Prof. Pap. Ser. (Biol.) (Nicholls State
Univ., Thibodaux, Louisiana) 3: 1-29.
Renaud, M. 1986. Hypoxia in Louisiana
coastal waters during 1983: implications
for fisheries. Fishery Bull. 84: 19-26.
Richards, F. A. 1954. Oxygen in the ocean.
Pages 185-238 in]. W. Hedgpeth (ed.),
Treatise on Marine Ecology and
Paleoecology. Geol. Soc. Amer. Mem.
Vol. 67, No. 1.
Sklar, F. H. and R. E. Turner. 1981.
Characteristics of phytoplankton
production off Barataria Bay in an area
influenced by the Mississippi River. Cont.
Mar. Sci. 24: 93-106.
Sen Gupta, B. K., R. E. Turner and N. N.
Rabalais. 1993. Oxygen stress in shelf
waters of northern Gulf of Mexico: 200-
year stratigraphic record of benthic
foraminifera. Page A138 in Geological
Society of America, 1993 Annual Meeting,
Abstract.
Sen Gupta, B. K., R. E. Turner and N. N.
Rabalais. 1996. Seasonal oxygen depletion
in continental-shelf waters of Louisiana:
Historical record of benthic foraminifers.
Geology 24: in press.
-------
Turner, R. E. and R. L. Allen. 1982. Bottom
water oxygen concentration in the
Mississippi River delta bight. Contr.
Marine Sci. 25:161-172.
Turner, R. E. and N. N. Rabalais. 1991.
Changes in Mississippi River water quality
this century. Implications for coastal food
webs. BioScience 41(3):140-147.
Turner, R. E. and N. N. Rabalais. 1994a.
Coastal eutrophication near the
Mississippi river delta. Nature 368:619-
621.
Turner, R. E. and N. N. Rabalais. 1994b.
Changes in the Mississippi River nutrient
supply and offshore silicate-based
phytoplankton community responses.
Pages 147-150 in K. R. Dyer and R. J.
Orth (eds.), Changes in Fluxes in
Estuaries: Implications from Science to
Management. Proceedings of
ECSA22/ERF Symposium, International
Symposium Series, Olsen & Olsen,
Fredensborg, Denmark
Wiseman, Jr., W. J. and F. J. Kelly. 1994.
Salinity variability within the Louisiana
Coastal Current during the 1982 flood
season. Estuaries 17(4):732-739.
Wiseman, Jr., W. J., N. N. Rabalais, R. E.
Turner, S. P. Dinnel, and A.
MacNaughton. submitted. Seasonal and
interannual variability within the Louisiana
Coastal Current: Stratification and
hypoxia. J. Mar. Systems.
-------
Figure 7.
Distribution of near-bottom water hypoxia (dissolved 02 < 2 mg T ) i
mid-summer for the dates indicated in 1992, / 993, / 994 and 1995.
Data from hypoxia monitoring studies
of N. N. Rabalais, R. E. Turner and W.j. Wiseman, jr.
(From Rabalais et al. in review.)
Salinity %» and Temperature °C
2022342329X32343
Salinity %e and Temperature °C
20222428X30323433
2
4.
6
s8
£3 10
O.
0>
0 12,
14.
56
18
20
f * J
t Stn C6B : '
' \ 7/IMS ? /
"^^'C'..
»' 4 XTJ
',• / V
x> 4 0
*' 4 0
,«' »' 9
t \
• 0
• --a-.'Onpmtn 0
* l^wAaH °
/ * * Q*"" ?
i ?
i i s
o
2
4
S
8
10
12
14
16 •
18
20
'
«? « f
•) 4 »
:: StnC7 : :
i i i
8/19/95 f *
* 'o- a v
~~-o^ i •'
^'Q, 4 4
\ 4 V
a, 4 i
~^\ *
4 ~«>
i 4
i i
/ ,•'",
£-4 a«
» ft' o
p' / i
• • « i
1 £ 0
i i' *
i «'' V
1 6 6
1 1 1 1 1 1 1
012345
Dissolved Oxygen (mg/L)
012345678
Dissolved Oxygen (mg/L)
Figure 8.
Structure of water column for a salinity-controlled pycnocline and profile of
dissolved oxygen (left panel) and a temperature-controlled pycnocline and profile
of dissolved oxygen (right panel). Data from hypoxia monitoring studies ofN. N.
Rabalais, R. £. Turner and W. J. Wiseman, Jr.
37
-------
TRANSECT C
1992
1993
9/16
8/19
4/12
(frera Rabalais et a)., 1991*)
Figure 9.
Cross-section of southeastern Louisiana shelf showing seasonal
progression and extent of bottom hytooxic zones during 1992 and 1993.
Stippled area indicates < 2 mg /" ; black areas indicate < I mg /" .
(From Rabalais et a/. / 994a.)
1990
Yes N Limitation Some
No Si Limitation Some
4:1 SUN Ratio -1:1
N Limitation
Si Limitation
Si:N Ratio
Some
No
>1:1
N Limitation
Si Limitation
Si:N Ratio
No
No
-1:1
N Limitation Yes
Si Limitation Mo
Si:N Ratio »1:1
DOCUMENTED
RESPONSE
PREDICTED RESPONSE
BSi Accumulation
C Accumulation
Hypoxia Increase
BSi Accumulation f BSi Accumulation ii BSi Decrease |
C Accumulation \ C Accumulation A A C Decrease 4
Hypoxia Increase T Hypoxia Severe A.A, Hypoxia Decrease *
Figure 10.
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.
(Modified from Rabalais et al. in press.)
-------
Presentation Discussion
Nancy Rabalais (Louisiana State
University—Baton Rouge, LA)
Alan Ballard (Gulf of Mexico Program—Stennis
Space Center, MS) asked Nancy Rabalais what
percentage of the Mississippi River outflow
flows into the hypoxic area.
Nancy Rabalais responded by saying that
William Wiseman addressed the amount of
freshwater in the shelf and showed the seasonal
progression of the freshwater in the area during
his presentation. Most of the Atchafalaya River
and 50 percent of the Mississippi River
discharge water flow west.
Scott Dinnell (USM Center for Marine
Scienc—MS) added that 10 percent of the total
water content of the shelf (although it is
seasonal) is freshwater.
Neil Armingeon (Lake Pontchartrain Basin
Foundation—Metairie, LA) noted that Nancy
Rabalais mentioned that one of the proposed
river diversions would have little or no impact
on the hypoxic zones in the Gulf. He then
continued by asking her opinion on the impact
that diversion would have on the quality and
management of the receiving waters.
Nancy Rabalais responded by saying that she
felt river diversions should be done to control
freshwater and sediments, not as a panacea for
the low oxygen conditions.
Mike Waldon (USL—Lafajette, LA) disputed
that 48,000 csf is 10 percent of the average flow
of the river. He said that the total average is
approximately 450,000 cfs. Therefore, 48,000
csf is slightly more than 10 percent. He added
that only some of the possible diversions were
presented.
Nancy Rabalais agreed that there are other
possible diversions. She had presented only
those diversions for which she had data, and
that most directly affected the southwestern
shelf. She continued by saying that she had
demonstrated the southwestern shelf down-
plume from the Atchafalaya River also
experiences extreme bouts of hypoxia and
stressed that those areas compared statistically
to the two-month time lag of peak river flow in
the Atchafalaya River system.
An unidentified gentleman from the audience
asked Nancy Rabalais if the growth of soybeans
instead ,of the over application of fertilizer could
be the primary contributor to the hypoxia
problem since the sources of nutrients in the
watershed were 50 percent from fertilizers,
30 percent of that being from leguminous
crops.
Nancy. Rabalais responded that there are many
nutrient sources to the river. Those sources and
the fate and transformation of those sources, as
they move down the watershed, are not yet well
understood.
Lon Sttong (U.S. Department of Agriculture/
Natural, Resources Conservation Service—Jackson,
MS) asked if the sewage discharge data
presented represented treated or untreated
wastewater.
He also asked Nancy Rabalais to comment on
her statement that sewage was not a significant
source of input. Wastewater treatment
discharges are continuous and the total nitrogen
content in the effluent ranges between
2—60mg/L, compared to most run-off from
cropland which is storm-event driven and lower
in nitrogen content.
Nancy Rabalais responded by saying that she
did not know all of the details of Bob
39
-------
Howatth's paper but that Bob Howarth and his
colleagues do not consider sewage an
anthropogenic input to the system. She
concluded by saying the budget presented was
for annual inputs for the whole watershed.
Paul LaViolette (Gulf Weather Corporation—
S'tennis Space Center, MS) asked if other effects of
modifying the flow of the river through the
Atchafalya had been studied.
Nancy Rabalais replied that it was one of the
management scenarios for coastal Louisiana.
40
-------
rt
We have, over the course of 16 years,
accumulated considerable informa-
tion on the responses of benthonic
and demersal organisms to hypoxia (< 2 mg/1
dissolved oxygen) via direct diver observations,
remotely operated vehicle (ROV) video systems,
and by sample collection using benthic cores
and grabs. As might be expected, the responses
of the fauna vary, depending on extent of
oxygen depression. In a progressive decrease
from 2.0 to about 0.2 mg/1, we have observed
progressive disappearance of motile organisms
(fish, cephalopods, and crustaceans), to
pronounced stress behavior of benthonic
organisms incapable of escape, to emergence of
deep burrowing benthonic organisms from their
burrows, to death of these organisms. At near
0.2 mg/1, the sediment becomes black and
sulfur-oxidizing bacteria form "cottony" mats
on the sediment surface. At 0.0 mg/1 there is no
sign of aerobic life, only black sediments.
Episodic hypoxic/anoxic stress may result in
temporary destabilization of the benthic
assemblage. Evidence suggests that the
community recovers to its former diversity and
abundance. Repeatedly stressed communities,
however, have low diversity, no species with
direct development, a few highly tolerant
species, low biomass and limited recovery
following abatement of hypoxia.
Hypoxia (< 2.0 mg/1 of dissolved oxygen
[D.O.]) and anoxia (0.0 mg/1 D.O.) occur in
many localities (see Tyson and Pearson 1991
and Diaz and Rosenberg 1995, for reviews). In
most cases the data generated by studies of
hypoxia have reported reduced abundances of
benthic fauna and/or the absence of nektonic
fauna. There are, however, relatively few
studies documenting the effects on these
organisms as the oxygen concentration
decreases from normoxic to below 2.0 mg/1 and
continues to decrease toward anoxia. Jorgensen
(1980) and Stachowitsch (1991), using scuba,
documented behaviors of benthonic organisms
during the onset of anoxia, including stress
behavior of actiniarians (anemones), ophiuroids
(brittlestars), gastropods and bivalves.
In the northwestern Gulf of Mexico (defined
here as the Louisiana coast west of the Missis-
sippi Delta and the Texas coast north of
Matagorda Bay), prior to 1979, there were a few
published reports of dead organisms being
collected by trawl or seen by divers (Harper
41
-------
et al. 1981, 1991), and there were anecdotal
reports of shrimp fishermen avoiding large
areas because nothing was being caught, but no
systematic studies had been conducted.
Methods
The behaviors of benthic infauna and epifauna,
and demersal nektonic organisms, were
observed direcdy by scuba divers, and by
"flying" ROV's. The responses of benthic
communities to hypoxia and anoxia were
determined by collecting benthic samples using
diver-operated, or remotely operated, grab
samplers and cores. In addition to direct
observation, divers used Nikonos cameras to
obtain wide angle photographs and macro-
photographs, and the ROV excursions were
documented on videotape via electronic signals
received from the on-board cameras. These
observations and photographic records were
made principally during two long-term studies.
The first occurred off Freeport, Texas, during a
7-year (1977—1984) study of the macrobenthos,
and the second occurred off Cocodrie,
Louisiana, during a 5-year (1989—1993) period
in which 1-week cruises were made each
summer on the Louisiana continental shelf.
Results
In June 1979, during a study of the macro-
benthos off Freeport, Texas, divers reported
seeing apparently dead infaunal organisms lying
on the bottom amidst various sized patches of
"cottony" material (probably the
sulfur-oxidizing bacterium Beggiatoa); the
divers also reported smelling hydrogen sulfide
in the bottom water (Harper et al. 1981, 1991).
Pavella et al. (1983) simultaneously collected
virtually no nekton in the area. Data collected
along cross shelf transects indicated the "dead"
water extended from about 6-m depth out to
about 30-m depth, a cross-shelf distance of 50
km (Harper et al. 1981; Pavella et al. 1983).
During, and immediately following, the event,
infaunal abundances decreased to the lowest
levels reported during the 7-year (1977-1984)
study. The deeper study..site (21-m depth)
recovered quickly to pre-hypoxia conditions
(Harper et al. 1991) At the shallower site (15-m
depth), however, the benthic community was
apparently destabilized, because a succession of
dominance occurred in which one species
became numerically dominant for 1 to 3
months,-then was replaced by another domin-
ant. This process continued for almost 2 years
until the spionid polychaete Paraprionospio
pinnata regained its pre-hypoxia dominance
(Harper etal. 1991).
Detailed studies of hypoxia began off Louisiana
in 1985 and continue to the present (Rabalais
and Harper, 1991, 1992; Rabalais et al. 1991,
1992, 1994a, b; 1996, in press). These studies
have documented that up to 18,000 km2 of the
Louisiana continental shelf may be impacted by
hypoxic/anoxic bottom water. Week-long late
summer cruises off Louisiana from 1989
through 1995 have documented several stages
of hypoxia/ anoxia, ranging from hypoxia onset
to virtual absence of oxygen and the presence of
hydrogen sulfide in the water column. These
observations have led to the creation of a step-
wise effects diagram (Figure 11). Above
2.0 mg/1 (normoxia) divers occasionally observe
fish, but fish, squid, and large mobile bottom-
dwelling organisms are routinely seen during
ROV tows (Figure 12). Another characteristic
of normoxic water is that it is generally turbid
and visibility is limited. As the oxygen level
decreases from 2.0 to around 1.5 mg/1 the
mobile organisms usually disappear. Very often
the turbidity in the water decreases. Further
reduction from 1.5 to 1.0 mg/1 causes stress
behavior in smaller bottom-dwelling organisms;
crabs and sea stars climb on top of high points,
brittlestars emerge from the sediment and use
42
-------
their arms to raise their disks off the substrate,
burrowing shrimp emerge from the bottom,
snails move about the bottom with their
siphons directed upward, large burrowing
worms emerge from the substrate (Figure 13).
All these actions are taken to position the
organisms' ventilation mechanisms (gills,
siphons, etc) above the microenvironment at
the sediment-water interface where the hydro-
gen sulfide concentration may be increasing. At
oxygen levels of 1.0 to 0.5 mg/1 the sediment
surface develops "fairy rings" of thin strands of
bacterial filaments that appear to expand out-
ward leaving black sediment in the center of the
ring (Figure 14). At this stage, even the most
tolerant burrowing organisms, principally
various types of worms, emerge partially or
completely from their burrows and lie motion-
less on the bottom (Figure 15). Often these
organisms can be revived if placed in oxygen-
ated water; several different motionless, and
apparently dead, species have been collected in
jars on the bottom, returned to the surface and
placed in aerated water, and have revived within
a half hour. As the oxygen concentration
decreases from 0.5 toward zero, bottom
organisms die. They do not, apparently decom-
pose rapidly, because their bodies continue to
litter the bottom (Figure 16). The absence of
large scavengers is also evidenced by the fact
that the corpses remain on the bottom and are
not eaten. At 0.0 mg/1 the sediment becomes
almost uniformly black and there is no sign of
life; even the strands of the sulfur-oxidizing
bacterium Beggiatoa are absent.
rbr
iscussion
One of the principal problems associated with
hypoxia and anoxia in the marine environment
is that the effects on the benthonic and nektonic
communities usually cannot be observed
directly; the effects must be inferred via
reductions in trawl catches or reductions in
collected bottom-dwelling organisms. Without
the visual impact of stressed, dying, or dead
organisms littering the bottom, the effects of
reduced oxygen do not generate the same level
of consternation that would occur if the same
type of catastrophe occurred on land. We
suggest that if, for two to three months each
summer, dead animals were strewn over
r)
18,000 kirr of land in Missouri or Iowa, people
would rightly be upset, and that efforts would
be immediately undertaken to correct the situa-
tion. Those in agricultural states who depend on
soils for their livelihood must realize that worms
in the marine environment serve the same func-
tion that earthworms do onland; they burrow
into the soil, cause mixing, and at the same time
allow oxygen to penetrate to deeper levels than
would be possible without them. The worms,
both terrestrial and marine, contribute greatly to
increased overall productivity of their respective
habitats.
It is generally accepted that the continental shelf
of Louisiana is impacted by hypoxia almost
annually, but it had been assumed that occur-
rences off Texas were infrequent to rare. Studies
of the benthos off Freeport, Texas, revealed
that hypoxia occurs more frequently off the
upper Texas coast than had previously been
believed. During the 7-year study off Freeport,
Texas, there were two confirmed events of
D.O. decreasing below 2 ppm, and three other
suspected incidents; if these latter occurred, the
events were fairly short-lived and were missed
due to the schedule of sampling cruises.
Because of the limited area being sampled we
do not know the area affected during the
catastrophic 1979 event. If, however, as we
suspect, the hypoxic water mass was imported
from the Louisiana shelf, and extended offshore
' o
to at least 28 km, an area of at least 5,400 km
was affected along the upper Texas coast. The
other event(s) was (were) fairly short-lived and
probably had minor effects. The shallower site
43
-------
benthic community was apparently destabilized
by the hypoxic event and required about two
years to return to pre-event conditions.
The Texas experience is in contrast to the
conditions that exist on the Louisiana shelf. The
benthos off Louisiana are subjected to hypoxia
almost annually, and the hypoxia often extends
over 3-4 months. Thus the benthos rarely, if
ever, attains a climax community. Rather, the
community is grossly reduced, or eliminated,
annually and must be reestablished following
break-up of the conditions producing hypoxia.
This prevents establishment of a climax
community.
References ;
Diaz, R. J. and R. Rosenberg. 1995. Marine
benthic hypoxia. A review of its ecological
effects and the behavioural responses of
benthic macrofauna. Oceanography and
Marine Biology Annual Review. In press.
Jorgensen, B. B. 1980. Seasonal oxygen
depletion in the bottom waters of a Danish
fjord and its effect on the benthic
community. Oikos, 34: 68—76.
Harper, D. E., Jr., L. D. McKinney, R. R. Salzer
and R. J. Case. 1981. The occurrence of
hypoxic bottom water off the upper Texas
coast and its effects on the benthic biota.
Contributions in Marine Science. 24: 53—79.
Harper, D. E., Jr., L. D. McKinney, J. M. Nance
arid R. R. Salzer. 1991. Recovery responses
of two benthic assemblages following an
acute hypoxic event on the Texas
continental shelf, northwestern Gulf of
Mexico, pp. 49—64. In: Modern and Ancient
Continental Shelf Anoxia. Geological
Society Special Publication No. 58. London.
Pavella, J. S., J. Ross and M. E. Chittenden.
1983. Sharp reduction in abundance of
fishes and benthic macroinvertebrates in the
Gulf of Mexico off Texas associated with
hypoxia. Northeast Gulf Science. 6:
167-173-
Rabalais, N. N. and D. E. Harper, Jr. 1991.
Studies of the benthic oxygen phenomenon
off Louisiana, pp. 57-63. In: International
Pacifica Scientific Diving ...1991. Pro-
ceedings of the American Academy of
Underwater Science Eleventh Annual
Scientific Diving Symposium, 25—30 Sept-'
ember 1991, Honolulu, Hawaii. H.-J.
Krock and D. E. Harper, Jr. (eds.).
Rabalais, N .N., R. E. Turner, W. J. Wiseman,
Jr. and D. F. Boesch. 1991. A brief summary
of hypoxia on the northern Gulf of Mexico
continental shelf: 1985-1988. pp. 35-47. In:
Modern and Ancient Continental Shelf
Anoxia. Geological Society Special Publica-
tion No. 58. London.
Rabalais, N. N. and D. E. Harper, Jr. 1992.
Studies of benthic biota in areas affected by
moderate and severe hypoxia. pp. 48—51. In:
Nutrient Enhanced Coastal Ocean Produc-
tivity. Publication No. TAMU-SG-92-109.
Texas Sea Grant College Program,
Galveston, Texas.
Rabalais, N. N., W. J. Wiseman, Jr. and R. E.
Turner. 1994a. Comparison of continuous
records of near-bottom dissolved oxygen
from the hypoxia zone of Louisiana.
Estuaries, 17:850-861
44
-------
Rabalais, N. N., R. E. Turner and W. J. Wise-
man, Jr. 1994b. Hypoxic conditions in
bottom waters on the Louisiana-Texas shelf.
pp. 50—54. In: Coastal Oceanographic
Effects of Summer 1993 Mississippi River
Flooding. Special NOAA Report, National
Oceanic and Atmospheric Administration,
Coastal Ocean Office, Silver Spring,
Maryland. M. J. Dowgallo (ed.).
Rabalais, N. N., R. E. Turner, D. Justic, Q.
Dortch, W. J. Wiseman, Jr. and B. K. Sen
Gupta. 1996. Nutrient changes in the
Mississippi River and system responses on
the adjacent continental shelf. Estuaries 19:
In press.
Rabalais, N. N., R. E. Turner and W. J.
Wiseman, Jr. In press. Hypoxia in the
northern Gulf of Mexico: Linkages with the
Mississippiiver. Proc., Large Marine
Ecosystems, Gulf of Mexico Program.
Stachowitsch, M. 1991. Anoxia in the northern
Adriatic Sea: rapid death, slow recovery.
pp. 119-129. In: Modern and Ancient
Continental Shelf Anoxia. Geological
Society Special Publication No. 58. London.
Tyson, R. V. and T. H. Pearson (eds.). 1991.
Modern and Ancient Continental Shelf
Anoxia. Geological Society Special
Publication No. 58. London. 470 pp.
45
-------
Dissolved Oxygen Limits for
Communities and Indicators of Stress
(Dead nekton)
Dead
portunids
Dead
majids
No large
living organisms
Dead benthos
S-ox
bact._
I
Larger
nekton
Shrimp, motile
larger invertebrates
Other large, less motile invertebrates
e.g., Persephone (purse crab)
Stressed larger benthos,
moribund benthos
Macrolnfauna reduced
0.0
Sulfldlc
stolments
0.2
-3%
0.5
-TV,
1.0 1.5
Dissolved Oxygen (mg/l)
-14% -21%
-27%
Diagram of components of the demersal and benthic communities and responses to varying levels of dissolved oxygen (mg/l) In
tha overlying bottom waters. Figure is non-quantitative along the y-nxis and represents estimates and ranges along the x-axis.
Compiled from benthic studies of N. N. Rabalais and D. E. Harper, Jr.
Figure II.
Step within decrease in oxygen concentration as observed on the
Louisiana continental shelf (from Rabalais and Harper 1992).
Figure 12. (Kioto courtesy of Franklin Viola)
Benthonic organisms photographed in normoxic water.
Cerianthid burrowing anemone
46
-------
figure 13.
Small unidentified anemone.
Figure 14.
Hermit crab.
47
-------
figure J5.
Cantharus cancellarus snail moving about substrate with siphon directed upward.
Figure 16.
Unidentified snapping shrimp emerged from burrow.
48
-------
figure 17.
Astropecten sea star atop a mound
Figure 18.
Unidentified brittlestars on the surface
with disks raised above the substrate.
SSffi-BfgSS,ss.1^^g[3gBaKSp^affl,ffl^^^
49
-------
Figure 19.
Unidentified brittlestars on the surface vwth disks raised above the substrate.
figure 20.
Choleia viridis potychaete
50
-------
Figure 21. (Photo courtesy of Frankffn Vfafa)
BeggJotoa filaments on the surface of the bottom. Moribund organisms lying
motionless on the bottom in oxygen concentrations of 1.0 to 0.5 mg/l.
«- -,-
22. (Photo courtesy of Franklin Viola)
Cerionth/d qnenome.
-------
Figure 23.
Unidentified polychaetes.
figure 24.
Unidentified polychaetes.
52
3a^^^
-------
Figure 25.
Hem/chore/ate.
Figure 26.
Dead organisms on the bottom in oxygen concentrations of < 0.5 mg/L
Unidentified potychaete.
^
53
-------
figure 27.
Unidentified crab.
Figure 28.
Unidentified burrowing shrimp.
54
-------
Presentation Discussion
concentration of fish above the bolus.
Don Harper (Texas A&M University—
Galveston, TX)
An unidentified woman from Texas stated that
there was some evidence that the episodic quality
of hypoxic waters in response to hurricanes pass-
ing to the east of the hypoxic area may cause
deterioration of benthic organisms on die eastern
coast of Texas. She asked Don Harper if he had
found any further evidence of this effect, and
referred to a quote he made in the Houston
Chronicle as saying perhaps one of the fish kills in
east Texas was associated with the 1983
phenomenon.
Don Harper said he had no evidence which sup-
ported diat observation and felt diat she had
misinterpreted his statement. He clarified the
Houston Chronicle quote by saying he was discus-
sing a fairly localized dinoflagellate bloom and he
was not aware of the effects it had on bottom
•waters.
Don Boesch (University of Maryland—
Cambridge, MD) asked Don Harper if he
understood the fate of mobile organisms such as
shrimp in areas where the oxygen stress is not
severe enough to cause complete mortality of
benthic organisms.
Don Harper discussed a series of five summer
cruises during which he and Nancy Rabalais used
an ROV to generate many videos and slides.
Because there was a sharp demarcation between
the hypoxic and normoxic waters, they expected
an abundance of fish and shrimp along the edge
of die zone. Instead, the last segment on the
video tape showed a bolus of hypoxic water in
the middle of the water mass and a huge
He continued by discussing a cruise that he and
Eugene Turner completed during which diey
crossed a boundary between hypoxic and
normoxic waters. The water mass was being
moved westward by water which was up welling
from the deep-water area off die Mississippi
Delta. Heading back toward Louisiana, there was
a gradual transition from normoxic to hypoxic
waters and no evidence of a strong concentration
of fish or shrimp. He believed that this was a
result of die shrimp traveling further up die water
column, making them ideal prey for sight feeding
predators.
He justified tiiis dieory by explaining diat when
bottom water begins to turn hypoxic it becomes
very clear because die picnocline (boundary)
prevents suspended material from breaking
dirough. The suspended material already in the
bottom water settles out, causing die water to
become very clear. In these conditions, shrimp
are easily seen by sight feeding predators.
Eugene Turner (Louisiana State University-
Baton Rouge, LA) added diat he had also
witnessed tiiis phenomenon. He has seen squid
dive into die hypoxic layer to feed on shrimp.
Don Harper commented diat the video tape
depicting Nancy Rabalais' "D-transect" showed a
concentration of fish in the upper portion and in
die lower portion tiiey were absent.
William Herke (Citizens for a Clean Environ-
ment—Baton Rouge, LA) commented that
when die water oxygen is normal tiiose types of
organisms are found in die bottom of the water
column. He asked Don Harper if diese organ-
isms were important as a base of die food chain
for other commercially important organisms.
55
-------
Don Harper replied that shrimp spend their time
on the bottom when they are feeding and work
their appendages into the bottom looking for
polychaetes and small crustaceans. Most of the
organisms that were viewed in die video were too
large to be prey for shrimp. However, benthic
communities are extremely important in shrimp
production.
Though he has not been able to establish a
scientific correlation, the period of time when
lowest benthic abundances were found was also
the period of greatest shrimp landings in the
Galveston region. Therefore, he theorized that
the shrimp are feeding on a lot of the benthic
organisms during that period. He intends to
continue his efforts to demonstrate a correlation
between the necessity of the benthic organisms to
die overall healtfi of the shrimp community.
56
i?j!^3r^a!iw^^
ff!^^
-------
-.
A«i^<^a^?js8jisgwts,»i^^«sw!^t,rfJ»J«4R f-f*r>*j-~ —--
MlK -l^l > > ^V - '"
Abstract
The inner shelf of the northern Gulf of
Mexico from the Mississippi River Delta
westward to the Texas-Louisiana border
is the site of the highly stratified Louisiana
Coastal Current fed by the Mississippi River
system. The initial efflux of water from the
mouths of the Mississippi River occurs as highly
stratified plumes. Lagrangian measurements
within the plume of Southwest Pass are
consistent with concurrent satellite imagery.
Nutrient distribution patterns are similar to
those of physical parameters and suggest
conservative dilution during the first day
following nutrient release from the river mouth.
After the waters attach to the coast to form the
Louisiana Coastal Current, they tend to flow
westward, although they do respond to wind
shifts on periods of a few days or longer. The
physical structure of this region is dominated by
the strength and phasing of river discharge and
wind forcing. This structure exerts a strong
control on the distribution of hypoxia and the
processes responsible for its spatial and
temporal variability.
Introduction
The Mississippi-Atchafalaya River system
constitutes the dominant control on the
oceanographic character of the Louisiana inner
shelf. This river system drains 43 percent of the
contiguous United States and parts of two
Canadian provinces. It delivers, on average,
approximately 635 km3 of fresh water to the shelf
each year, as well as massive amounts of
suspended sediment and dissolved nutrients.
Peak discharge occurs in the spring. Thirty
percent of this water enters the Gulf through the
Atchafalaya River and the remainder through the
Mississippi River delta. Of this latter volume, the
amount flowing onto the west Louisiana shelf is
uncertain, but often quoted as being
approximately 50 percent. Much of this fresh
•water entering the west Louisiana shelf hugs the
coastline and flows westward as a narrow current,
the Louisiana Coastal Current. During most of
the year, this current flows westward into Texas
and even Mexican waters. During the summer
months, though, strong southerly winds along the
south Texas coast tend to push water back onto
the Louisiana shelf and the current may reverse
for a month to six weeks, particularly near the
Texas-Louisiana border.
57
-------
It is commonly accepted that the Mississippi
River system discharge is intimately related to
the development of summer hypoxia on the
Louisiana inner shelf (Rabalais et al., in press).
In its most simplistic form, the paradigm for the
development of summer hypoxia is that the
Mississippi and Atchafalaya Rivers load the
coastal waters with massive amounts of new
nutrients. Rapid phytoplankton growth results.
These phytoplankton cells sink to the bottom
and utilize oxygen, either through respiration or
as they decay. Reoxygenation of the near-
bottom waters is prevented by a strong density
interface in the water column which results
from light, low salinity water from the river
system lying over heavy, saltier shelf waters. In
order to understand the dynamics of hypoxia on
die shelf, it is necessary to understand the
physics of the inner shelf.
The Atchafalaya River empties into Atchafalaya
Bay, a broad, shallow estuary, and then onto a
broad, shallow region of the shelf. It dissipates
energy through bottom friction and entrains
ambient fluid through lateral mixing (Wang,
1984). In contrast, plumes from the dominant
passes of the Mississippi River delta expand
buoyantly and entrain water both from the sides
and from below (Wright and Coleman, 1971).
Studies of the Mississippi River plumes are
maturing (e.g. Hitchcock et al., in review), while
those of the Atchafalaya input are in their
infancy.
The water mass modifications important to
hypoxia should be described in a Lagrangian
sense, i.e. following a water particle. We have
only begun to address this type of description of
processes on the inner shelf. Recent plume
studies (Hitchcock et al., in review) have tracked
water from the mouth of Southwest Pass for
time periods of the order of one day, roughly
die time necessary for the plume waters to
merge into the Louisiana Coastal Current. Near
surface current speeds can be as high as 1 m/s.
Vertical entrainment rates at the base of the
plume are estimated to range between 0.25 and
1 m/hr. Modification of nutrient concentra-
tions (NO3, SiO4) during these early periods
after water is released from the mouth of the
river is not inconsistent with conservative
mixing. The subsequent modification of waters
entering the Louisiana Coastal Current have not
been studied in a Lagrangian sense. Conse-
quently, further downstream changes in water
characteristics must be inferred from Eulerian
measurements.
The seasonal changes in runoff to the Louisiana'
Coastal Current alter the stratification of the
waters of the current (Wiseman et al., 1986).
Fresh water floats atop salty water. Even during
winter, when the river and nearshore waters are
colder than the deeper and offshore waters, the
freshness of the coastal waters maintains their
lower density. The timing of floods and stormy
weather is usually such that just as spring runoff
is peaking, storminess is diminishing (Dinnel
and Wiseman, 1986; DiMego et al., 1976).
Thus, mechanical stirring and mixing of the
water column by the wind is diminishing and
maximum stratification establishes itself.
The distribution of stratification is modulated
by other processes besides mixing. Winds cause
currents which result in waters flowing towards
or away from the coast, as well as parallel to the
coast (Grout, 1982, Dagg, 1988). During 1986,
a transect of stations was occupied 17 times
across the inner shelf offshore of Cocodrie
(Figure 29). Well mixed conditions were
observed during low-runoff and high wind
conditions at the beginning of the year. As the
discharge increased and wind stirring
diminished, stratification developed. Even
under low winds, though, the stratification
strength varied significantly due to upwelling
and downwelling (Wiseman et al., in review).
58
-------
During the summer, a secondary density
interface developed near the bottom. Some
years this second density interface is absent
from our data. When it is present, it can be
either due to salinity, due .to temperature, or due
to both. It is weaker than the main density
interface, but very important to the distribution
of hypoxia.
While water column stratification may be
moderately constant for extended periods of
time, strong currents may still be flowing
through the region. These occur on a variety of
time scales. Tides in the Gulf of Mexico are
weak and so are the tidal currents (Science
Applications International Corporation, 1989).
Nevertheless, tidal currents stir the waters and
interact frictionally with the bottom (Dinnel,
1988). Other currents also occur on similar
time scales: inertial oscillations (Daddio et al.,
1976) and currents driven by the sea-breeze
system. On much longer scales, of the order of
many weeks, the density gradients resulting
from the spatial distribution of light, fresh water
and heavy, salty water are associated with
geostrophic currents. The observed shears due
to these low frequency currents are very similar
to those expected from theoretical
considerations (Wiseman et al., in review). The
most important current variability occurs on
time scales of a few days to a few weeks (Grout
et al., 1984; Chuang and Wiseman, 1983;
Science Applications International Corporation,
1989; Wiseman and Kelly, 1994). These
fluctuations are driven by wind forcing. They
are also the strongest currents generally
observed over the inner shelf. Flow reversals
may occur in less than a few hours. Thus,
cruise data that takes a week to collect may be
sampling totally different flow conditions at the
beginning and the end of the cruise. This
makes data interpretation difficult. If one sees
spatial variability, it is not clear whether this is
the result of local processes or advection.
Strong, persistent seasonal stratification is a
necessary condition for the occurrence of
hypoxia (Figure 30); it is not clear that this is a
sufficient condition. Furthermore, this strong
stratification does not necessarily determine the
structure of the hypoxic water mass. The
oxygen sinks in the system are concentrated in
the near-bottom regions of the water column.
Weak near-bottom density interfaces confine
the low oxygen waters near the bottom
(Figure 31).
In summary, the physical processes active over
the Louisiana inner shelf are important to the
dynamics of hypoxia in the region. Mixing of
river effluent with ambient shelf waters occurs
rapidly after discharge. These waters then flow,
generally, westward along the Louisiana coast
carrying with them dissolved and suspended
material from the rivers. The density structure
of the waters within this flow are intimately tied
to the occurrence, persistence, and structure of
hypoxia. While the principal determinant of
this density structure is the discharge from the
Mississippi-Atchafalaya River system, winds and
solar heating also modulate the stratification.
There remain numerous open questions:
• How much of the observed variability is the
result of local mixing (and biological
processes) as opposed to advection?
• Wh^t processes are responsible for cross-
frontal exchange in the Louisiana Coastal
Current?
• How are the processes associated with and
influenced by the Atchafalaya River
discharge different from those near the
Mississippi Delta discharge?
• What processes control the secondary
density structures observed in the waters of
the Louisiana Coastal Current?
59
-------
Acknowledgments
This work has been supported by NOAA's
Coastal Ocean Program.
References
Chuang, W.-S. and Wm. J. Wiseman, Jr.. 1983.
Coastal sea level response to frontal
passages on the Louisiana-Texas shelf. J.
Geophys. Res., 88(C4):2615-2620.
Grout, R. L.. 1982. Wind-driven, near-bottom
currents over the west Louisiana inner
continental shelf. Ph. D. dissertation,
Department of Marine Sciences, Louisiana
State University, Baton Rouge, LA. 117pp..
Crout, R. L., Wm. J. Wiseman, Jr. and W.-S.
Chuang. 1984. Variability of wind-driven
currents, west Louisiana inner continental
shelf: 1978-1979. Contributions in Marine
Science. 27:1-11.
Daddio, E., Wm. J. Wiseman, Jr., and S. P.
Murray. 1978. Inertial currents over the
inner shelf near SON. J. Phys. Oceanogr.,
8(4):728-733.
Dagg, M.J.. 1988. Physical and biological
responses to die passage of a winter storm
in the coastal and inner shelf waters of the
northern Gulf of Mexico. Cont. Shelf Res.,
8(2):167-178.
DiMego, G. J., L. F. Bosart, and G. W.
Endersen. 1976. An examination of the
frequency and mean conditions surrounding
frontal incursions into the Gulf of Mexico
and Caribbean Sea. Monthly Weather
Review, 104:709-718.
Dinnel, S. P.. 1988. Circulation and sediment
dispersal on the Louisiana-Mississippi-
Alabama continental shelf. Ph. D.
dissertation, Department of Marine
Sciences, Louisiana State University, Baton
Rouge, LA. 173pp..
Dinnel, S. P. and Wm. J. Wiseman, Jr.. 1986.
Fresh water on the Louisiana and Texas
shelves. Cont. Shelf Res., 6(6):765-784.
Hitchcock, G. L., W. J. Wiseman, Jr., W. C.
Boicourt, A. J. Mariano, N. D. Walker, T. A.
Nelsen and E. Ryan, in review. Property
fields in an effluent plume of the Mississippi
River. J. Mar. Sys..
Rabalais, N. N., R. E. Turner, Q. Dortch, Wm,
J. Wiseman, Jr., and B. Sen Gupta, in press.
Nutrient changes in the Mississippi River
and system responses on the adjacent
continental shelf. Estuaries.
Science Applications International Corporation.
1989. Gulf of Mexico physical
oceanography program, Final report: Year 5.
Volume II: Technical report. OCS
Report/MMS - 89-0068, U. S. Department
of the Interior, Minerals Management
Service, Gulf of Mexico OCS Regional
Office, New Orleans, LA. 333 pp..
Wang, F. C.. 1984. The dynamics of a river-
bay-delta system. J. Geophys. Res.,
89(C5):8054-8060.
Wiseman, Wm. J., Jr. and F. J. Kelly, 1994.
Salinity variability within the Louisiana
Coastal Current during the 1982 flood
season. Estuaries, 17(4):732-739.
Wiseman, Wm. J., Jr., S. P. Murray, J. M. Bane,
and M. W. Tubman. 1982. Physical
environment of the Louisiana Bight.
Contributions in Marine Science, 25:109-
120.
Wiseman, Wm. J., Jr., N. N. Rabalais, R. E.
Turner, S. P. Dinnel and A. MacNaughton.
in review. Seasonal and interannual variabil-
ity within the Louisiana Coastal Current:
stratification and hypoxia: J. Mar. Sys..
Wright, L. D. and J. M. Coleman, 1971.
Effluent expansion and interfacial mixing in
the presence of a salt wedge, Mississippi
River delta. J. Geophys. Res., 76:8649-8661.
60
-------
STATION C6
10 15
SALINITY DIFFERENCE
Figure 29.
Time series of salinity along an inner shelf transect offshore ofCocodrie, LA in
1986 (upper), time series of the associated Brunt-Vaisala period (middle), and
time series of river discharge for 1986.
E'*r iff- fUt VEAK
Figure 30.
Scatter plot of surface to bottom oxygen difference versus surface to bottom
density difference from multiple occupations of the same station in 20 meters
of water offshore ofCocodrie, LA in all seasons of the year. The straight line is
a least squares fit to the data.
61
-------
10 15 20 25 30 35 40
Depth of deepest density gradient > .015/m [m]
45
50
Figure 31.
Scatter plot of the depth of the 2 mg/l dissolved oxygen surface versus the
depth of the deepest point where the density gradient exceeds 0.015 /m for 9
years of mid-summer cruise data from the west Louisiana inner shelf. The solid
line has a slope of unity.
Presentation Discussion
William J. Wiseman (Louisiana State
University—Baton Rouge, LA)
Len Bahr (Louisiana Governor's Office-
Baton Rouge, LA) asked William Wiseman to
discuss die mechanical impacts of significandy
shortening the southern-most tributary of the
Mississippi river and releasing the nutrient
enriched water in an uncontrolled fashion into
either the Barataria Basin or the Berdn Sound
area, which are much shallower and less
susceptible to stratification.
William Wiseman responded by saying that if
the resulting stratification were reduced, the
flow may short-circuit into deep water fairly
quickly; particularly if it flowed through
Barataria Bay, where there is a rapid increase in
depth close to the shore. If there were a
reduction in the salinity deficit of the water
flowing onto the shelf, then the stratification of
the Louisiana coastal current would be reduced
and the bottom waters would be more easily re-
oxygenated by weaker winds. If the River were
allowed to flow past Morgan City, Louisiana
and out through the Atchafalaya, the hypoxia
conditions east of the Atchafalaya Delta would
certainly be improved, and there would not be a
major freshwater cap on top of that water. This
process could potentially move the problem
downstream into Texas waters.
Paul LaViolette (Gulf Weather Corpora-
tion— Stennis Space Center, MS) asked how
satellite data would be used to help define the
area of hypoxia.
62
-------
William Wiseman did not know how satellite
data would be used to define the hypoxia area,
though he said it could be used to track plumes,
plume dynamics, and occasionally the inner
coastal current to analyze dynamics and mixing
characteristics. There is a possibility the data
could be used to gain an understanding of the
distribution of phytoplankton in the surface
layers which may indicate where the carbon
source will settle.
63
-------
Trends in Shrimp CatcOT
Northern Gulf of Mexico
Roger Zimmerman, James Nance, and JoXli|n
.National Marine Fisheries Service i;'s':'1ifl
iGafveston Laboratory ; sfs||
!Ga/veston, Texas ,,;i :y|
Introduction
An effect of hypoxia on shrimp landings is
expected, both through reducing catch in
areas of high hypoxia and concentrating
catches in adjacent areas. Investigations of
seasonal hypoxia in Louisiana offshore reveal that
infauna are killed and fish and shrimp are often
sparse or absent (Rabalais and Harper, 1991 and
1992; Renaud, 1986). Comparison of hypoxic
areas in the New York Bight and the northern
Gulf of Mexico indicate similar reductions in
abundance of infauna (Boesch and Rabalais,
1991). Although infauna typically recover during
months without hypoxia, the community remains
in an early successional state because of mortali-
ties during the summer every year (Boesch and
Rabalais, 1991). The affected area off of Louisiana
is large, covering up to 9500 km2 (Rabalais et al.,
1991), which coincides with historical white
shrimp (Pe/taenssetifems'} and brown shrimp
(P. asgectis) fishing grounds (Lindner and
Anderson, 1956; Christmas and Etzold, 1977).
These shrimp rely upon benthic infaunal foods as
the mainstay of their diets (McTigue and
Zimmerman, 1991).
The National Marine Fisheries Service has a
database on monthly shrimp landing statistics for
the Gulf of Mexico going back to 1960. The
database is used to follow shrimp landings trends
and estimate shrimp trawling effort for manage-
ment of penaeid shrimp resources in the federal
Exclusive Economic Zone (EEZ). Subareas of
reported landings and effort include historical
areas of hypoxia and thus may be useful in com-
parisons of interrelationships. However, it must
be recognized that the shrimp statistics database
was not designed for detecting effects of hypoxia
and reported subareas of landings may be undesir-
ably large for ideal analysis. Notwithstanding this
shortcoming, the number of data entries are rela-
tively large and cover many years including the
past decade when the area of hypoxia has been
measured annually. With retrospective analysis,
we may be able to observe trends that suggest
relationships between shrimp landings or effort
and the annual extent of hypoxia.
The data on shrimp landings are gathered by 21 port
agents in major fishing ports from Key West to
Brownsville (Figure 32; Poffenberger, 1991). These
port agents canvas 450-500 dealers each month,
record landings (a mandatory reporting requirement
for the dealers) and assign landings to areas in a
statistical grid system designed for the Gulf of Mexico
(Figure 33). The statistical subareas are numbered and
subdivided into depth zones in 5 fathom increments
out to 25 fathoms and landings data is entered for
each subarea by depth zone (Figure 34). This census
of dealers provides information on the size of catch
and the number of trips by area and depth. Some
major ports for shrimp landings occur between
Freeport, Texas and New Orleans, Louisiana and, not
64
-------
surprisingly, prominent shrimping subareas overlap
with areas of seasonal hypoxia in the northwestern
Gulf. In particular, statistical sub-areas 13, 14, 15, 16,
and 17 incorporate waters offshore of Louisiana and
uppermost Texas where hypoxia has been
documented.
In order to establish catch-per-unit effort (CPUE),
the port agents interview shrimp fishermen,
collecting information on trip duration, time fished,
and location fished (Figure 35). Since this reporting
is not mandatory, the reliability of these data is
dependent upon access to cooperative fishermen. A
simple equation incorporating interview data and
landings data is used to calculate shrimp fishing
effort. Landings from the dealers divided by the
CPUE from sample interviews is equivalent to
effort, which is usually reported in days fished
(Figure 36). CPUE is estimated in instances where
landings have no associated interviews. At present,
about 70% to 80% of the shrimp pounds landed
have CPUE interview data and the other 20% to
30% is estimated (Figure 37 and 38).
To evaluate relationships between hypoxia, size-
of-catch and CPUE, respective data in each
statistical cell off the coast of Louisiana was
calculated. A cell corresponds a depth zone with a
subarea. For example, the cell closest to shore in
subarea 17 is zero to 5 fathoms; the next cell
offshore is 5 to 10 fathoms, and so on. Louisiana
subareas are demarked longitudinally and depth
zones are roughly latitudinal. We entered this
information into our geographic information
system program (GIS) and color-coded catch to
represent average monthly pounds of shrimp
landed during July and August each year. The
color scale grades from gray, representing very
little or no catch, to dark red, representing more
than 600,000 pounds of shrimp tails. Intermediate
values were represented by shades of orange. The
average July and August catch in each statistical
cell, months of high hypoxia, was determined for
each year between 1985 and 1994 and entered
into the GIS. Annual area of hypoxia was plotted
from data reported by Rabalais et al. (1991, 1992
and unpublished). The hypoxic area was entered
in the GIS, calculated for each statistical cell and
superimposed on the image of mean July and
August catch for each year. A step-wise regression
with catch as the dependent variable and depth,
subarea, East/West, years, and percent area of
hypoxia in cells as independent variables was
performed.
Results and Discussion
Hypoxia and July/August shrimp catch are
depicted in Figures 39 through 45, presenting the
years 1985,1986, 1990, 1991, 1992, 1993 and
1994. Offshore hypoxia developed to greater or
lesser extent every summer during this ten year
period. A number of relationships were evident.
Shrimp catch nearshore was always significantly
higher thkn catch offshore regardless of the
extent of; hypoxia. In addition, a significant
relationship in catch between subareas occurred
from West to East across years. Catches were
significantly higher near the Texas-Louisiana
border compared to the Mississippi delta. This
corresponds to the generalized distribution of
hypoxia which is usually greater in the eastern
sector compared to further west. But during years
when area of hypoxia was large in the west,
shrimp catch was diminished there too. Area and
distribution of hypoxia between consecutive years
often varied significantly as did overall shrimp
catch. The year 1990, with less hypoxia, had
higher catch than 1991, with greater hypoxia,
including a large hypoxic area which coincided
with a reduction in catch in the nearshore
5 fathom zone. During 1993, with hypoxia
broadly distributed along the coastline, catch was
high nearshore and markedly lower offshore.
However, this relationship was not as strong
during a similar large hypoxic event during 1994.
The magnitude of catch (strength of the year
65
-------
class) was different between 1993 and 1994, but
the pattern of catch distribution related to
hypoxia remained similar.
Catch plotted against the percent area of hypoxia in
each cell reveals a significant negative relationship.
Importantly, CPUE plotted against percent area of
hypoxia in cells demonstrates that CPUE does not
change relative to area of hypoxia (Figure 46). The
regression of catch against hypoxia is affected by an
interaction of a significant relationship between
catch and depth. The highest catches always occur
nearshore in cells with a very low percentage of
hypoxia area. In cells with a very high percentage of
hypoxia, we never observe high catch despite similar
CPUEs. This is interpreted as meaning that shrimp
fishermen do not trawl in those areas within a cell
that have high hypoxia. Converting the data to catch
per unit hectare demonstrates the same relationship
(Figure 47). However, we also observed low catches
in offshore cells with low hypoxia, i.e., the
relationship of diminution in catch from nearshore
to offshore corresponding to increasing depth.
There was a negative, albeit not always significant,
relationship between shrimp catch and area of
hypoxia. This may have very important historical
implications for the Louisiana shrimp fishery. The
traditional inshore and nearshore fishery appears
to be promoted and the offshore fishery is dis-
couraged by hypoxia. Catches in offshore waters
beyond the hypoxic area are always as low as
those in the hypoxic area. High catches nearshore
are always in cells with low hypoxia. The interpre-
tation is that the large hypoxic area in inter-
mediate depth zones concentrates shrimp near-
shore. This is supported by laboratory evidence
that shrimp move away from low oxygen water
(Renaud, 1986a) and field evidence of low den-
sities (Renaud, 1986b). Moreover, the phenomena
of concentration of nekton avoiding hypoxia in
other areas, the so-called jubilees in Mobile Bay,
has been known for years (Loesch, 1960). Since
shrimp seem to avoid hypoxia, the hypoxic area
would effectively block a large part of the popu-
lation from moving offshore. This blocking
phenomena may in part explain persistent low
catches in offshore Louisiana beyond the hypoxic
zone. Moreover, since CPUE doesn't change
relative to percent hypoxia in statistical cells, we
take this as evidence that shrimpers do not trawl
in unproductive waters, and that waters offshore
of the hypoxic zone are indeed unproductive.
More simply stated, it is economic reality that as
the shrimp move so do the shrimp fishermen and
those who do not catch anything in their trawls
move on in order to profit. Thus, wherever
shrimp fishermen chose to stay and trawl their
CPUE is relatively high. The statistical cells with
low catch associated with hypoxia and offshore
waters beyond mean that shrimpers are actively
avoiding these areas. By contrast, higher catches
occur in comparable offshore depth contours in
Texas where hypoxia does not exist. Indeed,
Texas has a very well developed offshore shrimp
fishery. An alternative hypothesis is that the
shrimp industry in Louisiana developed around
white shrimp which is an inshore and nearshore
species, and Louisianans did not build vessels big-
enough to trawl offshore whereas, the Texas
shrimp industry developed around brown shrimp
which is more an offshore species.
Although the landings data are coarse for the
purpose and analyses would benefit from a speci-
fically designed study, evidence of a negative
relationship between hypoxia and shrimp catch
appears to exist. Overall, offshore areas of
extensive hypoxia during the summer months
yield lower shrimp catches in July and August than
nearshore areas with less hypoxia. Importantly, the
shrimp appear to concentrate in shallow waters
near shore between the hypoxic zone and the
shoreline and the effect of diminished catch
extends offshore well beyond the hypoxic
zone.
66
-------
References "
«• _, ,„ _, '_, , _f _ , «. ;_ „ . f".
Boesch, D. F. and N. N. Rabalais 1991. Effects of
hypoxia on continental shelf benthos:
comparisons between the New York Bight
and the northern Gulf of Mexico, pp. 27-34.
In: R. V. Tyson and T. H. Pearson (eds.)
Modern and Ancient Continental Shelf Anoxia.
Geological Society Special Publication.
No. 58.
Christmas,]. Y. and D. J. Etzold 1977. The
shrimp fishery of the Gulf of Mexico: a
regional management plan. Gulf Coast
Research Laboratory, Tech. Rep. Ser. 2,
Ocean Springs, Mississippi, 128 pages.
Lindner, M. J. and W. W. Anderson 1956.
Growth, migration, spawning and size
distribution of shrimp, Penaeus setiferus." U.S.
Fish and Wildlife Service, fishery Bulletin 56:553-
645.
Loesch, H., 1960. Sporadic mass shoreward
migrations of demersal fish and crustaceans in
Mobile Bay, Alabama. Ecology 41: 292-298.
McTigue, T. A,, and R. J. Zimmerman 1991.
Carnivory vs. herbivory in juvenile Penaeus
sett/ems (Linnaeus) and Penaeus a^tecus (Tves).
Journal of Experimental Marine Biology and Ecology
151: 1-16.
Poffenberger, J. R., 1991. An overview of the data
collection procedures for the shrimp fisheries
in the Gulf of Mexico. Southeast Fisheries
Science Center Report, Miami. 20 pages.
Rabalais, N. N., R. E. Turner, W. J. Wiseman, Jr.,
and D. F. Boesch 1991. A brief summary of
hypoxia on the northern Gulf of Mexico
continental shelf: 1985-1988. pp. 35-47. In:
R.V. Tyson and T.H. Pearson (eds.) Modern
and Ancient Continental Shelf Anoxia. Geological
Society Special Publication No. 58.
Rabalais, N. N., and D. E. Harper, Jr 1991.
Studies of the benthic oxygen depletion
phenomenon off Louisiana.pp. 57-63. In: H.
J. Krock and D. E. Harper, Jr.(eds).
International Pacifica Scientific Diving...1991.
Proceedings of the American Academy of
Underwater Sciences Eleventh Annual
Scientific Diving Symposium.
Rabalais, N. N., and D. E. Harper, Jr. 1992.
Studies of benthic biota in areas affected by
moderate and severe hypoxia. pp. 150-153. In:
Nutrient Enhanced Coastal Ocean
Productivity. Proceedings of a NOAA
Workshop, Louisiana Universities Marine
Consortium, Sea Grant Program, Texas A&M
University, Publication TAMU-SG-92-109,
June 1992.
Rabalais, N. N., R. E. Turner, and W. J. Wiseman
Jr., 1992. Distribution and characteristics of
hypoxia on the Louisiana shelf in 1990 and
1991. In: Nutrient Enhanced Coastal Ocean
Productivity, Proceedings of a NOAA
Workshop, Louisiana Universities Marine
Consortium, Sea Grant Program, Texas A&M
University, Publication TAMU-SG-92-109,
June 1992.
Renaud, M. L., 1986a. Detecting and avoiding
oxygen deficient sea water by brown shrimp,
Penaeus a^tecus (Tves), and white shrimp Penaeus
setiferus (Linnaeus). Journal of Experimental
Marine Biology and Ecology 98: 283-292.
Renaud, M. L., 1986b. Hypoxia in Louisiana
coastal waters during 1983: implications for
fisheries. Fishery Bulletin 84(l):19-26.
67
-------
Port Arthur
New Orleans
Pascagoula
Bayou LaBatre
L
{• Port Isabel
&Brownsville
Panama City
Golden Meadow
Houma St. Petersburg
New Iberia Fort Myers
Key West
Figure 32.
Port Agents.
Landings Data
Canvass between 450-500 dealers each month
Record landings by trip
Assign grid zones
A census of catch and trips
Figure 33.
Statistical Areas far Reporting Shrimp Catch
Figure 34.
68
-------
CPUE Data
Interview
Data Items
- Trip duration
- Time actually fished
- Area fished
A census of catch and trips
Figure 35.
Time Fished
(Present Method)
_ Landings
Effort - CPUE
Sampled
Censused
Figure 36.
Cells With Current Interview Data
70-80% of the shrimp pounds have an average
CPUE associated with them
What about the other 20-30% of the pounds?
Figure 37.
69
-------
«*
Statistical model used to
estimate the current CPUE (one
model/month)
log CPUE(ij) =
location
ij) + year(i)
G)
Figure 38.
t>B )
<»
106*11
figunt 39.
Toto/ Shrimp Catch (pounds)
JulylAugust 1985
70
-------
Total Shrimp Catch (pounds)
July/August 1986
Total Shrimp Catch (pounds)
julylAugust 1990
71
-------
Figure 42.
Total Shrimp Catch (pounds)
JulylAugust 1991
figure 43.
Total Shrimp Catch (pounds)
July/August 1992
72
-------
.WE Mi
Figure 44.
Totaf Shrimp Catch (pounds)
July/August 1993
w
figure 45.
Total Shrimp Catch (pounds)
July!August 1994
-------
20001
UJ
3
Q.
o
1000-
o
4.000+6-
3.000+6-
2.008+6-
1.000+6-
O.OOe+0
y = 563.60 - 0.10154x RA2 = 0.000
• • ••
*•*». •
#V.-. .::
••• .•
y = 6.4827e+5-5586.6X RA2 = (
20 40 60 80 100
Percent Hypoxia
20 40 60 80 100
Percent Hypoxia
Figure 46.
0.003
2.
8 0.002 -
o
Q)
X
o 0.001 -
(0
o
0.000
'•••
20 40 60
Percent Hypoxia
Figure 47.
80 100
74
-------
-=" ^,-x™- ~ ^^ Jx~xz.~ .,,. --^sr ^jg-V-jj- -^~- -^ —^~ .( — ~ <-
^ JP.esjBDLtatioji. Discussion _ .
Roger Zimmerman (NMFS—Galveston, TX)
Bob Anderson (The Advocate—Baton Rouge,
LA) asked Roger Zimmerman if anyone had looked
at the total shrimp catch per year during the years
when there is a large hypoxia zone present.
Roger Zimmerman answered that although there
appeared to be a very weak relationship between
years, when he performed a step-wise regressional
analysis, he could not demonstrate a strong
relationship.
Eddie Funderberg (Louisiana State University
Agricultural Center—Baton Rouge, LA)
commented that the shrimp catch in cells of
Zimmerman's map with 80 to 100 percent hypoxia
appeared to be as good as in cells of zero percent
hypoxia. In light of that data, he asked Roger
Zimmerman to elaborate on his comment that the
catch per unit effort (CPUE) did not vary because
fishermen were not fishing in the area of hypoxia.
Roger Zimmerman responded that even in cells of
100 percent hypoxia there may be areas or times
that the hypoxia fluctuates. If the hypoxia is absent
and shrimp, which are very mobile move in, then it
is possible to have a catch in those cells that have
been identified as 100 percent hypoxic. Also, some
of those cells are only 60 percent hypoxic. The
shrimping activity in that cell is averaged over the
whole cell. Unfortunately, there is no ability to
separate within a cell.
Don Boesch (University of Maryland— Cam-
bridge, MD) asked if the inability to catch shrimp
in hypoxic waters was a result of a reduction in the
shrimp population, or a decreased ability to catch
that population. He also asked if Roger Zimme-
rman had an approach or strategy to study this
question with either existing data or new
observations.
Roger Zimmerman felt that there were two
potential possibilities.
• The first possibility was a correlation between
catch data and depth. That relationship could be
analyzed by comparing shrimp catches in the
1960's to the nearshore and offshore abundances
of shrimp. Hopefully, that would demonstrate
that during periods of hypoxia there would be a
reduced number of shrimp or a lower percentage
of catch relative to near shore.
• The second, and most likely possibility is that the
hypoxic zone could be causing the shrimp to
migrate up against the shoreline.
William Wiseman (Louisiana State
University—Baton Rouge, LA) asked if the data
presented had been normalized. For example,
shrimping activity in Louisiana, is basically localized
in the nearshore. It is not considered an offshore
fishery. Therefore, comparing the data to shrimping
activity in Texas may be inconclusive, because the
effort may be distributed differently.
From the research he has conducted on shrimp
populations in Louisiana, environmental factors in
the spring, (i.e., water temperatures and salinity
regimes) have a great deal of influence on what
production is going to be and what kind of
recruitment we have in late spring and early
summer. The fact that hypoxia does not really set
up offshore during the spawning and larval
migration periods inshore does not seem to be an
influence. The ultimate growth and survival of
juveniles, which is really dependent upon inshore
conditions, does not seem to be affected either.
Roger Zimmerman confirmed that the data have
been normalized by effort. He agreed with William
Wiseman that the strength of the year ckss is more
dependent upon the conditions in the nursery and
inshore than it is in the offshore conditions. He
thought they might be just redistributing the year
ckss. It is possible to argue that as the organisms
grow and move offshore as sub-adults, and if there
is a large hypoxic zone where all the worms are
dead, the feeding ground has obviously been
impacted. It is similar to lower salinity impacts in
the estuaries, which could reduce production in
certain parts of the nursery. It is possible that the
offshore feeding ground is being eliminated and
that could affect growth rates.
75
-------
Distribution, Abundance, F
of Fish Larvae Associated with t(
Discharge Plume, and the
Churchill B. Grimes
National Marine Fisheries Service
Panama City, FL 32408
Abstract
From 1986—1992 we made cruises
during all seasons to the Mississippi
River discharge plume to collect
ichthyoplankton using a neuston net (1 x 2 m;
.947 mm) and tucker trawl (1 x 1 m; 0.335 mm);
CTD casts were made to collect environmental
data. Sampling stations were positioned using
AVHRR satellite imagery (visible channel) along
15—25 km transects that radiated from the delta
outward.
The hydrography in the vicinity of the plume
consists of three distinct water masses, a shallow
lens of low salinity plume water, high salinity
Gulf of Mexico shelf water and the 6—8km wide
frontal or mixing zone between plume and shelf
waters.
Fish larvae are abundant in the vicinity of the
discharge plume in general, but are especially
concentrated in the frontal region, e.g., average
neuston catches were 6-fold higher than at
plume stations and 12-fold higher than at shelf
water stations.
The described hydrography promotes strong
hydrodynamic convergence within the frontal
zone. We used surface drifters to measure
apparent surface convergence rates at turbidity
fronts of up to 0.8 m/sec. We used an
advection diffusion model to simulate larval
densities in surface waters at the frontal
convergence zone that approximated the mean
and median observed densities within the
frontal zone.
We deployed radio tracked surface drifters and
repeatedly sampled nearby over time to deter-
mine if larvae could be retained in the vicinity of
the plume, or were advected westward by the
average surface flow off the Mississippi Delta. A
clockwise circulation with a radius of curvature
of about 50km was identified and acted to retain
larvae in the vicinity of the plume.
The diet of striped anchovy, Anchoa hepsetus,
larvae was investigated to determine if fish
larvae in the frontal zone were deriving a
trophic advantage from the potential food
resources concentrated there. The diet consisted
of a wide array of prey items including various
microcrustaceans (e.g., amphipods, cladocerans,
copepods and ostracods) diatoms, larvaceans
and polychaete larvae, but according to both
frequency of occurrence and number of prey
items, copepods and diatoms were the dom-
inant foods consumed. Diatoms occurred more
frequently and more numerous at plume and
shelf stations (43 percent and 39 percent and
72 percent and 75 percent, respectively), but
copepods were the most frequently occurring
(49 percent) and most numerous (48 percent)
prey items at frontal stations. Because cope-
pods are larger and have a slightly higher
C:N ratio striped anchovy larvae in frontal
waters may consume a more nutritious diet.
76
-------
Otolith microstracture techniques were used to
determine age and estimate growth rates to
determine if larvae in the vicinity of the plume
in general, or the frontal waters in particular,
grow faster. King mackerel, S comberomoras cavalla,
larvae from the Mississippi plume region grew
significantly faster than larvae from other areas
(0.95 vs. 0.79 mm/d), while Spanish mackerel,
S. maculatus, and little tunny, Euthynnus alletteratus,
from the plume did not. Spanish mackerel,
yellowfin tuna, Thunnus ablacares, and striped
anchovy larvae in the vicinity of the discharge
plume grew faster at intermediate salinities, i.e.,
the frontal zone (1.0 vs. 1.3, 0.75 vs. 0.6 and
1.05 vs. 0.85 mm/d, respectively). We regressed
loge of SL on age for the descending limb of
plots for these same species to estimate daily
instantaneous mortality rates. Mortality rates
were higher in the vicinity of the plume for
Spanish mackerel, little tunny, and striped
anchovy (0.6 vs. 0.3, 0.9 vs. 0.7 and 0.23 vs.
0.12, respectively).
A relative survival model
Nt = N0e-2t
where
z — daily instantaneous mortality rate
t = Lmax/growth rate (mm)
— a 25-mm size refuge
*—
was used to evaluate the advantage of faster
growth vs. the disadvantage of increased
mortality for Spanish mackerel larvae. Survival
was much more sensitive to changes in
mortality than growth, suggesting that the
specific demographics prevalent in the plume
environment may not favor survival and
recruitment in the Mississippi River discharge
plume.
J/ntrqduction ' - '/ „>//:
There is considerable circumstantial evidence
worldwide that river plumes influence the
mechanisms underlying fish production (i.e.,
growth, mortality and recruitment), recruitment
being the most important since it is the largest
contributor to variation in fish production.
'Major fisheries have been eliminated or have
declined :when river flows were controlled. For
example.! filling of the Aswan Dam began on
the Nile River in 1965 and was completed in
1969, during which time the flow was decreased
by 40 km3 yr "*, with a concomitant decline in
primary production off the delta. Egyptian
fishery catches in the Mediterranean Sea
declined from 37,800t in 1962 to 7,142t in 1976,
with an attendant decline in community struc-
ture (Debars and Lasserre, 1983). The largest
river in North America, the Mississippi, is no
exception. A major feature influencing the
ocean environment of the Gulf of Mexico—it
annually1 discharges an average 1.83 x 10 m S~
(Gunter, 1979) of freshwater, nutrients, and
suspended materials. Fishery landings from the
fertile fishery crescent surrounding the Missis-
sippi River delta are extraordinary, accounting
for approximately 80 percent of the total com-
mercial landings from the Gulf of Mexico
(NMFS, 1994).
There is, concern about how the periodic
occurrence of a major hypoxia zone off the
Mississippi River may influence the valuable
fisheries associated with the river. The purpose
of this brief report is to summarize the results
of research on the recruitment dynamics, i.e.,
abundance, feeding, growth, and mortality, of
fish larvae associated with the Mississippi River
discharge plume, and discuss the possible
significance of these results to hypoxia and its
potential impact on fish production.
77
-------
; Results And Discussion
From 1986—1993 we conducted research cruises
during both low (summer-fall) and high (spring)
flow regimes to the Mississippi River discharge
plume. Plankton samples were collected at
stations 4—6 km apart along 15—25 km transects
that radiated out from the delta, and were
positioned using AVHRR satellite imagery
(visible channel) to cross from the plume into
Gulf of Mexico shelf waters. Plankton was
collected with neuston net (1 x 2 m; 0.947 mm)
and Tucker trawl (1 x 1 m; 0.335 mm); CTD
casts were made to obtain environmental data at
each station.
The water column in the vicinity of the dis-
charge plume has a characteristic hydrographic
structure created by the abutment of water
masses with distinctly different densities
(Grimes and Finucane, 1991; Govoni and
Grimes, 1992). Lighter plume waters are
represented by a shallow lens of low salinity
water overlying heavier high salinity Gulf of
Mexico shelf waters; the 6-8km wide frontal, or
mixing, zone between these two water masses is
where isohalines are closely spaced and
approach the surface (Figure 48). Turbidity
fronts, represented by sharp color discon-
tinuities, are the seaward projection of concen-
trated suspended particulate matter, and they are
often nested within the frontal zone (Garvine
and Monk, 1974).
Phytoplankton, zooplankton and fish larvae are
concentrated in the vicinity of the plume in
general and the frontal region in particular. For
example, average surface phytoplankton bio-
mass, macrozooplankton displacement volume
and neustonic ichthyoplankton catch rates were
about 4, 2 and 6 fold greater in frontal waters
than in adjacent plume and shelf waters (Grimes
and Finucane, 1991; see also Govoni eta/., 1989
and Govoni and Grimes, 1992).
Surface waters converge at plume fronts, pri-
marily due to strong horizontal density gradients
and resulting pressure gradients that are pro-
duced within and below the frontal layer. Cross
frontal circulation is characterized by vigorous
convergence on both sides of the front, typically
higher on the high density (seawater) side than
on the low density (plume) side, e.g., average
0.2 and 0.1 m sec for the Mississippi River
plume (Govoni and Grimes, 1992). As surface
waters converge, planktonic organisms move
passively with the water toward the front where
converging water masses move downward with
gravity. Surface seeking and buoyant organisms
accumulate at the surface as they resist down-
ward movement. This is local, but important,
transport mechanism that can concentrate larval
fish and zooplankton and account for the high
densities of these properties observed at fronts.
Govoni and Grimes (1992) measured surface
convergence velocity in the Mississippi River up
to 0.8 m sec"1. Observed velocity was always
greater than the velocity calculated from the
density alone (Figure 49) because the observed
velocity is the sum of the density driven velocity
plus the tidally driven velocity inherent in shelf
waters. They used the advection diffusion
model of (Olson and Backus (1985) to simulate
surface densities of fish larvae at the front that
agreed well with observed values (Figure 50).
Having observed the distribution and abun-
dance of these properties, as well as the hydro-
graphic structure and hydrodynamics Grimes
and Finucane (1991) developed a modification
of the short food chain hypothesis explaining
how the Mississippi River might act to enhance
recruitment of associated fish larvae. This hypo-
thesis states that fish larvae concentrated in the
vicinity of the plume in general, and the frontal
region in particular, would take advantage of
abundant food resources and consume a
superior diet, grow faster and thus experience a
shorter larval stage duration and survive better.
Implicit in this hypothesis is that fish larvae in
78
-------
the vicinity of the discharge plume are not
advected away from the rich plume environ-
ment by the low average westward flowing
surface currents that prevail off the Mississippi
River delta during the period of interest
(summer-fall) (Wiseman and Dinnel, 1988).
Radio tracked surface drifters were deployed
and repeatedly sampled nearby over time to
determine if fish larvae were retained in the
vicinity of the plume or were advected away
(Grimes and Wiseman unpublished). The
surface drifters were entrained in a tongue of
Gulf of Mexico shelf water that intruded into
the Louisiana Bight and rotated clockwise with a
radius of curvature of about 50km (Figure 51).
The average variation in abundance of fish
larvae in surface collections suggested that the
same assemblage of fish larvae was repeatedly
sampled nearby the surface drifters because the
coefficient of variation in total abundance along
drifter tracks was two to four fold less than the
variation among samples collected along tran-
sects that intentionally crossed plume, front and
shelf waters (Table 3). These results suggest
that, at least in this instance, a clockwise
circulation existed in the vicinity of the plume
that acted to retain fish larvae.
Research findings thus far are not totally in
accord with the first element of the short food
chain hypothesis, i.e., it cannot be stated
unequivocally that fish larvae associated with
the Mississippi River plume are conferred a
trophic advantage. Spot, Leiostomus xanthurus,
larvae collected off the Mississippi River plume
ate twice as many food organisms as did larvae
in Gulf of Mexico shelf waters (Govoni and
Chester, 1990). However, organisms within the
plume were mostly small (tintinnids, copepod
nauplii, pelecypod veligers and invertebrate
eggs), whereas organisms eaten in shelf waters
were larger (copepodites and adult copepods).
Because the volume and nutritional quality of
gut contents of larvae from the two areas were
roughly equivalent, they concluded that larvae in
the plume gained no trophic advantage.
Similarly, Powell etal. (1990) used morpho-
logical, gut content and recent growth criteria to
evaluate nutritional condition of spot larvae
associated with the Mississippi discharge, and
could not consistently demonstrate an advan-
tage. A diet study on striped anchovy, Anchoa
hepsetus, collected along transects crossing
plume, front and shelf waters showed that
diatoms and copepods were by far the dominant
food items, and that the larger more nutritious
copepods occurred more frequently and
accounted for the highest percentage of food
items in guts of larvae collected in frontal
waters, followed by plume waters then shelf
waters (McNeil and Grimes, 1995). A suite of
biochemical indices to nutritional condition
(RNA/DNA ratio, percent protein and CS and
LDH enzyme systems), were examined on
striped anchovy collected along the same
transects off the Mississippi plume; larvae
collected in frontal waters were in the highest
nutritional conditions (Torres et al. unpub-
lished). Furthermore, in a recent review of the
influence of riverine plumes worldwide on fish
larvae Grimes and Kings ford (in press) found
that certain taxa, e.g., small opportunistic
species, appear to be associated with plumes
and may be better adapted than larger more
competent larvae of other species to take
advantage of abundant food resources around
plumes and their fronts.
The second element of the hypothesis states
that fish larvae that are conferred a trophic
advantage will respond by growing faster, and
there is isome evidence that growth of some fish
larvae, as determined from otolith
microstructure, may be enhanced. Growth of
king mackerel, Scomberomorus cavalla, was higher
off the Mississippi River plume (0.95 mm d "*)
than at other locations in the Gulf of Mexico
(0.79 mm d'1) (DeVries etal., 1990). However,
superior growth off the plume was not
79
-------
demonstrated for Spanish mackerel, S. maculatus,
(DeVries et a/., 1990), or little tunny, Euthynnus
alleteratits, (Allman and Grimes unpublished).
Other results on Spanish mackerel, (Grimes and
De Vries unpublished, Figure 52) as well as
those on yellowfin tuna, Thyntms albacares, (Lang
et a/., 1994) and striped anchovy, Anchoa hepsetus,
(Day, 1993), suggest that larvae associated with
the Mississippi plume grow faster at inter-
mediate salinities, i.e., frontal waters (0.6 vs.
0.75 mm d , respectively for yellowfin tuna and
stripped anchovy).
The final element of the hypothesis states that
faster growth leads to shorter duration of the
larval stage and better survival, with the caveat
that the same dynamics that concentrate prey of
fish larvae might also concentrate their
predators. There is little evidence to evaluate
this element of the hypothesis. Grimes and
DeVries (unpublished) estimated instantaneous
rates of natural mortality for Spanish mackerel
and king mackerel using a catch-curve approach
(i.e., regressing the log of frequency on age of
the descending limb of age-frequency
histograms). Instantaneous natural mortality
estimates were approximately 0.3d "1 away from
the plume and 0.6d-l or higher in the vicinity of
the plume. For little tunny, instantaneous
natural mortality was slighdy higher in the
vicinity of the Mississippi River plume (0.94d"1)
than in the Gulf of Mexico off Panama City,
Florida (O.SSd'1) (Allman and Grimes
unpublished). Similar analyses for striped
anchovy in water masses off the Mississippi
River suggest that natural mortality in the front
(0.13d"1) and plume (0.23d'1) may be higher
than that experienced in shelf waters (0.09d-
(Day, 1993). Conversely, yellowfin tuna (Lang et
a/., 1994) experience higher natural mortality at
fronts (0.41 d"1) than in the plume area in
general (O.^d"1). These differences in mortality
rates should be interpreted with caution.
Application of catch curve or survivorship
analysis to estimate instantaneous mortality rates
assumes equal vulnerability to capture by the
sampling gear for all ages used in the analysis.
Faster growth rates might lead to biased
mortality estimates because fast growing larger
larvae become less vulnerable to capture and
may be under represented at the older ages used
in the analysis. Although these results are
tentative they do not support the contention
that higher growth rates of larvae associated
with river plumes lead to better survival.
To summarize the results of evaluating the
elements of the short food chain hypothesis
with respect to the Mississippi River, it appears
that some species of fish larvae, opportunistic
ones, are able to take advantage of abundant
prey resources. Also, some species of fish larvae
appear to grow faster, but mortality rates may
also be higher. So, whether this hypothesis is
valid and the population dynamics of fish larvae
in the vicinity of river plumes favor recruitment,
depends upon the relative magnitude of growth
and mortality. A simple and convenient way of
evaluating the relative importance of growth
and mortality is to use the expression of
exponential decay in population size with time
Nt - N0 ezt
where
Nt = population at time t
N0 = initial population
z = instantaneous mortality.
Z can be directly estimated, while t = Lc/g
(where LC = a critical size refuge where
mortality decreases markedly) and g = growth
rate that is also directly estimated. The product
of zt is an exponent that determines the
decrease in N0. Obviously, the effect of z on N0
(survival to the critical size, L,^ is much greater
than g, because z is a direct multiplier and g is a
fractional multiplier (the divisor of LJ. Thus,
incremental changes in mortality will have a
80
-------
much larger effect on survival and recruitment
than incremental changes in growth rate. So, if
physical and biological conditions in the vicinity
of the Mississippi River plumes aggregate larval
fish prey that results in a trophic advantage and
faster growth, but also aggregates predators and
increases the mortality rate on larvae, the disad-
vantage of increased mortality may well out-
weigh the advantage of faster growth, and
increased survival and recruitment will not be
the result. However, I emphasize that accurate
larval mortality rate estimates are difficult to
obtain using the time specific approach usually
taken, due mainly to sampling bias associated
with gear selectivity and the contagious distri-
bution of fish larvae in time and space.
In summarizing their review of the effects of
riverine plumes on fish larvae and their
recruitment dynamics, Grimes and Kingsford
(in press) offer two alternatives to the short
food chain hypotheses. One alternative possibly
explaining the apparent favorable effect of river
plumes on recruitment, the total larval produc-
tion hypothesis, is that trophic conditions
support such high total production of fish
larvae that negative effects of unfavorable
dynamics are overridden. That is, high primary
and secondary production associated with
plumes may simply support such high total
production of fish larvae that the specific
population dynamics at plumes are not often
relevant. A second alternative is that plumes and
associated circulations facilitate the retention of
larvae within an area. The presence of food
would of course be important, but variation in
physical retention rather than production may
explain variation in recruitment; as argued for
the member/vagrant hypothesis of Sinclair
(1988).
The relationship between hypoxia and the
recruitment dynamics of fish larvae in the
vicinity of the Mississippi River discharge is in
fact unknown, but there are several potentially
important possibilities that can be discussed.
The present understanding is that the hypoxia is
due to both the effects of stratification of fresh
and marine waters that restricts vertical reoxy-
genation of bottom waters, and the oxygen
consuming breakdown of organic material
mostly derived from high plankton production
driven by river borne nutrients. While the
hypoxia problem is believed to have been
exacerbated by nutrient enrichment of the Mis-
sissippi River, the high nutrient load of the river
and resulting high productivity associated with
the discharge area is also vital to maintaining
valuable Gulf of Mexico fisheries (e.g., approxi-
mately 80 percent of commercial fishery land-
ings are taken from the region of riverine
influence).
Hypoxia and the larvae of valuable fishery
species may sometimes co-occur in time and
space, almost certainly leading to larval mortality
when this occurs. The hypoxic zone is generally
located from off the Mississippi River delta
westward along the Louisiana coast, and
although it can occur in the winter and fall it
most consistently occurs from mid-May to
mid-September. A number of valuable species
spawn at this time in the gulf, and their larvae
are abundant off the Mississippi River delta,
e.g., both king mackerel (Grimes eta/., 1990)
and Spanish mackerel (Grimes and DeVries
unpublished), cobia (Ditty and Shaw, 1992),
dolphin (Ditty eta/., 1994) and yellowfin tuna
(Lang ef a/., 1994). Several valuable reef fishes,
e.g., gray snapper, vermilion snapper and red
snapper also spawn at that time, and while their
larvae are collected off the delta (Grimes et a/.
unpublished, Lyczkowski-Shultz unpublished,
Comyns unpublished) the adults spawn over
hard bottom and thus spawning is probably not
associated with the river discharge. Fish larvae
are most abundant in near surface waters off the
Mississippi plume (Govoni eta/., 1989), and the
hypoxia is typically associated with bottom
waters but it can extend up into the water
¥j!!a«s^^
81
-------
column. Thus, differing vertical distributions of
the hypoxic water and fish larvae may ame-
liorate potential negative impacts of hypoxia on
larvae. However, the concentration of larvae in
the frontal region may extend deeper into the
water column because the hydrodynamic con-
vergence that acts to concentrate larvae (Govoni
and Grimes, 1992) continues unabated to the
bottom on the high density (seawater) side of
the convergence zone (Govine and Monk,
1974).
References
Debars, M. I., and Lasserre, G. 1983. Analysis of
the Egyptian marine and lagoon fisheries
from 1962—1976, in relation to the construc-
tion of the Aswan Dam (completed in
1969). OceanologicaActa 6:417-26.
Day, G. R. 1993. Distribution, abundance,
growth and mortality of striped anchovy,
Anchoa hepsetits, about the discharge plume of
the Mississippi River. MS Thesis, University
of West Florida, Pensacola, Florida.
DeVries, D. A., Grimes, C. B., Lang K. C., and
White, D. B. 1990. Age and growth of king and
Spanish mackerel larvae and juveniles from the
Gulf of Mexico and U.S. south Atlantic.
Environmental Biology of 'Fishes 29: 135-43.
Ditty,]. G. and Shaw, R. L. 1992. Larval
development, distribution and ecology of
cobia, Racbjcentron canadum (Family:
Racbjcentridae) in the northern Gulf of
Mexico. Fishery Bulletin, U.S. 90: 668-677.
Ditty, J. G. Shaw, R. F., Grimes, C. B. and
Cope, J. S. 1994. Larval development,
distribution and abundance of larval dolphin
fishes, Corpbaena bippurus and C. equiselis, in
the northern Gulf of Mexico, Fishery
Bulletin, U.S. 92:275-91.
Garvine, R.W., and Monk, J.D. 1974. Frontal
structure of a river plume. Journal of
Geophysical Research 79: 2251-59.
Govoni, J.J., Hoss, D.E., and Colby, D.R. 1989.
The spatial distribution of larval fishes about
the Mississippi River Plume. Umnology and
Oceanography 34:17887.
Govoni, J.J., and Chester, AJ. 1990. Diet
composition of larval Leiostomus xanthurus in
and about the Mississippi River plume.
Journal of Plankton Research \2: 1265-76.
Govoni, J.J. and Grimes, C.B. 1992. Surface
accumulation of larval fishes by hydro-
dynamic convergence within the Mississippi
River plume front. Continental Shelf Research
12: 1265-76.
Grimes, C.B., and Finucane, J.H., Collins, L.A.,
and DeVries, D. A. 1990. Young king
- mackerel, Scomberomorccs cavalla, in the Gulf
of Mexico, a summary of the distribution
and occurrence of larvae and juveniles, and
spawning dates for Mexican juveniles.
Bulletin Marine Science 46: 640-654.
Grimes, C.B. and Kingsford, M.J. in press. How
do riverine plumes of different sizes
influence fish larvae: do they enhance
recruitment? Journal of Freshwater and
Marine Research.
Gunter, G. 1979. The annual flows of the
Mississippi River. Gulf Research Reports 6:
283-90.
Lang, K.L., Grimes, C.B., and Shaw, R.F. 1994.
Variations in the age and growth of
yellowfin tuna larvae, Thunnus albacares,
collected about the Mississippi River plume.
Environmental Biology of Fishes 39: 259-70.
82
-------
McNeil, C.S. and Grimes, C.B. 1995. Diet and
feeding ecology of striped anchovy, Anchoa
hepsetus, associated with the Mississippi River
discharge plume. In 'Nutrient-Enhanced
Coastal Ocean Productivity'. (Eds. D.K.
Atwood, W.F. Graham, and C.B. Grimes.)
pp. 81-89. Louisiana Sea Grant College
Program: Baton Rouge.
Olson, D.B. and Backus, R.H. 1985. The
concentrating of organisms: A cold water
fish and a warm core Gulf Stream ring.
Journal of Marine Research 43: 113-37.
Powell, A.B., Chester, A.J., Govoni, J.J., and
Warlen, S.M. 1990. Nutritional condition of
spot larvae associated with the Mississippi
River plume. Transactions_of the American
Fisheries Society 119: 957-65.
Sinclair, M. 1988. 'Marine populations: An essay
on population regulation and speciation'.
University of Washington Press, Seattle.
Wiseman, W.J. and Dmnel, S.P. 1988. Currents-
near the mouth of the Mississippi River.
journal Physical Oceanography 18: 1287—91.
Table 3.
Comparisons of the average variation
(Coefficient of variation, CV) in total
ichthyoplankton catch at stations along drifter
tracks to CV's for catches made at stations
along transects that intentionally sampled
plume, front and shelf waters.
Along drifter tracks
• Gimes and Wiseman
(unpublished)
Among plume, front and shelf
waters
• Grimes and Finucane (1991)
• Govoni and Grimes (1992)
CV
61 (fall)
235 (fall)
140 (fall)
129 (spring)
23
129
Station number
Figure 48.
83
-------
••M
1
o
CD
CO
E
0.6
0.5
0.4
0.3
0.2
0.1
0
Observed Q
Predicted/Calculated H
D
H3 D BD
. • " •
Sep 4 Sep 5 Sep 6 Sep 7 Sep 8 Sep 9 Sep 10
Date
Figure 49.
1.837
*k
1
o5
1300
1100
i
900
700
500
300
100
V 0.05
123456789 10 123456789 10
Diffusivity (m 2 sec "1)
Figure 50.
84
-------
Figure SI.
1.9
E 1.7
I1-5
o
05 1.3
cd
cd
o
1.1
0.9
21
15
261
I
71
30
6
81
J L
20 22 24 26 28 30 32 34
Salinity (0/00)
Figure 52.
85
-------
Presentation Discussion
ChurchiU Grimes (NMFS—Panama City, EL)
Fred Bryan (National Biological Service—Baton
Rotige, LA) commented that between 11-19 days
of age, fish may be approaching the size where
they are no longer equally disposed to the gear
being used. Therefore, the estimate of mortality
could be indicative of the organisms having
become more nektonic than planktonic and
thus are no longer available to a neuston or
Tucker trawl.
Churchill Grimes agreed with Fred Bryan's
observation. He said that estimating larval
mortality is an extremely difficult task. He
attempted to estimate larval mortality by
following the same patch of larvae through a
specified period of time. However, because the
variation in catch was so great, it was impossible
to complete the study.
86
-------
Potential
SUl £ V*r-%V
/^*> ~m -• * ft \ Afim** '- --*~-'^3?^i**^--m*ffi^-;m
ames G. Honifen, WillitmM^^
Louisiana Department
Marine Fisheries Division
P.O. Box 98000 ....
Baton- Rouge, LA 70898-900g\3
Abstract
The harvest of marine fishery resources of
Louisiana, valued at over one-half billion
dollars annually and dependent upon the
State's nutrient-rich estuaries, potentially are
subject to impact from hypoxia. Hypoxic
conditions periodically develop in near shore
waters in many areas of the world, including the
northern Gulf of Mexico. Salinity and
temperature stratification in the nearshore Gulf of
Mexico results in conditions conducive to the
development of hypoxic and anoxic bottom
waters west of the Mississippi River delta. Stable
summer weather patterns, freshwater inflow from
the Mississippi River and local precipitation, and
nutrient enrichment from these sources
contribute to increased bacterial decomposition
and oxygen demand in near-bottom waters. The
magnitude of the phenomenon, in terms of
depression of dissolved oxygen concentration and
areal extent along the coast, varies annually.
Likewise, impacts to fisheries vary. The
distribution of fishery species is affected by
displacement of demersal nekton and mobile
epibenthic species assemblages and communities
to areas with sufficient dissolved oxygen, and
disruption of species movement patterns. Plank-
tonic stages of fishery species are subject to stress
and mortality in hypoxic waters. Preliminary
indications are that pelagic species have not been
impacted, although severe hypoxic conditions
extending: high into the water column may have
affected their distribution and move ment pat-
terns. Other potential impacts to Louisiana's
fisheries include: concentration of fishing effort
resulting in increased harvest; localized overfish-
ing in some areas; shellfish mortality if hypoxic
conditions impinge on coastal bay waters, local-
ized mortality of finfish and shellfish in shoreline
areas; and decreased recruitment due to impacts
to zooplankton species assemblages. Changes in
the relative amounts of nutrients can affect phyto-
plankton community dynamics, resulting in
changes throughout the food web; replacement of
diatoms with dinoflagellates may result in devel-
opment of red and brown tides with resultant
adverse impacts to fisheries. Changes in the dis-
tribution and abundance of fish species will result
in a loss of commercial and recreational harvest
opportunities and a net economic loss to the
State. Economically marginal commercial par-
ticipants may leave the fishery, and some recrea-
tional participants may elect not to fish. Fishery
management decisions that are based on fishery
independent data from resource surveys that do
not take hypoxia into account may result in a loss
of precision in assessment of fishery stocks.
introduction ~ J'.,,,£• j
The harvest of the commercial and recreational
marine fisherv resources of Louisiana is valued at
87
-------
over one-half billion dollars annually, and gener-
ates over 1.2 billion dollars in economic activity in
the state of Louisiana (David LaVergne, Econo-
mist, Louisiana Department of Wildlife and
Fisheries (LDWF), personal communication).
These fisheries, the products of which are used
nationally and internationally, are dependent upon
the State's nutrient-rich estuaries and nearshore
waters of the Gulf of Mexico. They potentially are
subject to impacts resulting from the occurrence
of hypoxic dissolved oxygen (D.O.) concentra-
tions of less than 2.0 mg/1) conditions in the bot-
tom waters of the nearshore Gulf of Mexico. This
phenomenon off Louisiana represents a potential
threat to the health and viability of these fisheries.
Hypoxic bottom waters offshore from the
Louisiana coast were reported initially from 1935
(Richards, 1957; Conseil Permanent International
pour 1'Exploration de la Mer, 1936; cited in
Bedinger, et al., 1980), and subsequently have
been studied and reported by numerous investi-
gators (Ragan, et al., 1978; Harris, et al., 1978;
Bedinger, et al. 1980; Harper, et al., 1981; Stuntz,
et al., 1982; Turner and Allen, 1982a; 1982b;
Rabalais, et al., 1985; Renaud, 1986; Rabalais, et
al, 1994; Schurtz and St. Pe1,1984). The Missis-
sippi River flood that ultimately resulted in an
unprecedented area of hypoxic water bottoms
during the summer of 1993 focused national
attention on the Louisiana coast (e.g., Holstrom,
1993).
LDWF has collected hypoxia-related data through
its Coastal Study Area fishery management pro-
gram since its inception in 1966. Other study-
specific programs, such as the Cooperative Gulf
of Mexico Inventory (Barrett, et al., 1971; Perret,
et al., 1971) also have contributed to the Depart-
ment's understanding of the relationships between
environmental factors and coastal fisheries.
LDWF noted the occurrence of hypoxic condi-
tions in 1973 and began regular monitoring of
Gulf waters in which hypoxic conditions occur in
1978 (Foote, 1982). In 1982, funding became
available from the National Marine Fisheries
Service (NMFS) for the Southeast Area Monitor-
ing and Assessment Program (SEAMAP), and
limited additional data were collected by LDWF
from across the central coast. The expansion of
SEAMAP in 1985 provided the opportunity to
collect data related to hypoxia and its potential
impacts to fisheries from water depths of 5 to
20 fathoms between the Mississippi River and
Atchafalaya Bay (Figure 53). The data presented
here were collected primarily for other fishery
management and environmental monitoring
purposes, not specifically for measuring the
impacts of hypoxia on fisheries. All fisheries-
related data were collected using otter trawls of
various sizes and configurations, and the findings
are presented as a descriptive summary. Datasets
have been combined to illustrate the potential
impacts of hypoxia and have not been subjected
to rigorous statistical analysis.
C ~~~~ — 4
^Physical and Chemical Data
Hypoxic conditions periodically develop in near-
shore waters in many areas of the world, including
the northern Gulf of Mexico. (Richards, 1957;
Faganeli, et al., 1983). Factors hypothesized to
contribute include organic sediment loads
(Richards, 1957; Ragan et al., 1978), reduced
vertical mixing of the water column due to salinity
and temperature stratification coupled with ben-
thic and planktonic respiration (Turner and Allen,
1982a; 1 982b). Stable summer weather patterns,
freshwater inflow from the Mississippi River and
local precipitation, and nutrient enrichment from
these sources contribute to increased bacterial
decomposition and oxygen demand in near-
bottom waters (Richards, 1957; Turner and Allen,
1982a; 1982b). Salinity and temperature stratifica-
tion in the nearshore Gulf of Mexico results in
conditions conducive to the development of
88
-------
hypoxic and anoxic bottom waters west of the
Mississippi River Delta (Figure 54). Typically,
only the near-bottom waters become
hypoxic/anoxic, principally during the months of
May through August (Figure 55). The phen-
omenon develops because of stable local summer
weather patterns that do not provide sufficient
energy in terms of wind and wave action to break
the density barriers established between the upper
and lower water column. The barriers thus estab-
lished by salinity and temperature isolate the
lower reaches of the water column where oxygen
demand from respiration and decomposition
deplete dissolved oxygen. The resulting
hypoxic/anoxic condition persists until the water
column again is mixed by a frontal passage,
tropical weather system, or other disturbance.
Once the disturbance has passed and stable
conditions again prevail, hypoxia generally
becomes re-established. The regular passage of
cold fronts beginning in the fall causes the water
column to mix and remain in that state until the
next summer.
The magnitude of the event varies annually in
terms of the size of the hypoxic area and the
depression of D.O. concentrations. This is
illustrated by the frequency of encountering a
location with hypoxic bottom water. The
Midwestern drought in the late 1980's resulted in
a relatively small number of nearshore hypoxic
observations. Conversely, during the Mississippi
River flood of 1993, a record number of hypoxic
stations were encountered during routine
sampling (Figure 56). The severity of the
depression of D.O. levels also indicates the
variability of the event. Hypoxic conditions might
develop over a large geographical area, but the
depression in D.O. levels might be slight as
compared to other years. For example, LDWF
data indicate that the 1984 event was moderately
severe with over 10 percent of samples containing
D.O. concentrations of less than 2.0 mg/1, or
approximately half the number of stations found
in the record event of 1993. However the mean
D.O. concentration from samples collected
during the summer of 1984 was greater than
3.0 mg/1 (Figure 57), indicating that the hypoxic
event that year was of shorter duration or covered
a smaller area than in other years. During 1993
over 20 percent of stations sampled were found
to be hypoxic, and D.O. concentrations were near
0.0 mg/1, indicating a large, severe event.
Concentrations of silicates in surface waters
during spring months, although not attributable
to hypoxia, show a relationship with its subse-
quent development. Silicates comprise the skele-
ton of diatoms, the dominant phytoplankton
group in the northern Gulf of Mexico. High
concentrations of silicates generally are followed
by low measurements of D.O., and vice versa
(Figure 58). High levels of silicates in the spring
may indicate the potential for a subsequent
bloom of diatoms. As these complete their life-
cycle and settle into the bottom waters, decompo-
sition consumes the available D.O. and contri-
butes to the hypoxic event. Nutrients remaining
in the surface waters are then available to other
phytoplankton groups such as cyanobacteria and
dinoflagellates. These organisms have been linked
to formation of hypoxic bottom waters (Dortch,
1994). Shifts in the species composition of this
community may make conditions favorable for
blooms of the noxious and toxic phytoplankton
that cause red tides.
Impactsrto Fisheries
The presence of hypoxic waters in an area can be
expected to have a variety of impacts to fisheries.
Nekton communities and assemblages, being
mobile, will move away from areas with insuf-
ficient D.O. and congregate along the borders of
the hypoxic area until conditions are conducive to
their return. Mobile epibenthic organisms simil-
arly will leave the area if possible. Planktonic
89
-------
communities that are unable to swim away from a
hypoxic water mass, or benthic communities
associated with specific water bottoms, would be
subject to stress and/or mortality depending on
the severity of the hypoxic event and their length
of exposure to it.
Observed Impacts
The distribution of fishery species is affected by
displacement of demersal nekton and mobile
epibenthic species assemblages and communities
to areas with sufficient dissolved oxygen, and
disruption of species movement patterns.
Numbers of demersal species, and their abun-
dance, are reduced greatly (Bedinger, et al., 1980;
Stuntz, et al., 1983). Pihl, et al. (1991 ) found that
demersal species tended to migrate to shallower
water to escape hypoxic bottom conditions.
Preliminary indications are that pelagic species
have not been impacted, although severe hypoxic
conditions extending high into the water column
may affect their distribution and movement
patterns. Stuntz, et al. (1983) speculated that
pelagic species may congregate around hypoxic
water bottoms to take advantage of feeding
opportunities on benthic, epibenthic, and demer-
sal organisms that are rendered vulnerable to pre-
dation by stress resulting from low oxygen levels.
Pihl, et al. (1992) observed a similar phenomenon
in demersal species in Chesapeake Bay.
Otter trawls of the type used in the LDWF
surveys are designed to collect organisms that
reside on or near the bottom. Therefore the catch
from this gear generally is comprised of epi-
benthic and demersal species. Reef-associated
species sometimes are caught if the gear passes
over an irregularly-contoured area of the bottom
that provides reef-like habitat, or if the animals
are moving between reef areas. Pelagic species can
be caught near the bottom if a school is encoun-
tered as the gear is fishing, or off-bottom as the
gear is being set or retrieved. The plankton net
data reported here were collected in a survey to
determine zooplankton species composition and
abundance in near-bottom offshore waters sub-
ject to hypoxic conditions. Seventy-five species of
finfishes, crustaceans, and cephalopods were
collected in the Department's fishery-independent
trawl surveys in the area where hypoxic condi-
tions occur. The LDWF data indicate that a
variety of epibenthic, demersal, pelagic, and reef
species were present in both trawl and plankton
net catch from the hypoxic zone (Table 4). Target
species for directed fisheries, both commercial
and recreational, were recorded in the catch
regularly.
Composition and abundance of species caught in
trawls decreased with D.O. concentration. The
LDWF data indicate that 35 percent of nearshore
trawl samples collected from hypoxic waters
during summer result in no live organisms being
caught. Large catches from hypoxic or anoxic
waters generally were comprised of pelagic
species. Trawl catch in waters with D.O. concen-
trations between 0.0 and 1.0 mg/1 was comprised
of 34 species (Figure 59). Although the number of
taxa encountered was only slightly less than that
found at D.O. concentrations above 3.0 mg/1, the
total number of individuals was significantly
depressed. Between 1.0 and 2.0 mg/1 of D.O., the
number of species caught decreased to 24, but the
number of individuals in the catch increased by
nearly an order of magnitude. Between 2.0 and
3.0 mg/1 of D.O. the number of species increased
to 46 while the number of individuals decreased
slightly. The number of species remained rela-
tively constant as D.O. concentrations rose from
3.0 to 4.0 mg/1, and the number of individuals
increased by approximately 33 percent. Weight of
the catch exhibited a nearly linear increase as
D.O. concentrations rose (Figure 60). Pelagic
species exhibited a nearly constant relative
abundance in trawl catches, remaining
-------
approximately at 20 percent except at D.O.
concentrations between 2.0 and 3.0 mg/1
(Figure 61). Of the 34 species found between
D.O. concentrations of 0.0 and 1.0 mg/1, approxi-
mately one-third "were epibenthic species that may
not have been able to avoid the development of
hypoxic conditions. The increase in numbers of
individuals between 1.0 and 2.0 mg/1 D.O.
suggests that animals may be congregating around
the margins of the hypoxic area, upon being
displaced as the event develops. The relative
abundance of pelagic species increased when
D.O. concentrations increased between 2.0 and
3.0 mg/1. This increase may be due not only to
availability of prey organisms under stress from
low D.O. concentrations, but also to increased
numbers of prey that have been concentrated by
hypoxia.
.Potential Impacts ;: -<< " i --••
The observed impacts to fisheries vary with the
decrease in D.O. concentration, and presumably
with the area of the water bottoms that are
affected by hypoxic conditions. Mobile organisms
move from hypoxic waters to congregate in areas
where D.O. concentrations are sufficient to sus-
tain life. Animals unable to move away from
developing hypoxic/anoxic waters, either because
of life stage or habit, are subject to mortality. The
observed redistribution of species, concentration
of abundance, and mortality that hypoxic con-
ditions cause may also contribute to other
potential impacts to Gulf of Mexico fisheries.
Numbers of pelagic species appear to increase in
response to availability of prey organisms concen-
trated at the edges of the hypoxic area. Numbers
of fishermen may increase in these areas for
similar reasons. The concentration of fishing
effort may result in increased harvest of target and
non-target species. If the relative abundance of
nontarget species is high in these areas, increased
bycatch mortality may result. Similarly, localized
overfishing in some areas may be possible if the
relative abundance of target species is high.
Additionally, if the fishery is prosecuted over a
limited area, much of which is affected by hypoxic
bottom water, the relatively high proportion of
target species can lead to local overharvest.
Particularly severe hypoxic events, or those that
impinge closely upon the shoreline, may leave no
escape for organisms displaced and crowded by
low D.O. concentrations. As a result, localized
mortality of finfish and shellfish in shoreline areas
may occur. Hypoxia was a factor in a 1973 jubilee
and fish kill at Grand Isle (Philip E. Bowman,
LDWF, personal communication). Further
impingement on the shoreline and encroachment
into bay waters may impact the oyster fishery.
Adult oysters will shut their valves to withstand
hypoxic conditions for short periods of time. If
hypoxic conditions persist for days, however,
mortality is likely. Furthermore the peaks of
oyster spawning are in early and late summer,
when hypoxic conditions may exist. Larval
oysters, as well as the eggs and larvae of other
fishery species, are components of zooplankton
assemblages and therefore potentially are subject
to stress and/or mortality from exposure to
hypoxic conditions (Earl Melancon, Nichols State
University, Thibodaux, Louisiana, personal com-
munication). Decreased recruitment to fishery
stocks, disease, or slow growth due to exposure to
hypoxia are other potential impacts to zooplank-
ton species assemblages. Changes in the relative
amounts of nutrients may affect phytoplankton
community dynamics, resulting in changes
throughout the food web; replacement of diatoms
with cyanobacteria or dinoflagellates may result in
development of red and brown tides with resul-
tant adverse impacts to fisheries.
Low catch rates have been demonstrated in
research and resource surveys (Pavela, et al., 1983;
Bedinger, et al., 1980). Similar decreases in
91
-------
directed fisheries have been reported (Krapf,
1994; Barton, 1995). Changes in the distribution
and abundance of fishery species may result in a
loss of commercial and recreational harvest
opportunities and a net economic loss to the
State. Fishermen may have to travel farther to
productive waters until the point where the
economic outlay does not equal the expected
return. The recreational fisherman's response may
be to reduce fishing effort, and therefore,
fishing-related expenditures. Decreased catch in
the commercial fishery may affect a suite of
related industries, including processors,
wholesalers, retailers and restaurants, resulting
ultimately in a loss in economic activity.
Economically marginal commercial participants
may leave the fishery. Hypoxia was cited as one of
the factors that led to the closing of the Zapata
menhaden processing plant in Dulac, Louisiana in
1995 (Barton,!995).
Fishery-independent data provide indices of
population levels to fishery managers. In a highly
regulated fishery these data provide the only
available measure of changes in population levels.
The changes in distribution of species resulting
from hypoxia, movement away and concentration,
may result in more resource survey samples with
high and low abundance. This will result in an
increase in the sample variances of resource
survey data. Population estimates
based on these data will be less precise. Fishery
managers then must allow for the larger
confidence interval surrounding population
estimates in formulating management decisions
regarding seasons and quotas.
Summary
Louisiana's fisheries, and to a large extent those of
the northern Gulf of Mexico, depend on the
Mississippi River for their existence. The
sediments and nutrients carried by the river built
the Louisiana coastal marshes. Today, as a result
of leveeing the river, nutrients and sediments that
once built and maintained Louisiana's coastal
marshes are being deposited off the Continental
Shelf in the abyssal depths of the Gulf of Mexico.
Decreasing the nutrient levels in the Mississippi
River may serve to lessen the severity of hypoxia
in coastal waters, but also may impact the food
web of the northern Gulf and decrease fisheries
production.
Distributional changes that can be related to
hypoxia have been observed in demersal fish and
invertebrate communities. Plankton communities
are subject to stress and mortality in hypoxic
waters since these organisms lack the ability to
avoid the area. Preliminary indications are that
pelagic species have not yet been affected, other
than a possible change in local population density
related to feeding behavior. However, an increase
in the severity of the phenomenon that leads to
hypoxic waters extending high into the water
column potentially can also impact the distribu-
tion and movement of these highly-mobile species.
Other potential impacts include a concentrating
of fishing effort resulting in increased harvest; low
catch rates in directed fisheries; localized over-
fishing; mortality; decreased recruitment due to
impacts to zooplankton. Changes in the concen-
trations of nutrients, such as silicate, can result in
a change in phytoplankton community dynamics,
and subsequent changes throughout the food
web. Diatoms replaced by dinoflagellates may •
result in development of red and brown tides with-
resultant adverse impacts to fisheries, such as
advisories against catching/keeping certain
species, off-flavors, and direct mortality. Changes
in the distribution and abundance of fish species
could result in a loss of commercial and recrea-
tional harvest opportunities and a net economic
loss to the State. Economically marginal
commercial participants may leave the fishery, and
recreational participants may reduce fishing effort.
92
-------
Fishery management decisions that are based on
fishery-independent data from resource surveys
that do not take hypoxia into account may result
in a loss of precision in assessment of fishery
stocks.
References ' ^ '
Barton, E.A. 1995. Dead zone blamed for plant's
closure. Houma Daily Courier. November 6,
1995.
Barrett, B.B., J.W. Tarver, W.R. Latapie, J.F.
Pollard, W.R. Mock, G.B. Adkins, W.J.
Gaidry, CJ. White, and J.S. Mathis. 1971.
Cooperative Gulf of Mexico estuarine
inventory and study, Louisiana. Phase II,
Hydrology and Phase III, Sedimentology.
Louisiana Wildlife and Fisheries Commission.
New Orleans, Louisiana. 191 pp.
Bedinger, C.A., J.W. Cooper, A. Kwok, R.E.
Childers, and K.T. Kimball. 1980. Ecological
investigations of petroleum production
platforms in the central Gulf of Mexico.
Volume 1: pollutant fate and effects studies.
Draft final report submitted to the Bureau of
Land Management, Contract No.
AA551-CT8-17. 149pp.
Dortch, Q. 1994. Changes in phytoplankton
numbers and species composition, pp. 4649.
in. Dowgiallo, M.J., Editor. Coastal
oceanographic effects of summer 1993
Mississippi River flooding. U.S. Department
of Commerce, National Oceanic and
Atmospheric Administration.
Faganeli, J., A. Avcin, N. Fanuko, A. Male], V.
Turk, P. Tusnik, B. Vriser, and A. Vukovic.
1985. Bottom layer anoxia in the central part
of the Gulf of Trieste in the late summer of
1983. Marine Pollution Bulletin. 16(2):75-78.
Foote, K.J. 1982. Observations of oxygen
depletion during LOOP environmental
studies. (Abstract). Presented at: Gulf of
Mexico Information Transfer Meeting. U.S.
Department of the Interior, Minerals
Management Service. August 24-26, 1982.
Harper, D.E., Jr, L.D. McKinney, R.R. Salzer, and
RJ. Case. 1981. The occurrence of hypoxic
bottom water off the upper Texas coast and
its effects on the benthic biota. Contributions
in Marine Science 24:53—79.
Harris, A., J. Ragan, and J. Green. 1978. The
recurrence of oxygen-deficient waters in a
shallow marine environment. (Abstract). In.
American Association for the Advancement
of Science. Abstracts of papers of the 144th
national meeting.
Holstrom, D. 1993. Mississippi flooding leaves a
toxic legacy: pollution in the Gulf of Mexico.
Christian Science Monitor. September 7, 1993.
Krapf, D. 1994. Gulf shrimp. National
Fisherman. April, 1994.
Pavela, J.S., J.L. Ross, and M.E. Chittenden. 1983.
Sharp reductions in abundance of fishes and
benthic macroinvertebrates in the Gulf of
Mexico off Texas associated with hypoxia.
Northeast Gulf Science 6(2):167-173.
Perret, W.S., B.B. Barrett, W.R. Latapie, J.F.
Pollard, W.R. Mock, G.B. Adkins, W.J.
Gaidry, and CJ. White. 1971. Cooperative
Gulf of Mexico estuarine inventory and study,
Louisiana. Phase 1, Area description and
Phase IV, Biology. Louisiana Wildlife and
Fisheries Commission. New Orleans,
Louisiana. 174pp.
Pihl, L., S.P. Baden, and RJ. Diaz. 1991. Effects
of periodic hypoxia on distribution of
93
-------
demersal fish and crustaceans. Marine
Biology. 108:349-360.
Pihl, L., S.P. Baden, RJ. Diaz, and L.C. Schaffner.
1992. Hypoxia induced structural changes in
the diet of bottom-feeding fish and Crustacea.
Marine Biology. 112:344361 .
Rabalais, N.N., R.E. Turner, W.J. Wiseman, Jr.,
and D.F. Boesch. 1985. Hydrographic,
biological, and nutrient characteristics of the
water column on the Louisiana shelf, July and
September,! 985. Louisiana Universities
Marine Consortium. Data Report Number 3.
Rabalais, N.N., R.E. Turner, and W.J. Wiseman,
Jr. 1994. Hypoxic conditions in bottom waters
on the Louisiana-Texas shelf, pp. 50—54. in.
Dowgiallo, M.J., Editor. Coastal oceano-
graphic effects of summer 1993 Mississippi
River flooding. U.S. Department of Com-
merce, National Oceanic and Atmospheric
Administration.
Ragan, J.G., A.M. Harris, and J.H. Green. 1978.
Temperature, salinity, and oxygen measure-
ments of surface and bottom waters on the
continental shelf off Louisiana during por-
tions of 1975 and 1976. Professional Papers
Series (Biology), Number 3. Nichols State
University, Thibodaux. Louisiana.
Renaud, M.L. 1986. Hypoxia in Louisiana coastal
waters during 1983: implications for fisheries.
Fishery Bulletin 84(l):19-26.
Richards, F.A. 1957. Oxygen in the ocean, pp
185-238 in: Hedgepeth, J.W., ed. Treatise on
marine ecology and paleoecology. Geological
Society of America. Memoir Number 67.
Volume 1.
Schurtz, M.H., and K.M. St. Pe'. 1984. Water
quality investigation of environmental
conditions in Lake Pontchartrain, report on
interim findings. Louisiana Department of
Environmental Quality, Water Pollution
Control Division. 85pp. Appendix.
Stuntz, W.E., N. Sanders, T.D. Leming, K.N.
Baxter, and R.M. Barazotto. 1983. Area of
hypoxic bottom water found in northern Gulf
of Mexico. Coastal Oceanography and
Climatology News. 4:37-38.
Turner, R.E., and R.L. Allen. 1982a. Bottom
water oxygen concentration in the Mississippi
River Delta Bight. Contributions in Marine
Science. 25:161-172.
Turner, R.E., and R.L. Allen. 1982b. Plankton
respiration rates in the bottom waters of the
Mississippi River Delta Bight. Contributions
in Marine Science. 25:173-179.
-------
Table 4.
Summary of species caught in LDWF trawl and plankton net samples from nearshore waters,
1978-1995. Bold type indicates a species sought either in the commercial or recreational fishery.
Epibenthic
Demersal
Pelagic
Other species
Plankton
_- " s' s^ _ f '' " ^f|S&M:VV-^% V> ', *-*-•- -w ,__ ^ /N - .
Mud creabs, purse crabs, spider crabs, other crabs, batfish, southern
flounder, ocellated flounder, other flounders, soles
White shrimp, brown shrimp, blue crab, mantis shrimp, other swimming
crabs, sharpnose shark, anchovies, lizardfishes, catfishes, cusk-eeis, spotted
seatrout, sand seatrout, silver seatrout, southern kingfish. croaker,
other drums, sea basses, searobins, puffers
Squids, Gulf menhaden, other herrings, spadefish, Spanish mackerel,
Atlantic bumper, other jacks, bluefish, Gulf butterfish, harvestfish
Berracudas, pinfish, red snapper, mojarras, seahorses, filefish, triggerfish,
remoras
Brown shrimp, seabob, other shrimp, blue crab, other swimming crabs,
anchovies, herrings, jacks, sand seatrout, spotted seatrout, red drum,
other drums, Spanish mackerel, other species
Figure 53.
Map of coastal Louisiana showing approximate side and location of 1985 and 1993 hypocix
events, and the location of the LDWF study area.
95
-------
SALINITY (PPT)
25.3
31.4
TEMPERATURE (CELSIUS)
1.0
7.1
132
METERS j-
19.2
25.3
31.4
DISSOLVED OXYGEN (PPM)
25.3
31.4
Figure 54.
Nearshore water column profile showing stratification based on salinity (top)
and depression of dissolved oxygen concentration in near-bottom waters
(bottom). LDWF data, July 1995, sample.
96
-------
O —i.
Jan Feb Mar Apr May Jun Jul Aug S«p Oct Nov Deo
Months
Figure 55.
Monthly mean concentration of dissolved oxygen in bottom waters at
nearshore LDWF stations, 1978-1995.
25-
_ 15 H
o
4)
78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
Figure 56.
Percent occurrence ofhypoxic conditions in bottom waters at nearshore
LDWF stations, 1978-1995.
97
-------
7 -•
: ! '• • I I I ! • i I I I . I • >, I
78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
Year
Figure 57.
Mean summer (June, July, and August) dissolved oxygen concentration
in bottom waters at nearshore LDWF stations, 1978-1995.
D.O. (Y1)
Silicate (Y2)
— 2
1.5
o'
fa
'— 1
— O.S
m
"ST
aa
ta
*+*
er>
! I i ) 1 I I 1 I I i ~
7S 79 80 81 82 83 84 85 86 87 88 89 9O 91 92 93 94 95
Year
Figure 58.
Concentrations of silicates in surface waters during spring months (April, May,
and June) and summer (June, and August) dissolved oxygen concentration in
bottom waters from LDWF nearshore stations, 1978-1995.
98
-------
so
4OOO
4O
ffofTaxa (Y1)
# of Individuals (Y2)
- 35OO
3OOO
25OO
2OOO
15OO
1OOO
socr
O.O to 1.0 1 .0 to 2.0 2.0 to 3.O
D.O. Concentration
-O
3.O to 4.O
z
c
3
sr
S3.
I
c
Figure 59.
Numbers oftaxa and total numbers of individual organisms caught in LDWF
trawl samples from nearshore waters, 19 78-1995.
so
— 20
o
KT
15 Si
JB
oa"
:?•
1O ^
03
10
# of Taxa (Y1)
Total Weight (kg) (Y2)
— s
O.O to 1 .O 1 .O to 2.O 2.O to 3.O
D.O. Concentration
3.O to 4.O
Figure 60.
Numbers oftaxa and total weight of catch from LDWF trawl samples a at
nearshore stations, I978-I995.
99
-------
so —
10 -
Taxa
Weight
O.O to 1.0 1.0 to 2.O 2.O to 3.O
D.O. Concentration
3.O to 4.O
Figure 61.
Numbers oftaxa and total weight of catch from LDWF trawl samples at
nearshore stations, 1978-1995.
Presentation Discussion
Jim Hanifen (Louisiana Department of Wildlife &
Fisheries)
Eugene Turner (Louisiana State University—
Baton "Range, LA) commented that the silicate and
low dissolved oxygen relationship illustrates many
of the problems encountered during large data set
investigations. He believed there could be a
surrogate which is represented by the silicate
relationship demonstrated by a lag between the
silicate and dissolved oxygen which have been
shown by several models and analyses to be
similar with the loadings in the river and hypoxia
offshore. The silicate may be a surrogate for
salinity or loading because the lower the salinity
the higher the silicate. The silicate concentration
has been relatively steady through the study year
compared with previous years. However, that may
be a result of stratification. The same relationship
may be demonstrated if salinity were plotted
against dissolved oxygen. Therefore, these types
of studies should be conducted for more than one
year. Unfortunately, funding for ten year studies is
rarely available.
100
-------
Estuarine Hypo|ji|i
Jonathan Pennock /-t'llfl
Dauphin Island Sea Lab . '.. .: :.;:E2IIlj
Dauphin Island, Alabama
Abstract " **:
W™*» ft ™f ^ ^ N
Mass migrations of motile epifauna
(e.g., crabs and demersal fish) up onto
the shore of Mobile Bay have been
documented in the popular literature since the
mid-1800s. These events, referred to popularly
as "jubilees," are now known to be a result of
the movement of parcels of hypoxic and anoxic
bottom waters towards shore caused by the
winds and tides. Recent research has shown that
these hypoxic/ anoxic waters are caused by high
rates of oxygen consumption at the sediment
surface combined with extremely strong salinity
stratification in the water-column that
effectively isolates the bottom water from
oxygen available in the surface waters. For
Mobile Bay, these factors serve to create an
environment that is hypoxic approximately 50
percent of the time during the summer period.
There does not appear to be a long-term trend
(either positive or negative) in the duration of
these low-oxygen events. Rather, the frequency
and duration of hypoxia/ anoxia is related to
short-term variations in the physical structure of
the water-column. Despite the frequency of
these events, fisheries landings in Mobile Bay
remain high and researchers are now addressing
the question of whether such events (that may
help maintain highly productive "pioneer" com-
munities) may have a beneficial effect on secon-
dary production in the ecosystem.
No Manuscript Submitted.
Presentation Discussion
John Pennock (Dauphin Island Sea L^b—Dauphin
Island, AL)
Don Boesch (University of Maryland—Cambridge,
MD) asked the speaker if he could estimate how
many of the Gulfs estuarine systems are at a
stage where additional nutrient input would
enhance useful production.
John Pennock replied that there are many
systems, especially pristine grass bed systems,
which presumably could sustain more plankton
production. However, theoretically, those
nutrient inputs could have a negative, as well as
positive, effect on the grass beds. He said that
more turbid systems, for example Delaware Bay,
can sustain nutrient inputs better than many of
the cleaner water systems on the Gulf.
Clive Walker (Natural Resources Conserva-
tion Service—Texas A&M University) stated
that they used plots of fertilizer application
versus corn production to demonstrate the
threshold of benefit from fertilizers. These plots
are very similar to the nutrient versus estuarine
productivity plots shown in the presentation.
John Pennock said that these two concepts
were very similar and it was necessary to link
these concepts to resolve both the farmers' and
the marine fisheries' issues.
101
-------
Causes and Effects of I
Putting the
Robert]. Diaz • -^mm
School of Marine Science • ;:i"'-'::;||l^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Virginia Institute of Marine Science . ,.; ^$
College of William and Mary . . ^3|HiB
Gloucester Point, Virginia 23062 ... <, '^ilBr^
Abstract
..-,..-. i
Global Patterns of Benthic Hypoxia
and Anoxia: Causes, Responses, and i
Altered Energy Flows !
Synthesis of literature pertaining to benthic
hypoxia and anoxia (Diaz and Rosenberg,
1995) revealed that community and
population responses to low dissolved oxygen
stress were similar across all ecosystems and
followed a hierarchical pattern. The occurrence
of hypoxia and anoxia is expanding with
significant structural and functional changes in
affected benthic communities. Benthic-pelagic
coupling is also adversely affected. No other
environmental variable of such ecological
importance to coastal marine ecosystems
around the world has changed so drastically in
such a short period as dissolved oxygen. While
hypoxic and anoxic environments have existed
through geological time, their occurrence in
shallow coastal and estuarine areas appears to be
increasing, most likely accelerated by human
activities. The oxygen budgets of most major
coastal ecosystems have been adversely affected
mainly through the process of eutrophication,
which acts as an accelerant or enhancing factor
to hypoxia and anoxia. Many ecosystems that
are now severely stressed by hypoxia appear to
be near or at a threshold of imminent collapse
(loss of fisheries, loss of biodiversity, lower
system ascendancy).
102
"."3ns
^Introduction
No other environmental variable of such eco-
logical importance to coastal marine ecosystems
around the world has changed so drastically in
such a short period as dissolved oxygen. While
hypoxic and anoxic environments have existed
through geological time, their occurrence in
shallow coastal and estuarine areas appears to be
increasing, most likely accelerated by human
activities. Synthesis of literature pertaining to
benthic hypoxia and anoxia (Diaz and Rosen-
berg, 1995) revealed that the oxygen budgets of
most major coastal ecosystems have been
adversely affected mainly through the process of
eutrophication, which acts as an accelerant or
enhancing factor to hypoxia and anoxia. Many
ecosystems that are now severely stressed by
hypoxia appear to be near or at a threshold of
imminent collapse (loss of fisheries, loss of bio-
diversity, lower system ascendancy). Hypoxic
events on the Louisiana shelf will be discussed
relative to world wide problems with low
dissolved oxygen.
System Summaries ,,.<
rfH^ * f. * ** ~>~& s *s™*«t4ss»
Limfjorden, Denmark
• Experiences seasonal summertime hypoxia.
• Oxycline can be sharp with >5.6 ml/1 at
0.5 m and 0.4 ml/1 at 0.05m above the
bottom.
_
-------
• Annual mass mortality and recolonization
occurs.
• Also, Beggiatoa spp. occur in some areas all
year.
Gullmarsfjord, Sweden
• Oxygen concentration has declined
gradually from the 1950's but remained
above 2 ml/1.
— Fauna appeared stable up to 1979.
• During the winter of 1979 severe Hypoxia
occurred, reaching 0.8 ml/1.
— Fauna remained stable.
• Hypoxia continued and in January 1980
hypoxia reached 0.2 ml/1.
— Mass mortality of fauna occurred.
Upwelling & Oxygen Minimum Zone
• Stable hypoxia associated with high quality
organics.
• Leads to low diversity, but stable high
abundance and biomass fauna.
Kiel Bay, Germany
• A declining trend in dissolved oxygen has
been observed since 1950's.
• Periodic hypoxia began in the 1960's.
— Mortality of fauna was observed.
• In 1981 and 1983 severe hypoxia occurred.
— Mass mortality of fauna.
— Shift in fauna to opportunistic species.
• The 1981 event was widespread in all parts
of Kiel Bay, with H2S and Anoxia at
> 20m.
— Event was several weeks long.
• The result was mass mortality of benthos,
99 percent of biomass died.
- 30,000 mt of macrofauna died, 750 km2.
Kattegat, Sweden-Denmark
• Classic description of benthic communities
from this area. Fisheries well developed.
• By the 1970's there was speculation that the
ecosystem was not doing well.
• Annual hypoxia began in 1980.
- First observations of fish and benthos
mortality.
— 1984 record high catches in trawl
fisheries.
• Hypoxia was severe by 1985 and worsening.
— Mortality of benthos, fisheries reduced
to low levels.
• In 1988, 3,000 km2 affected.
— Mass mortality of benthos and fisheries,
poor recovery.
LA-TX Continental Shelf
• Chronic mild hypoxia may have existed, no
long-term data.
• The first measured hypoxic event was in
1973.
— Reductions in fauna.
— No mass mortality.
• In 1978 severe hypoxia occurred.
— Mass mortality of benthos.
— Shift to opportunistic species.
— Low fishery catches.
103
-------
• Currently 8,000-9,500 km2 are affected
annually.
— Mass mortality of benthos.
— Low fisherj' catches.
Black Sea
* Average depth of 1270 m, the largest mass
of "naturally occurring" permanently anoxic
water on earth.
« About 90 percent of its 5.4 x 105 km3
volume is anoxic beginning at depths of 150
to 250 m.
• Permanently hypoxic below about 100m
« Ukrainian northwestern Black Sea shelf is
critically eutrophic, periodic hypoxia and
anoxia is wide-spread encompassing all of
the Sea of Azov and up to 95 percent of the
Ukrainian northwestern shelf.
• Periodic events are distinct from the per-
manent anoxic layer and lead to mass
mortalities of benthic populations that
colonize the area during normoxic periods.
• In 1991, anoxia along the Romanian coast
eliminated an estimated 50 percent 6f the
demersal fish populations.
• Since the 1960's increasing hypoxia and
anoxia have been blamed for the replace-
ment of the highly valued demersal fish
species with planktic omnivores.
• Of the 26 commercial species fished in the
1960's only 6 still support a fishery.
Baltic Sea
• Trend of declining oxygen concentrations
was documented from the 1930's to the late
1960's in the deep basins.
Beginning in the 1960's and lasting up to the
present, large deep bottom areas of the
Baltic Sea have been mostly permanently
hypoxic or anoxic and devoid of benthic
macrofauna.
Below the halocline, at about 70 m, approxi-
mately 100000 km2 of the bottom is more
or less permanently hypoxic.
No significant change in the bottom water
oxygen content has occurred up to 1993.
Biomass "missing" in the anoxic areas
estimated to be 1.7 X 106 t wet wt
Periodic hypoxia in the mesohaline Born-
holm Basin in the south Baltic was reported
as early as 1948.
Benthic communities were reduced and
even eliminated during periods of hypoxia
or anoxia and how bottoms were
recolonized following a subsequent return
of normoxia.
^Reference
Diaz, R.J. and R. Rosenberg. (In Press) Marine
benthic hypoxia—review of ecological
effects and behavioral responses on macro-
fauna. Oceanogr. Mar. Biol. Ann. Rev.,
vol. 33.
Discussion
Robert Diaz (Virginia Institute of Marine
Science— Gloucester Point, VA
No questions/discussion following Mr. Diaz
presentation
104
-------
TableS.
Summary ofbenthic effects for hypoxic systems around the world. Several of these systems also experience
anoxia. In the case of many fjords, and coastal and oceanic oxygen minimum zones (OMZ) there is an anoxic
zone within which no macrofauna occur. The absence of fauna from these anoxic zones is not considered a
community response but a consequence of stable anoxia. Hypoxia is typed as: Aperiodic—events that are
known to occur at irregular intervals greater than a year; Periodic—events occurring at regular intervals shorter
than a year, related to tidal stratification/destratification cycles (Haas, 1977); Seasonal—yearly events relate to
summer or autumnal stratification; Persistent—year-round hypoxia. Levels of hypoxia are: Moderate— oxygen
decline to about 0.5 ml/1; Severe—decline to near anoxic levels, could also become anoxic. Time trends of
hypoxia, areal and or intensity, for the systems are: -= Improving conditions; + = Gradually increasing; ++ =
rapidly increasing; 0 = stable;. = no temporal data. Benthic community response is categorized as:
None—communities appear similar before and after hypoxic event; mortality—moderate reductions of
populations, many species survive; Mass Mort.—drastic reduction or elimination of the benthos. Benthic
recovery is: No change—dynamics appear unrelated to hypoxia; Some—recolonization occurs but community
does not return to prehypoxic structure; Slow—gradual return of community structure taking more than a year;
Annual—recolonization and return of community structure within a year.
System Level Response to Hypoxia
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
System
Deep Texas Shelf
German Bight, North Sea
New York Bight, New Jersey
Shallow Texas Shelf
Sommone Bay, France
North Sea, W. Denmark
Peru/Chile, El Nino, shallow
York River, Virginia
Rappahannock River, Virginia
Seto Inland Sea, Japan
Louisianna Shelf
Saanich Inlet, British Columbia
Bomholm Basin, S. Baltic
Oslofjord, Norway
Kattegat, Sweden-Denmark
German Bight, North Sea
Main Chesapeake Bay, MD
Port Hacking, Australia
Tolo Harbor, Hong Kong
Tome Cove, Japan
Laholm Bay, Sweden
Gullmarsfjord, Sweden
Swedish West Coast Fjords
Pamlico River, North Carolina
Limfjord, Denmark
Kiel Bay, Germany
Lough Ine, Ireland
Hillsborough Bay, Florida
Gulf of Trieste, Adriatic
Elefsis Bay, Aegean Sea
Black Sea NW Shelf
Arhus Bay, Denmark
Loch Creran, Scotland
Byfjord, Sweden
Black Sea (except NW shelf)
Idefjord, Sweden-Norway
Baltic Sea, Central
Fosa de Cariaco, Venezuela
Caspian Sea
Peru/Chile Upwelling Deep
Santa Maria Basin, California
Central California OMZ
Volcano 7, Pacific OMZ
Gulf of Finland, Deep
Hypoxia
Type
Aperiodic
Aperiodic
Aperiodic
Aperiodic
Aperiodic
Aperiodic
Aperiodic
Periodic
Periodic
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Persistent
Persistent
Persistent
Persistent
Persistent
Persistent
Persistent
Persistent
Persistent
Persistent
Persistent
Persistent
Hypoxia Time
Level Trends
Moderate 0?
Mod./Severe +
Severe
Severe +
Severe 4-?
Severe 4-
Severe 0?
Mod./Severe 0
Severe 4-
Moderate
Mod./Severe 4-
Mod./Severe 0
Mod./Severe 4- **
Mod./Severe 4-
Mod./Severe + 4-
Severe 4-?
Severe 4-
Severe
Severe
Severe
Severe 4- 4-
Severe 4-
Severe 4- +
Severe
Severe 4-
Severe 4-
Severe 0
Severe
Severe 4- 4-
Severe
Severe -f 4-
Sevcre 4-
Severe 0
Severe 0
Severe 4-
Severe +#
Severe 4- 4-
Severe
Mod./Severe 0
Mod./Severe 0
Mod./Severe 0
Mod./Severe 0
Mod./Severe 0
Mod./Severe
Bent. Com.
Response
Mortality
Mass Mort.
Mass Mort.
Mass Mort.
Mass Mort.
Mortality
Mass Mort.
None
Mortality
Mortality
Mortality
Mortality
Mass Mort.
Mortality
Mass Mort.
Mortality
Mortality
Mortality
Mass Mort.
Mortality
Mortality
Mass Mort.
Mortality
Mass Mort.
Mass Mort.
Mass Mort.
Mass Mort.
Mass Mort.
Mass Mort.
Mass Mort.
Mass Mort.
Mass Mort.
Mass Mort.
Mortality
No Benthos
Mortality
Mortality
Reduced
Mortality
Biomass Increase
Reduced
Biomass Increase
Biomass Increase
Reduced
Benthic
Recovery
Annual
Annual
Slow
Slow
Slow
Annual
???
No Change
Annual
Annual
Annual
Annual
Slow
Annual
Slow
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Some
Annual
Annual
Annual
Annual
Annual
Slow
Annual
Annual
Slow
No Change
Some
No Change
Some
Some
No Change
Some?
No Change
No Change
No Change
No Change
Slow
Fisheries
Stocks
Stressed
Surf Clam loss
Stressed
Collapse of Q
Stressed
Stressed
Stressed
Stressed
Stressed
Reduced
Collapse Norw
Stressed
Stressed
Stressed
Stressed
•Stressed
None
Stressed
Stressed
Reduced
Pelagic only
Pelagic only
Stressed
Enhanced?
Reference
Harper et al. 1981, 1991
Dethlefsen & Westemhgen 1983
i Boesch & Rosenberg 1981, Carlo et
al. 1979, Sindermann & Swanson 1980
Harper et al. 1981, 1991
dde fishery Desprez et al. 1992
Dyeretal. 1983,
Westemhagen & Dethlefson 1983
Rosenberg et al. 1983, Amtz &
Fahrbach 1991
Pihl et al. 1991, Diaz et al. 1993
Llans6 1990
Imabayashi 1986
Boesch&Rabalais 91, Rabalais et al.1991
Tunnicliffe 1981
Tulkki 1965, Leppakoski 1969
Peterson 1915, Mirza & Gray 1981
Rosenberg et al. 1987
y Lobster Baden et al. 1990a, Josefson
& Jensen 1992. Rosenberc et al. 1992
Niermann et al. 1990
Holland etal. 1987
Rainer & Fitzhardinge 1981
Wu 1982
Tsutsumi 1987
Baden et al. 1990b,
Rosenberg & Loo 1988
Josefson & Widbom 1988
Josefsen & Rosenberg 1988
Tenore 1972
Jargensen 1980
Arntz 1981, Weigelt 1990
Kitching et al. 1976
Santos & Simon 1980
Stachowilsch 1991
Friligos and Zenetos 1988
Zaitsev 1993
Fallesen & Jargensen 1991
Pearson 1981
Rosenberg 1990
Tolmazin 1985, Mee 1992
Rosenberg 1980
Andersin et al. 1978
Nichols 1976
Zenkevitch 1963
Amtz & Fahrbach 1991
Rosenberg et al. 1983
Hyland et al. 1991
Mullins et al. 1985
Levin et al. 1991
Andersin &. Sandier 1991
* Stable oxygen gradient associated with organic enrichment.
** These systems are currently in a persistent hypoxic state.
# Recent imporvements in oxygen concentrations due to pollution abatement.
105
-------
Evidence for Nutrient
Causing Hypoxia on t
'Coastal Ecology Institute
Department of Geology and Geophysics
Louisiana State University
Baton Rouge, Louisiana 70803 ,
3
'
^Abstract
The conclusion that there has been an
increase in the severity or areal coverage
of summer oxygen depletion is
dependent on knowledge of other coastal
systems, nutrient limitations on the modern
phy toplankton, and analysis of sedimentary
records. Five hypotheses may explain these
changes: 1) overland flow through coastal
wetlands has been severely restricted this cen-
tury, 2) Increased nutrient concentration in the
Mississippi River since the 1950s, 3) intrusions
of offshore waters causing a natural long-term
variability that is misinterpreted as a permanent
change, 4) Short- or long-term climate changes
(riverine fluctuations), and, 5) increased loadings
from estuarine sources. Based on the available
data, the strong inference is that only Hypo-
thesis No. 2 is sufficient to explain these
changes in an efficacious and non-contradictory
way.
Introduction
The conclusion that there has been an increase
in the severity or areal coverage of summer
hypoxia (e.g., Rabalais, et al., this volume) is
dependent on knowledge of other coastal
systems, nutrient requirements of phyto-
plankton, and analysis of sedimentary records.
106 __»«______»___
Six hypotheses are proposed to explain these
changes: 1) overland flow through coastal wet-
lands has declined severely this century,
2) increased nutrient and organic loadings from
estuarine sources, 3) intrusions of offshore
waters causing a natural long-term variability
that is misinterpreted as a permanent change,
4) short- or long-term climate changes (riverine
fluctuations), 5) organic loading from the Mis-
sissippi River causes the lack of oxygen, and,
6) the increased nutrient concentration in the
Mississippi River since the 1950s. Each of these
hypotheses has been tested, and the results
outlined in the following discussion.
ieses
Hypothesis No. 1
Overland flow through coastal wetlands has
been severely restricted this century by
navigation and flood control levees on the
Mississippi River. The consequence of this
disruption in the natural (geologic) hydrology
is to reduce the removal of nutrients from
water flowing over and through coastal
wetlands.
These hypotheses can be tested by examining
the amount of flow restricted by these levees,
determining the likelihood of nutrient removal
-------
in the area available, and comparing nutrient
concentrations in the Mississippi River and
receiving water bodies.
The amount of flow reduction from human-
made levees was determined by Kesel (1988) as
part of an effort to measure the effects on
suspended sediment transport. He said:
"The proportion of water discharge above bankfull
was computed from daily records and that proportion
used to determine the suspended load carried by
above bankfull flows. The amount of sediment
during this period that would have been available for
overbankflow was estimated to be 163.4 x 1Cr
metric tons. This amounted to 14 percent of the
suspended sediment carried during flood flows, but
only 2.6 percent of the total suspended load carried
during the entire 34-yrperiod." (Kesel, 1988).
A 2.6 percent reduction in suspended sediment
(and therefore nutrient) flow is thus an insigni-
ficant proportion of the total flow, which has
also tripled in nitrate concentration. Further-
more, there is a mismatch of overland flow
potential and river stage. Water levels on the
marsh peak in late summer, whereas the peak
river stage is in the spring. This mismatch mini-
mizes, rather than maximizes any potential
ability of wetlands to remove nutrients limiting
phytoplankton growth.
Furthermore, the ability of coastal wetlands to
absorb nutrients is not equal among wetland
types, and, in fact, most of Louisiana's coastal
wetlands appear to export the dissolved nutrient
forms that limit phytoplankton growth
(Table 6). In addition, the Louisiana experience
is that the conditions necessary for optimum
nutrient removal are not met (Table 7). There
are simply not enough forested freshwater
wetlands to remove even 10 percent of the
historically low nutrient concentration (prior to
1950) through overland flow.
Table 6.
Import (I) and Export (E) of nitrogen and phosphorus from
wetlands through overland flow (g m'2 y~).
"7, ''"toca'tipn' . igi
Fourleague Bay
Bayou Chevreuil
Barataria Bay
Fourleague Bay
Bonnet Carre'
;>!- "" Wetlan
-------
Table 7.
Nutrient removal optimization.
Favored by or indicated by
Long contact time (days/week)
Sufficient area
Higher loading =less retention
Subsurface flow higher removal than
surface flow
* ^"y- *^ /
Louisiana Experience--
Short ( 1 day, Bonnet Carre', an engineered crevasse)
Area restricted/limited by existing upland development and
landowner concerns; not extensive relative to loading rates
Not documented in Louisiana, but is a general experience
nationwide
Nitrate and phosphate exported from all coastal wetlands
but swamps
The conclusion is to reject Hypothesis No. 1.
Hypothesis No. 2
Nutrient and organic loadings from estuarine
sources has released organic matter offshore
in increasing amounts and caused hypoxic
water formation offshore.
This hypothesis can be tested by examining the
historical progression of estuarine eutrophica-
tion, the likelihood of estuarine-offshore
exchanges, and the net fluxes.
Estuarine exchanges with offshore waters
clearly exist. Evidence for this is the inverse
relationship between estuarine salinity and
Mississippi River discharge (Wiseman et al.
1990). Eutrophication of the estuaries has also
occurred (Rabalais et al. 1995). Therefore, it is
possible to exchange nutrients from inshore
estuaries to offshore. However, if there were
significant dominance of nutrients in either
direction (offshore to inshore, or vice-versa),
then the sedimentary record of diatom pro-
duction in nearshore and estuarine sediments
would be similar. The deposition/ accumulation
of biogenic silica (a surrogate for diatom pro-
duction) is strikingly different in both end
members. The accumulation rate of BSi in
estuarine waters reflects the use of fertilizer in
the estuarine basin and the accumulation in
offshore waters is coincidental with the nutrient
loading from the Mississippi River (Turner and
Rabalais 1994; Turner et al., unpublished).
Thus, there is no coherence between nutrient
loadings in the estuarine and offshore waters,
and Hypothesis No. 2 is not supported. Further,
a crude nutrient and carbon budget for estuarine
and offshore waters is dominated by the in situ
loadings, not the estuarine sources.
The conclusion is to reject Hypothesis No. 2.
Hypothesis No. 3
Intrusions of offshore waters cause a natural
long-term variability that is misinterpreted as
a permanent change.
This hypothesis can be examined by document-
ing physical connections between the oxygen
minimum layer (OML; Figure 62) found
throughout the open Gulf of Mexico and the
continental shelf and determining the respira-
tion rate in the OML.
108
-------
0
-500
1000
1= 1500
N-*'
.C
g. 2000
Q
2500
3000^
3500
*
*
01 2345678 9 10
Oxygen (mg/I)
Figure 62.
The oxygen minimum layer in the
open Gulf of Mexico.
Throughout years of data collection we cannot
find a physical connection between the hypoxic
water masses found in the OML within the
GOM waters (Figure 62) and in continental
shelf waters. Furthermore, the oxygen consum-
ption rates in the OML are insufficient (by
several orders of magnitude) to account for the
observed seasonal decline in oxygen concentra-
tion on the shelf.
The conclusion is to reject Hypothesis No. 3.
Hypothesis No. 4
Short- or long-term climate changes (riverine
fluctuations) occur and are mis-interpreted as
an increase in hypoxia.
Hypothesis No. 4 is not supported by
examination of the river discharge records
(Figure 63) or the sea level rise records, which
act as surrogates of major physical forcing
functions on the continental shelf.
2000
Figure 63.
The discharge of the Mississippi River
at Vicksburg, Miss, from the middle of the
last century to present
The conclusion is to reject Hypothesis No. 4.
Hypothesis No. 5
Organic loading from the Mississippi River
causes hypoxic water mass formation.
The amount of organic loading in the Missis-
sippi River is not large enough to account for
the observed decline in oxygen over such a large
area. There is much more oxygen removed
each summer than can be supported by carbon
introduced by the river. Also, the chemical
signature of the carbon (12'13C isotopic ratio)
found in material from the collection devices
placed in offshore waters, is different than in
the carbon from the river.
The conclusion is to reject Hypothesis No. 5.
Hypothesis No. 6
Nutrient loading from the Mississippi River
causes hypoxic water mass formation.
It appears that the nutrients in the Mississippi
River have changed in the same scale and in the
109
-------
amounts necessary to cause the observed
hypoxia (Figure 64; Turner and Rabalais 1991;
Rabalais et al., 1996, in press). Indicators of
oxygen stress are coincidental with the changes
in increased organic loading, as well (Sen Gupta
et al. 1996; Rabalais et al., 1996).
The conclusion is not to refect Hypothesis No. 6.
250
,-, 200
uj
w
r- 150
£ 100
a,
50
• Nitrate
o silicate
1960
1980
2000
Year
Figure 64.
Nitrate and silicate concentrations
in the Mississippi River
(from Rabalais et al., in press).
Summary
Based on the available data, there is a strong
inference that only Hypothesis No. 6 is
sufficient to explain these changes in a
efficacious and non-contradictory way.
Management measures based on Hypotheses
No. 1—5 are likely to be wasted efforts.
References
Dortch, Q., N. N. Rabalais, R. E. Turner and G.
T. Rowe. 1994. Respiration rates and
hypoxia on the Louisiana shelf. Estuaries
17(4): 862-872.
Dowgiallo, M. J. (ed.). 1994. Coastal Oceano-
graphic Effects of Summer 1993 Mississippi
River Flooding. Special NOAA Report,
NOAA Coastal Ocean Office/National
Weather Service, Silver Spring, Maryland.
Geyer, R. A. 1950. The occurrence of pro-
nounced salinity variations in Louisiana
coastal waters. J. Mar. Res. 9: 100-110.
Justic', D., N. N. Rabalais, R. E. Turner, and W.
J. Wiseman, Jr. 1993. Seasonal coupling
between riverborne nutrients, net produc-
tivity and hypoxia. Mar. Poll. Bull. 26(4):
184-189.
Justic', D., N. N. Rabalais, R. E. Turner and Q.
Dortch. 1995a. Changes in nutrient struc-
ture of river-dominated coastal waters:
Stoichiometric nutrient balance and its
consequences. Estuar. Coast. Shelf Sci. 40:
339-356.
Justic', D., N. N. Rabalais and R. E. Turner.
1995b. Stoichiometric nutrient balance and
origin of coastal eutrophication. Mar. Poll.
Bull. 30(1): 41-46.
Kesel, R. H. 1988. The decline in the suspended
load of the lower Mississippi River and its
influence on adjacent wetlands. Environ-
mental Geology and Water Science 11:
271-281.
Lohrenz, S. E., M. J. Dagg and T. E. Whitledge.
1990. Enhanced primary production at the
plume/oceanic interface of the Mississippi
River. Continental Shelf Res. 10: 639-664.
Nelson, D. M. and Q. Dortch. in press. Silacic
acid depletion and silicon limitation in the
plume of the Mississippi River: Evidence
from kinetic studies in spring and summer.
Mar. Ecol. Progr. Ser. (in press).
110
-------
Qureshi, N. A. 1995. The role of fecal pellets in
the flux of carbon to the sea floor on a
river-influenced continental shelf subject to
hypoxia. Ph.D. Dissertation, Department of
Oceanography & Coastal Sciences,
Louisiana State University, Baton Rouge,
255 pp.
Rabalais, N. N., R. E. Turner, W. J. Wiseman,
Jr. and D. F. Boesch. 1991. A brief summary
of hypoxia on the northern Gulf of Mexico
continental shelf: 1985--1988. Pages 35-46 in
R. V. Tyson and T. H. Pearson (eds.),
Modern and Ancient Continental Shelf
Anoxia. Geological Society Special Publ.
No. 58. The Geological Society, London.
Rabalais, N. N., W. J. Wiseman, Jr. and R. E.
Turner. 1994. Comparison of continuous
records of near-bottom dissolved oxygen
from the hypoxia zone of Louisiana.
Estuaries 17(4): 850-861.
Rabalais, N. N., R. E. Turner, D. Justic', Q.
Dortch, W. J. Wiseman, Jr. and B. K. Sen
Gupta. 1996. Nutrient changes in the
Mississippi River and system responses on
the adjacent continental shelf. Estuaries
19(2B): in press.
Rabalais, N. N., R. E. Turner, D. Justic', Q.
Dortch, W. J. Wiseman, Jr., and B. Sen
Gupta, in press. Gulf of Mexico biological
system responses to nutrient changes in the
Mississippi River. Proc. Estuarine Synthesis
Workshop, Irvine, Calif. J. Hobbie, ed.
Rabalais, N. N., Q. Dortch, D. Justic1, M. B.
Kilgen, P. L. Klerks, P. H. Templet, R. E.
Turner, B. E. Cole, D. Duet, M. Beacham,
M. Parsons, S. Rabalais, and R. Robichaux.
1995. Status and trends of eutrophication,
pathogen contamination, and toxic
substances in the Barataria-Terrebonne
estuarine system. Barataria-Terrebonne
National Estuary Program Publ. No. 22.
BTNEP Office, Nichols State Univ.,
Thibodaux, Louisiana, 265. pp plus
appendices.
Sen Gupta, B. K., R. E. Turner and N. N.
Rabalais. 1993. Oxygen stress in shelf waters
of northern Gulf of Mexico: 200-year
stratigraphic record of benthic foraminifera.
Page A138 in Geological Society of America,
1993 Annual Meeting, Abstract.
Sen Gupta, B. K., R. E. Turner and N. N.
Rabalais. 1996. Seasonal oxygen depletion in
continental-shelf waters of Louisiana:
Historical record of benthic foraminifers.
Geol. (in press).
Sklar, F. H. and R. E. Turner. 1981.
Characteristics of phytoplankton production
off Barataria Bay in an area influenced by
the Mississippi River. Cont. Mar. Sci. 24: 93-
106.
Turner, R. E. and N. N. Rabalais. 1991.
Changes in Mississippi River water quality
this century. Implications for coastal food
webs. BioScience 41(3): 140-147.
Turner, R. E. and N. N. Rabalais. 1994a.
Coastal eutrophication near the Mississippi
river delta. Nature 368: 619-621.
Turner, R. E. and N. N. Rabalais. 1994b.
Changes in the Mississippi River nutrient
supply and offshore silicate-based phyto-
plankton community responses. Pages
147-150 in K. R. Dyer and R. J. Orth (eds.),
Changes in Fluxes in Estuaries: Implications
from Science to Management. Proceedings
of ECSA22/ERF Symposium, International
Symposium Series, Olsen & Olsen,
Fredensborg, Denmark.
-------
Wiseman, Jr., W. J., V. J. Bierman, Jr., N. N.
Rabalais and R. E. Turner. 1992. Physical
structure of the Louisiana shelf hypoxic
region. Pages 21-26 in Nutrient Enhanced
Coastal Ocean Productivity. Publ. No.
TAMU-SG-92-109, Texas Sea Grant
College Program, Galveston, Texas.
Wiseman, W. J.,Jr.5 E. M. Swenson, andj.
Power. 1990. Salinity trends in Louisiana
estuaries. Estuaries 13: 265-271.
P r es en tation
Eugene Turner (Louisiana State
University— Baton Rouge, LA)
Bob Wayland (US. Environmental Protec-
tion Agency — Washington D.C.) commented
that Eugene Turner seemed to have taken a
bottom-of-the-funnel approach to assessing
wetland destruction or hydro modification of
the river. He asked whether other changes
which took place further up the river system
might have also contributed to the problem and,
therefore, reversal of those upriver problems
might also be considered as part of the solution.
Eugene Turner agreed that the nitrate loading
is a cause-and-effect relationship upriver as well.
However, he pointed out that regardless of the
cause (land-use changes are strongly implicated),
the major issue is the loading at the river mouth.
Robert Wayland replied that he understood
Eugene Turner was discussing overland flow
and loss of the attenuation capacity of the wet-
lands in the lower river, but he was suggesting
that the loss of riparian wetlands upstream
might have also been a contributing factor.
Eddie Funderburg (LSU Agricultural
Center— Baton Rouge, LA) opened his com-
ment by saying that there have been multiple
sources of nutrients identified as contributors to
the river, yet the focus is primarily on fertilizers,
particularly nitrogen fertilizers. He asked if there
have been correlations developed for other
sources, for example, wastewater treatment
plants which have been brought on-line since
1950, automobile emissions, or gasoline con-
sumption. It is possible that the graphic
relationship would look similar to the relation-
ship shown for nitrogen and phosphorous use.
Eugene Turner replied to his comment by say-
ing that for each of the other sources there is
not the same type of relationship or quantity
from atmospheric sources as there is from ferti-
lizer use. Atmospheric sources can, in fact,
come off the farmland as ammonia and deposit
as nitrate. The numbers from sewage plants
demonstrate that the contributions are not as
significant as the source material in terms of
quantity.
Eddie Funderburg then asked Eugene Turner
how 20 percent was derived as the amount of
nutrient contribution from farmlands to the
Mississippi River.
Eugene Turner replied that they assessed the
change in nutrient loading and measured how
much fertilizer was applied in each of those
county/states in the drainage basin. To explain
the changes of nitrate in the river, it is necessary
to assess if enough fertilizer was applied in the
drainage basin to account for the nitrate change
in the river.
Finally, he said that there is enough fertilizer
applied in the Mississippi River Basin to affect
that change and that means that there needs to
be only a 20 percent leakage from the system.
No system has 100 percent retention efficiency.
This is what is typically found in other rivers as
well for agriculture systems.
112
-------
Eddie Funderburg agreed there is some
leakage, but questioned that Eugene Turner
could demonstrate, using field studies, that
20 percent was moving from land into the
Mississippi River.
Eugene Turner countered by saying that
people have traced nitrogen isotopes and
determined whole nitrogen budgets. In fact,
one study was conducted in Iowa in the
1960's.
This conference may or may not decide that to
reduce the load it is necessary to more quanti-
tatively determine the exact sources so that risk
analyses and cost benefit analyses can be
conducted.
Len Bahr made a general comment saying the
last two papers presented touched upon an
important issue. Unfortunately, the format of
the conference was not set up to really discuss
the theoretical questions. He said the coastal
restoration program is not driven by the
hypoxic zone. The need to restore and mitigate
wetland losses of 35 square miles per year is
reason enough for the program to exist. The
theoretical bonus of being able to reduce some
of the eutrophication of the nearshore waters is
real and should be explored. It should be a
major part of the next conference on the
hypoxia area.
Lon Strong (USDA/NRCS-Jackson, MS)
reiterated a comment made by Eugene Turner
that Mitch and his co-author stated that sub-
surface flow systems were a lot more efficient
than surface flow systems. He asked Eugene
Turner if he was referring to man-made or
natural systems. -
Eugene Turner said that because the contact
time is increased, it is a general observation that
natural systems, where effluent flows below
ground, were more efficient than above ground
systems.
Ann Burruss (Coalition to Restore Coastal
Louisiana—Baton Rouge, LA) asked how the
nitrogen concentration in the upper part of the
Barataria Bay would compare to the concentra-
tion in the Mississippi River.
Eugene Turner replied that the nitrogen con-
centration in the upper part of Barataria Bay is
much lower than in the Mississippi River. The
concentrations were calculated by doing a trans-
ect for the past year and will continue for
another year or two. There are 37 stations off-
shore as far as Bayou Seville.
113
-------
Estimated Responses of Wat
Louisiana Inner Shelf to Nu-
in the Mississippi and AtchA
• • - • . . : .^«ill!iiilll
Victor/. B/ermon, Jr.
Umno-Tech, Inc.
South Bend, Indiana 46637
Abstract
The addition of anthropogenic nutrients
from sewage, industrial sources,
agriculture and overland runoff has
contributed to development of eutrophication
in the coastal waters of the northern Gulf of
Mexico. The principal source of these nutrients
is the Mississippi-Atchafalaya River (MAR)
system, the largest single source of freshwater
and nutrient inputs to the coastal waters of the
United States. An extensive, persistent zone of
seasonal hypoxia has been documented in the
nearshore bottom waters of the Louisiana-Texas
continental shelf.
As part of the NOAA Nutrient Enhanced
Coastal Ocean Productivity (NECOP) Program,
a mass balance water quality model was applied
to the Louisiana Inner Shelf (LIS) portion of the
northern Gulf of Mexico. The model was cali-
brated to field data representing summer
average conditions in 1985,1988 and 1990. As
part of the EPA Gulf of Mexico Program,
predictive simulations were conducted with the
calibrated model to estimate responses of
dissolved oxygen and chlorophyll concentra-
tions to potential reductions in nutrient loadings
from the MAR system. The objectives of this
analysis were to determine whether water quality
on the LIS was sensitive to changes in MAR
nutrient loadings and to estimate the approxi-
mate magnitudes of potential reductions in
nutrient loadings that might be necessary to
improve present water quality conditions.
Results indicated that dissolved oxygen and
chlorophyll concentrations on the LIS were
responsive to reductions in MAR nitrogen and
phosphorus loadings. For a given reduction in
MAR nutrient loadings there were large uncer-
tainties in response magnitudes. These uncer-
tainties were due primarily to uncertainties in
relationships among MAR nutrient loadings,
seaward boundary conditions and sediment
oxygen demand, and to inter-annual variability
in hydrometeorology.
The addition of anthropogenic nutrients from
sewage, industrial sources, agriculture and
surface runoff has contributed to development
of eutrophication in the coastal waters of the
northern Gulf of Mexico. The principal source
of these nutrients is the Mississippi—Atchafalaya
River (MAR) system, the largest single source of
freshwater and nutrient inputs to the coastal
waters of the United States. An extensive, per-
sistent zone of seasonal hypoxia has been docu-
mented in the nearshore bottom waters of the
.Louisiana—Texas continental shelf.
As part of the NOAA Nutrient Enhanced
Coastal Ocean Productivity (NECOP) Program,
114
-------
a mass balance water quality model was
developed and applied to the Louisiana Inner
Shelf (LIS) portion of the northern Gulf of
Mexico (Figure 65) (Bierman et al., 1994). The
model was calibrated to field data representing
summer average conditions in 1985, 1988 and
1990. As part of the EPA Gulf of Mexico Pro-
gram, predictive simulations were conducted
with the calibrated model to estimate responses
of dissolved oxygen and chlorophyll concentra-
tions to potential reductions in nutrient loadings
from the MAR system (Limno— Tech, Inc.,
1995).
'Objectives ^
The objectives of this analysis were to deter-
mine whether water quality on the LIS was
sensitive to changes in MAR nutrient loadings
and to estimate the approximate magnitudes of
potential reductions in nutrient loadings that
might be necessary to improve present water
quality conditions, especially seasonal hypoxia.
The purpose of this analysis was not to establish
target nutrient loading objectives, but to deter-
mine the potential range of nutrient loading
reductions that may need to be evaluated in
future studies. An important part of this analysis
was investigation of uncertainties due to differ-
ences in environmental conditions and external
boundary conditions.
The conceptual framework for the modeling
approach is shown in Figure 66. State variables
in the model include salinity, phytoplankton
carbon, phosphorus, nitrogen, dissolved oxygen
and carbonaceous biochemical oxygen demand.
The spatial domain of the model is represented
by a 21 -segment •water column grid extending
from the Mississippi River Delta west to the
Louisiana Texas border, and from the shoreline
seaward to the 30—60 meter bathymetric
contours (Figure 67). The spatial-segmentation
grid includes one vertical layer nearshore and
two vertical layers offshore. The temporal
domain of this model application represents
steady-state, summer-average conditions.
^ " * /
.Approach to. Predictive Sirnulations
The calibrated water quality model was run for a
series of predictive simulations. These simula-
tions involved a range of reductions from 10 to
70 percent on nitrogen and phosphorus load-
Jngs from the MAR system. Emphasis was
placed on comparison of results to base cali-
bration conditions, not on absolute predictions.
To address uncertainties due to differences in
environmental conditions, separate simulations
were conducted for July 1985, August 1988 and
July 1990 for each load reduction. The most
important differences among these three sum-
mer average calibration periods were differences
in MAR inflows and freshwater advective flow
magnitudes and directions on the LIS. To
address-uncertainties in specification of external
boundary conditions, each load reduction simu-
lation was conducted under two separate sets of
assumptions: first, all seaward and sediment
boundary conditions held constant at base cali-
bration values; and second, all seaward and sedi-
ment boundary conditions reduced by the same
percentage as the nutrient loading in each
simulation.
The rationale for two different assumptions on
boundary conditions was twofold: first, these
forcing functions are not computed by the
model but must be externally specified using
available field data; and second, values for these
forcing functions are not independent of MAR
nutrient loadings, but can be expected to
decrease as MAR nutrient loadings decrease.
115
-------
This approach was intended to bracket results
of the predictive simulations between present
conditions and estimates of future conditions
for these forcing functions.
Assumptions
Results of the predictive simulations in this
summary are premised on the following
principal assumptions:
1. The actual environmental system is fully
represented by the conceptual framework of
the model.
2. Nitrogen and phosphorus are the only
nutrients that potentially limit primary
productivity.
3. The actual environmental system is
represented at the coarse spatial scale of the
model segmentation grid. Near-field
gradients in the vicinity of the Mississippi
and Atchafalaya River plumes, and
near-bottom hypoxia, are not explicitly
represented.
4. The actual environmental system is repre-
sented in terms of a single "snapshot" in
time corresponding to an assumed summer
average, steady-state period. The potential
influences of meteorological events, shelf-
edge upwellings and mesoscale shelf
circulation are not explicitly represented.
5. All predictive results represent estimates of
future states of the system and do not
contain any information on the time frame
required for the system to fully respond to
imposed changes in nutrient loadings.
6. All predictive results for reduced boundary
conditions assume that seaward and sedi-
ment boundary conditions will eventually
hange by the same percentage as the
imposed changes in nutrient loadings.
The results presented in this summary are
preliminary results from an ongoing research
program and should be considered provisional
in nature.
Jesuits ofPredictive Simulations
The principal water quality response parameters
were bottom water dissolved oxygen concen-
trations and surface water chlorophyll concen-
trations. Results are presented in terms of com-
parisons among different years, different
response parameters, and loading reductions for
different nutrients. All comparisons are made
using the average of dissolved oxygen responses
for individual bottom offshore segments (Seg-
ments 15—21) and the average of chlorophyll
responses for individual surface offshore seg-
ments (Segments 8-14).
Predicted responses of average dissolved oxy-
gen and chlorophyll concentrations to nitrogen
load reductions are strongly dependent on
assumptions for boundary conditions. For
example, in response to 70 percent nitrogen
loading reductions for 1985 conditions, average
dissolved oxygen concentrations increase by less
than 10 percent for constant boundary condi-
tions and 35 percent for reduced boundary con-
ditions (Figure 68). For the same predictive
simulations, average chlorophyll concentrations
decrease by less than 10 percent for constant
boundary conditions and 60 percent for reduced
boundary conditions (Figure 69).
There are substantial differences in responses of
average dissolved oxygen concentrations among
different years. For example, in response to
70 percent nitrogen loading reductions, average
116
-------
dissolved oxygen concentrations increase by
150 percent for 1990 conditions and
35-40 percent for 1985 and 1988 conditions
under reduced boundary conditions (Figure 70).
In contrast to dissolved oxygen responses, there
are not large differences in average chlorophyll
concentration responses among different years.
In response to 70 percent nitrogen loading
reductions, average chlorophyll concentrations
decrease by 60-70 percent under reduced
boundary conditions (Figure 71).
In general, there are not large differences in
responses of dissolved oxygen or chlorophyll
concentrations between nitrogen and phos-
phorus loading reductions. There was a
tendency, however, for responses to be
somewhat greater for nitrogen loading reduc-
tions than phosphorus loading reductions,
especially for dissolved oxygen responses under
reduced boundary conditions.
There was no evidence of significant inter-
actions between nitrogen and phosphorus
loading reductions in the predictive simulations.
Results of simulations in which nitrogen and
phosphorus loadings were reduced simul-
taneously were generally consistent with results
of simulations in which the more limiting of the
two nutrients was reduced by itself. That is, if
nitrogen was more limiting than phosphorus for
a particular load reduction and set of boundary
conditions, then results for this simulation were
not significantly different when nitrogen and
phosphorus loadings were reduced simul-
taneously by the same percentage.
For 1985 hydrometeorological conditions and
reduced boundary conditions, average chloro-
phyll concentrations are less responsive than
average dissolved oxygen concentrations at
intermediate (10 to 30 percent) nitrogen loading
reductions, and more responsive at higher
(50 to 70 percent) nitrogen loading reductions
(Figure 72). Differences in responses for 1988
conditions follow patterns very similar to those
for 1985 conditions. In contrast to results for
1985 and 1988, average dissolved oxygen
responses for 1990 are much greater than
average chlorophyll responses for a given
nitrogen loading reduction under reduced
boundary conditions (Figure 73). These differ-
ences occur across the entire range of nitrogen
loading reductions from 10 to 70 percent.
Differences in maximum responses between
these two cases are plus 150 percent (dissolved
oxygen) and minus 70 percent (chlorophyll).
Discussion _ ">-
The responses of dissolved oxygen and chloro-
phyll concentrations to reductions in nutrient
loadings from the MAR system are complex
functions of internal model processes and
external model forcing functions. Part of this
complexity is due to the fact that chlorophyll
and dissolved oxygen are non-conservative and
are each tightly coupled to other state variables
in the model. Another aspect of this complexity
is that dissolved oxygen concentration is much
more strongly influenced by sediment boundary
conditions, primarily sediment oxygen demand,
than is chlorophyll concentration. Finally, there
is considerable uncertainty in seaward and sedi-
ment boundary conditions for both the model
calibration periods and the prediction
simulations.
The reasons for differences in responses
between dissolved oxygen and chlorophyll
concentrations are very complex. One reason is
that the relative influence of MAR nutrient
inputs, seaward boundary conditions and bot-
tom boundary conditions differ between the
dissolved oxygen and chlorophyll state variables
in the model. Another reason is that dissolved
oxygen is coupled to more state variables in the
117
-------
model than chlorophyll. Under reduced
boundary conditions, for example, dissolved
oxygen responses represent the integrated
effects of simultaneous changes not only in
dissolved oxygen processes per se, but also of
changes in carbonaceous biochemical oxygen
demand, phytoplankton carbon (through endo-
genous respiration) and ammonia nitrogen
(through nitrification). Still another factor is that
surface chlorophyll and bottom dissolved oxy-
gen concentrations are coupled through the
dependence of underwater light attenuation on
phytoplankton self-shading. That is, reductions
in surface water chlorophyll concentrations can
stimulate bottom water primary productivity
due to increased light penetration, and hence
cause increases in bottom water dissolved
oxygen concentrations.
-Conclusions
The following principal conclusions were drawn
from the data synthesis and modeling simula-
tions conducted in this study:
1. Dissolved oxygen and chlorophyll concen-
trations on the LIS appear responsive to
changes in MAR nitrogen and phosphorus
loadings.
2. For a given reduction in MAR nutrient load-
ings, there are large uncertainties in the
magnitudes of dissolved oxygen and chloro-
phyll concentration responses.
3. Uncertainties in the magnitudes of dissolved
oxygen and chlorophyll concentration
responses are due to three principal factors:
a. uncertainty in the relationship between
MAR nutrient loadings and seaward
boundary conditions
b. uncertainty in the relationship between
MAR nutrient loadings and sediment
oxygen demand
c. inter-annual variability in hydro-
meteorological conditions on the LIS.
4. Responses of average dissolved oxygen
concentrations were more sensitive to
differences in sediment oxygen demand
than to differences in any other
boundary conditions.
5. Responses of average chlorophyll con-
centrations were more sensitive to
differences in seaward boundary con-
ditions than to differences in sediment
nutrient boundary conditions.
6. Although differences in results between
nitrogen and phosphorus loading reduc-
tions were generally not large, there was
a tendency for responses to be some-
what greater for nitrogen loading reduc-
tions than phosphorus loading reduc-
tions, especially for dissolved oxygen
under reduced boundary conditions.
7.
to
Estimates of water quality responses ro
changes in MAR nutrient loadings must
be premised on specific assumptions for
hydrometeorological conditions on the
LIS.
•BBS"*' *w «£ZM£4^F*-& A^i a- jt / *• ^ ^~ w jftrf- v
L@£omjmj&n.aatiojns . /,
On the basis of the data synthesis and modeling
simulations conducted in this study, the follow-
ing principal recommendations are made:
1. The temporal domain of the present water
quality model should be extended to include
a time-variable representation of water
quality conditions on the LIS during the
period of vertical stratification.
2. The vertical scale of the present model seg-
mentation grid should be refined to better
represent near-bottom hypoxia on the LIS.
118
-------
3. The spatial domain of the present model
segmentation grid should be extended so
that its seaward boundaries are beyond the
influence of freshwater and nutrient inputs
from the Mississippi and Atchafalaya Rivers.
4. Advective flows and dispersive mixing
coefficients in the model should be deter-
mined using the output of a hydrodynamic
model.
5. The conceptual framework of the model
should be expanded to include dissolved
oxygen processes in the sediment and an
explicit dissolved oxygen mass balance
between water column and sediment
segments.
6. The conceptual framework of the model
should be expanded to include diatom and
non-diatom phytoplankton functional
groups, and silicon as a potential limiting
nutrient.
Bierman, V.J., Jr., S.C. Hinz, D. Zhu, W.J.
Wiseman, Jr., N.N. Rabalais and R.E.
Turner. 1994. A preliminary mass balance
model of primary productivity and dissolved
oxygen in the Mississippi River Plume/
Inner Gulf Shelf region. Estuaries.
17(4):886-899.
Limno-Tech, Inc. 1995. Estimated Responses
of Water Quality on the Louisiana Inner
Shelf to Nutrient Load Reductions in the
Mississippi and Atchafalaya Rivers. Report
prepared for Louisiana State University and
A&M College, Baton Rouge, Louisiana, and
submitted to U.S. Environmental Protection
Agency, Gulf of Mexico Program Office,
Stennis Space Center, Mississippi.
119
-------
»t
I
Gulf of Mexico
92
I
Figure 65.
Location map of study area.
External Source
Loads
Conceptual Framework for Water Quality Model
Advaction and
Dispersion
Boundary
Conditions
Sediment Flux
Temperature
Light
120
Figure 66.
Schematic diagram of principal model state variables and processes.
-------
94
9J
—I—
92
91
90
T~
69
30
29
30 -
29
94
'
92
90
89
—I—
92
91
Si
30
29
30 -
29
94
- I
SO
89
Figure 67.
Model spatial segmentation grid for the Louisiana Inner Shelf.
121
-------
1985 - N Reductions
(D
V)
CO
CO
CD
f>
•*-*
JO
CD
DC
(D
O)
C
CO
-C
O
tz
CD
O
o>
Du
10
20 30 50
Percent Load Reduction
Constant BC
Reduced BC
Figure 68.
Predicted responses of average dissolved oxygen concentrations to nitrogen loading
reductions for 1985 conditions under constant and reduced boundary conditions.
1985 - N Reductions
CD
V)
CO
CO
CO
CC
CD
CD
c
O
O
CD
D_
0
-10
-20
-30
-40
-50
-60
-70
-80
-90
-100
—i—
10
y
20 30 50
Percent Load Reduction
70
Constant BC
Reduced BC
Figure 69.
Predicted responses of average chlorophyll concentrations to nitrogen loading reductions
for 1985 conditions under constant and reduced boundary conditions.
122
.
-------
N Reductions -
Boundaries Reduced
CD
\&S
03
CD
O
•* — »
CD
.5:
"CD
oc
CD
CO
c
cs
O
«#—*
CD
O
rll
200 -^
1 80 -i
160-
140-
120 -
100-
80-
60-
40-
20-
|
!
>
0 J£-
Q_
20 30 . 50
Percent Load Reduction
70
1985
1990
1988
Figure 70.
Predicted responses of average dissolved oxygen concentrations to nitrogen loading
reductions for 1985, 1988 and 1990 conditions.
(S3
V)
CCJ
CD
o
,*««
0)
JXf
CD
DC
CD
O)
C
CO
jr
O
CD
O
l_
CD
Q.
N Reductions - All Boundaries Reduced
20 30 50
Percent Load Reduction
70
1985
1990
1988 Wb
Figure 71.
Predicted responses of average chlorophyll concentrations to nitrogen loading reductions
for 1985, 1988 and 1990 conditions.
123
-------
N Reductions - All Boundaries Reduced
CD
V)
CO
CO
CD
CD
DC
CD
CD
c
CO
JC
O
•+"-*
c
c:
o>
o
i_
CD
Q.
200-
160-
120 -
80-
40-
0-
-40-
-80-
-120-
-160-
-200
10
20 30 50
Percent Load Reduction
70
••I DO Chlorophyll
Figure 73.
Predicted responses of average dissolved oxygen and chlorophyll concentrations to
nitrogen loading reductions for 1990 conditions.
124
-------
t j*resejntattorT Discussio.n, _
Vic Bierman (Limno-Tech, Inc.—South
Bend, IN)
Mr. Bierman left conference due to an
emergency.
125
-------
John Day
Louisiana State University
Baton Rouge, Louisiana 70803
Abstract
No abstract submitted.
No manuscript submitted.
-Presentation Discussion ]
John Day (Louisiana State University—Eaton
Rouge, LA)
Nancy Rabalais (Louisiana State
University—Baton Rouge, LA) asked if John
Day knew what the partitioning was of the
freshwater outflow through the Atchafalaya Bay
because her offshore data (which is limited)
indicates that the nutrients (in Fourleague Bay)
are central and west of study area due to the
long shore current movement and they are
entering the Gulf through the Atchafalaya Bay
area.
John Day responded that the estimates of
partitioning are from five to ten percent flow
through the study area, and it is possible that
the flow goes in the direction she indicated, but
there is a lack of information to determine for
certain.
Daniel Ray (The McKnight
Foundation—Minneapolis, MN) asked if all
the wetlands need to be located in the Gulf or if
sufficient uptake would occur if the wetlands
were located closer to the flood plain.
John Day responded that the solution to the
problem is not just in the Delta, bust also for
126 , -
upstream, and that nitrate can be removed very
rapidly.
Don Boesch (University of Maryland-
Cambridge, MD) commented that some of
what has been shown is the sink of nitrogen and
some of it is conversion of organic nitrogen
into ammonium. He then asked what the net
permanent loss was through de-nitrification or
burial of nitrogen, in Fourleague Bay as
opposed to a marsh.
John Day replied that he has talked to some
people regarding this topic and has found that
usually among concentrations of nitrogen,
ammonium, and TKN, nitrate is much higher;
in excess of 50 percent. Someone has looked at
their wetland systems and almost 90 percent
goes to denitrification. It is possible that more
than 50 percent is lost through denitrification in
Fourleague Bay.
Len Bahr (Louisiana Governor's Office -
Baton Rouge, LA) commented that if the
losses in the Atchafalaya are approximately 30
percent, and the lower Mississippi losses are 70
percent, and one is constrained during the
spring, and one has no seasonal constraints, a
synoptic study of nutrient dynamics down both
branches extending to the nearshore area would
be useful. There should be more nutrient
uptake calibrated for the different flows of
sediment levels in the Atchafalaya.
John Day agreed, saying that the study was
absolutely necessary before further work on the
project can proceed.
-------
Abstract
/ . _ t. •„ _. /•
The quantification of the regional
transport of nutrients to the Gulf of
Mexico is important to developing
management strategies for reducing the hypoxic
zone observed in recent summers on the
Louisiana coastal shelf. Although existing
research clearly identifies the Mississippi and
Atchafalaya Rivers as the primary conduits for
nutrients, the origin (type and location) of the
sources of nutrients in these rivers is less cer-
tain. Better estimates of the quantities of point-
and nonpoint-source nutrients delivered to the
Gulf of Mexico from interior watersheds could
improve the efficiency of management
strategies.
To assist in identifying the origin of stream
nutrients nationally, we developed a water-
quality model of nutrient flux in rivers of the
United States. This model allows us to estimate
the origin of point and nonpoint source nutrient
flux at numerous locations on the coastal mar-
gin including the outlets of the Mississippi and
Atchafalaya Rivers. The regression-based water-
quality model relates monitored nutrient flux
from 430 watersheds to various measures of
upstream pollutant loadings, such as industrial
and municipal discharges, fertilizer application,
animal manure, and atmospheric deposition.
The monitored watersheds range in size from
several hundred to several tens of thousands of
square miles. Flux estimates are developed from
regularly-collected season nutrient measure-
ments and daily estimates of streamflow using
log-regression rating curve techniques. The
estimated loadings of nonpoint-source nutrients
to streams include the effects of watershed
physical characteristics including precipitation,
soil permeability, and topography. The model
also estimates the first-order decay of nutrients
during the transport of point and nonpoint
sources through a digital stream network of
nearly one million kilometers and 60,000
reaches. These decay rates reflect time-of-travel
estimates from field studies and the residence
time of water in major reservoirs.Through
application of the model to unmonitored
reaches, we estimate the quantities of point and
nonpoint source nutrients delivered to the Gulf
from several interior watersheds of the
Mississippi and Atchafalaya Basins, including
the Missouri, Arkansas, Upper Mississippi, and
Ohio River Basins. All model predictions are
accompanied by estimates of statistical error.
t ' " • "~ ' ^ '""
Introduction ,
In recent summers, a large area of very low
dissolved oxygen concentrations (i.e., the
hypoxic or "dead" zone) has appeared on the
Louisiana coastal shelf (Rabalais and others,
1994). Although researchers have observed
linkages between the nutrient-enriched waters
of the Mississippi and Atchafalaya Rivers and
127
-------
spatial and temporal variations in the hypoxic
zone (e.g., Rabalais and others, 1994; Justic and
others, 1993), the origin of nutrients in these
rivers is uncertain. The development of efficient
management strategies for reducing the hypoxic
zone depends on better estimates of the point
and nonpoint-source nutrients delivered to the
Gulf from interior watersheds of these rivers.
We developed a water-quality model of nutrient
flux in rivers of the United States to assist in
quantifying the origin of stream nutrients
nationally. This model allows us to estimate the
origin of point- and nonpoint-source nutrient
flux at numerous locations on the coastal
margin including the outlet of the Mississippi
River. The regression-based water quality model
relates monitored total nitrogen flux from
430 watersheds to various measures of upstream
pollutant loadings, including industrial and
municipal treatment plant discharges, fertilizer
application, animal manure, and atmospheric
deposition (R2= 0.83). The monitored water-
sheds range in size from several hundred to
several tens of thousands of square miles. Mean
estimates of flux for the period 1985-88 were
computed from regularly-collected seasonal
nutrient measurements and daily estimates of
streamflow using log-regression rating curve
techniques. The estimated loadings of
nonpoint-source nutrients to streams include
the effects of watershed physical characteristics
including precipitation, soil permeability, and
topography. The model also estimates the
first-order decay of nutrients during the trans-
port of point and nonpoint sources through a
digital stream network of nearly one million
kilometers and 60,000 reaches. These decay
rates reflect time-of-travel estimates from field
studies and the residence time of water in major
reservoirs.
Through application of the model to the
Mississippi River and its tributaries, we
estimated the quantities of total nitrogen
delivered to the Gulf from several interior
watersheds including the Missouri, Ohio,
White/Red, and the Upper, Central, and Lower
Mississippi River Basins (Figure 74). These
estimates indicate that more than 70 percent of
the total nitrogen delivered to the Gulf by the
Mississippi River originates above the con-
fluence of the Ohio and Mississippi Rivers. This
nitrogen is transported over distances of more
than 1000 miles. The Upper and Central Mis-
sissippi Basins, which include portions of the
states of Minnesota, Wisconsin, Iowa, Missouri,
and Illinois, account for the largest quantity of
nitrogen (39 percent) delivered to the Gulf.
Smaller fractions originate in the Ohio
(22 percent) and the Missouri (11 percent) River
Basins. Downstream from the Mississippi/Ohio
River confluence, the Lower Mississippi Basin,
which drains portions of the states of Tennes-
see, Arkansas, Missouri, Mississippi, and Louisi-
ana, contributes nearly a quarter of the nitrogen
to the Gulf, whereas the White/Arkansas River
Basins contribute six percent of the nitrogen.
The development of economically efficient
nitrogen removal strategies in the Mississippi
River Basin requires consideration of many
factors, including the benefits (i.e., nitrogen
reductions) expected from the application of
controls in different interior watersheds. Esti-
mates of each watershed's contribution of
nitrogen to the Gulf per unit of drainage area
(i.e., yield; see Figure 75) may be used to
approximate the level of benefits to the Gulf
expected per unit of drainage area receiving
controls. Accordingly, the largest benefits to the
Gulf per unit area controlled would be expected
from nutrient controls applied in the Lower,
Central, and Upper Mississippi River Basins
where per unit area nitrogen contributions
exceed those in the Ohio, White/Arkansas, and
Missouri Basins by more than a factor of two.
Although the Lower Mississippi Basin accounts
128
-------
for less than 25 percent of the Mississippi
River's nitrogen contribution to the Gulf, the
benefits of nutrient controls in this basin would
be expected to be more than twice as large as
those in other watersheds. The large per unit
area nitrogen contribution from the Lower Mis-
sissippi Basin (Figure 75) reflects the compara-
tively small drainage area of the basin and its
proximity to the Gulf which lead to fewer losses
of nitrogen.
In applying the nutrient model, we separately
tracked the contributions of point and nonpoint
sources of stream nitrogen to the Gulf. On the
basis of these analyses, we estimate that
approximately 90 percent of the nitrogen
delivered to the Gulf by the Mississippi River
originates from nonpoint sources consisting
predominantly of nitrogen in agricultural runoff
and atmospheric deposition. More detailed
applications of the nutrient model will be
needed to resolve the relative importance of
these two sources. Only about one percent of
the nitrogen comes from point sources in the
effluent of municipal treatment plants and
industries. The remaining nitrogen delivered to
the Gulf (9 percent) is from unknown sources
as estimated by the intercept of the regression
model. These unspecified sources may poten-
tially include inputs of nitrogen from ground
water.
- Refefencesf - ,_ ;
Justic, D., Rabalais, N.N., Turner, E.R., and
Wiseman, W.J., 1993, Seasonal Coupling
Between Riverborne Nutrients, Net Pro-
ductivity and Hypoxia, Marine Pollution
Bulletin, v. 26, no. 4, pp. 184-189.
Rabalais, N.N., Turner, R.E., Wiseman, W.J.,
Justic, D., Dortch, Q., and Gupta, B.S.,
1994, Hypoxia on the Louisiana Shelf and
System Responses to Nutrient Changes in
the Mississippi River: A Brief Synopsis, in
National Oceanic and Atmospheric Admini-
stration, Nutrient-Enhanced Coastal Ocean
Productivity, Proceedings of the 1994
Synthesis Workshop, Baton, Rouge,
Louisiana.
129
-------
Total Nitrogen Flux to the Gulf of Mexico from
Interior Watersheds of the Mississippi River Basin
Standardized for Drainage Basin Size
S States
n Yield (kg/km2/yr)
|I7J 83 - Missouri
108 - White/Arkansas
437 - Ohio
708 - Upper Miss.
1652 - Central Miss.
2072 - Lower Miss.
Figure 74.
The U.S. Geological survey recently estimated the quantities of total nitrogen delivered to the Gulf of Mexico
from several interior watersheds of the Mississippi River. These estimates, based on river monitoring data and
a regression-based water-quality model, indicate that more than 70 percent of the nitrogen originates above
the confluence of the Ohio and Mississippi Rivers and is transported over distances of more than 1000 miles.
Downstream from the Mississippi/Ohio River confluence, the Lower Mississippi and the White/Arkansas River
Basins contribute nearly 30 percent of the nitrogen. Model estimates also indicate that approximately 90
percent of the nitrogen delivered to the Gulf by the Mississippi River originates from nonpoint sources. Of the
remaining nitrogen, one percent is from point sources and nine percent is from unknown sources.
130
aK3TW«s«as5sa««^^
-------
Percentage of the Mississippi River
Total Nitrogen Flux to the
Gulf of Mexico from Interior Basins
States
3asin Share in Percent
6 - White/Arkansas
8 - Central Miss.
11 - Missouri
22 - Ohio
23 - Lower Miss.
31 - Upper Miss.
Figure 75.
Estimates of the quantities of total nitrogen delivered to the Gulf of Mexico from several interior watersheds
of the Mississippi River per unit of drainage area (kilograms/square kilometer/year) can be used to
approximate the level of benefits (i.e., nitrogen reductions) to the Gulf expected per unit of drainage area
receiving controls. Developing economically efficient nitrogen removal strategies in the Mississippi Basin
requires consideration of many factors, including the benefits expected from the application of controls in
different interior watersheds. The largest benefits to the Gulf per unit area controlled would be expected from
nutrient controls applied in the Lower, Central, and Upper Mississippi River Basins where per unit area
nitrogen contributions exceed those in the Ohio, White/Arkansas, and Missouri Basins by more than a factor
of two. Although the Lower Mississippi Basin accounts for less than 25 percent of the Mississippi River's
nitrogen contribution to the Gulf (see Figure 74), the benefits of nutrient controls in this basin would be
expected to be more than twice as large as those in other watersheds due in part to the Lower Mississippi
Basin's proximity to the Gulf.
^xr^^
131
-------
^Presentation Discussion j
Richard Alexander (U.S. Geological Survey—
Reston, VA)
Fred Bryan (National Biological Survey/
LSU) asked if the value Richard Alexander
presented for Nitrogen contribution to the river
load by the Mississippi River Basin represented
a cumulative value or had the contribution of
the rest of the watershed been subtracted.
He also speculated that the relative contribution
of point and nonpoint source loadings in the
lower basin could be estimated by examining
the Red-Atchafalaya River System. Most of the
point sources in the lower basin are in the
industrial corridor between Baton Rouge and
New Orleans. Only three or four point sources
from Alexandria are in the Red-Atchafalaya
system and since the Atchafalaya is 70 percent
Mississippi River water by volume, the nonpoint
source load could be estimated by comparing
the loading in the Red-Atchafalaya system with
the loading in the Mississippi River below New
Orleans.
Richard Alexander responded to Fred Bryan's
question by saying that 23 percent that he cited
for total nitrogen is a net contribution. The
Arkansas and White Rivers, as well as the upper
basin, are subtracted from that value.
In response to Fred Bryan's comment, Richard
Alexander replied that they chose not to look at
the Atchafalaya, mainly because they do not
have a monitoring station at the outlet that
meets the criteria for load estimation. He agreed
that it is something that needs to be examined
and if they did have a monitoring station there it
could be included in the budget and the analysis
Fred Bryan referred to could be conducted.
132
-------
Abstract -, -__1 " \t kX
Concentrations of dissolved nutrients
(nitrate, nitrite, ammonium and
orthophosphate) were measured on
surface-water grab samples collected at ten mile
intervals along the entire length of the navigable
portion of the Mississippi River during three
cruises in June—July 1991, September—October
1991 and March—April 1992. Samples were also
collected at the mouths of some of the major
tributaries and, at selected points, three-sample
cross sections were collected across the river to
measure cross-channel variability.
Simultaneously with collecting each sample, the
discharge of the Mississippi River was
estimated, permitting the calculation of the
nutrient load in the Mississippi River at each
sample point. The large number of samples
collected (between 179 and 207 per cruise) give
a picture of the instantaneous longitudinal
variation of nutrients in the Mississippi River
during three different seasons.
Both nitrate and orthophosphate loads appear
to increase or remain constant downriver for
each cruise, indicating that mechanisms for the
removal of these compounds are not as rapid as
their introduction into the river. In addition,
below the confluence of most of the major
tributaries, the loads show a "step" increase
caused by the nitrate and orthophosphate
contributions from the tributaries. From this
data, one can identify the possible sources of
nitrate and orthophosphate arriving at the Gulf
of Mexico from the Mississippi River. During
each of the three cruises (at three different
seasons), the majority of the nitrate and ortho-
phosphate appears to have originated from the
Upper Mississippi River Basin, above the con-
fluence of the Missouri River.
Ammonium and nitrite loads appear to originate
as point-sources, but disappear within approxi-
mately one hundred miles of their introduction,
probably as a result of conversion to nitrate
and/or nitrogen gas. Any nitrite or ammonium
from the Mississippi River deposited into the
Gulf of Mexico therefore probably originated
within one hundred miles of the Gulf.
Introduction and
Three sampling cruises were taken on the Mis-
sissippi River, originating near New Orleans,
Louisiana and finishing in Minneapolis, Min-
nesota during June 23 to July 2, 1991, Sept-
ember 25 to October 4, 1991 and March 25 to
April 4, 1992. One purpose of these cruises was
to study how concentrations and transport of
the dissolved nutrients nitrate, nitrite, ammo-
nium and orthophosphate ions varied longitudi-
nally in the Mississippi River and at the mouths
133
-------
of some of its major tributaries. Concentrations
of these dissolved nutrients were measured on
surface-water grab samples collected at ten-mile
intervals along the entire length of the navigable
portion of the Mississippi River. These data can
be used to evaluate the potential effects that
nutrients have on the Mississippi River system
and on the Gulf of Mexico.
Grab samples collected from the upper three
meters of the river were taken every ten miles
(about once an hour) from the center of the
river channel. In addition, samples were col-
lected from the mouth of some of the major
tributaries; at selected locations, three samples
were collected along cross-sections to measure
cross-channel variability. Samples were collected
from the river with a clean 2-liter Teflon bottle
placed in a weighted aluminum holder and were
then transferred into pre-cleaned 250-ml opaque
polyethlene bottles. All samples were immedi-
ately filtered through nylon or Nuclepore poly-
carbonate membrane filters with a 0.4-um
nominal pore diameter, and then either chilled
(for immediate analysis) or frozen for transport
to the laboratory. No chemical preservatives
were used.
Analyses were performed on an Alpkem air-
segmented continuous-flow colorimetric analy-
sis system, Model RFA-300, as described in
greater detail by Antweiler et al. (1994).
Determinations were always performed in
duplicate; if the two determinations did not
agree within the variance of the method, the
sample was reanalyzed, again in duplicate.
Analyses were supplemented by the determina-
tion of standard reference materials and cali-
bration standards to evaluate precision and
accuracy. Details of all the above information
are given by Antweiler et al. (1995a).
The discharge of the river was estimated at the
time of sample collection (Moody, 1995),
permitting the calculation of the nutrient load in
the Mississippi River at each sample point. The
large number of samples collected (about
200 per cruise) in the short time periods (about
ten days per cruise) give a picture of the near
instantaneous longitudinal variation of nutrients
in the Mississippi River during three different
seasons.
At three sites on the Mississippi River—
Clinton, Iowa, Thebes, Missouri and Baton
Rouge, Louisiana—and near the mouths of
three tributaries (the Ohio River at Grand
Chain, Illinois, the Missouri River at St. Charles,
Missouri and the Illinois River at Valley City,
Illinois), nutrient samples were collected
biweekly from April 1991 to September 1992.
These samples were discharge-weighted
laterally-composited samples collected in glass
or stainless steel containers. They were filtered
through a 0.45-um membrane filter immediately
after collection, preserved with mercuric chlor-
ide and shipped chilled to the USGS National
Water Quality Laboratory, Denver, CO, for
analysis. These samples were analyzed using an
automated colorimetric procedure (Fishman and
Friedman, 1989). Details of both the sampling
protocol and analyses can be found in Coupe et
al. (1995).
The results of these studies are tabulated by
Antweiler et al. (1995a) and Coupe et al. (1995)
and described by Antweiler et al. (1995b). The
distribution of nitrate and orthophosphate ion
concentrations appear to be similar for all three
cruises. Concentrations are low above Minne-
apolis, increase rapidly below the confluence
with the Minnesota River (due to its relatively
high concentrations compared to the Missis-
sippi River, around mile 1800), generally
decrease through southern Minnesota and
134
-------
Wisconsin (miles 1500-1800), increase through
Iowa (miles 1250-1500), remain nearly constant
through Missouri (miles 950-1250), decrease at
the confluence with the Ohio River (mile 950),
and generally remain constant downstream to
the Gulf of Mexico (Figure 76). In terms of
transport (or loads: they are synonymous terms),
these two compounds demonstrate some
important features. Both nitrate and ortho-
phosphate loads appear to either increase or
remain constant downriver for each cruise,
indicating that mechanisms for the removal of
these compounds are not as rapid as their
introduction into the river. In addition, below
the confluence of most of the major tributaries,
the loads show a "step" increase caused by
contributions from the tributaries (Figure 77).
The data collected biweekly at six sites in the
Mississippi River Basin also provide informa-
tion concerning the transport of nutrients in the
Mississippi River. Nitrate transports based on
these data are shown in Figures 78 and 79. The
pattern of nitrate transport of nitrate at Thebes,
Missouri (just above the confluence with the
Ohio River) is similar to the pattern at Baton
Rouge, Louisiana, during the entire sampling
period although the quantity of nitrate at
Thebes is less than at Baton Rouge (Figure 78).
In contrast, the quantity of nitrate at the mouth
of the Ohio River (Figure 79) is considerably
less than in the Mississippi River at Thebes, and
comparable with that from the Missouri and
Illinois Rivers.
From the biweekly and upriver cruise data, one
can postulate the possible sources of nitrate and
orthophosphate arriving at the Gulf of Mexico
from the Mississippi River. By integrating the
biweekly data over the course of the year, April
1991 to April 1992, and assuming that 30 per-
cent of the Mississippi River is diverted into the
Atchafalaya River above Baton Rouge, Louisi-
ana, it is apparent that the majority of the water
originates from the Ohio River (38 percent) and
the Lower Mississippi River—the Mississippi
River below the confluence with the Ohio River
(28 percent) (Figure 80). However, the majority
of the nitrate appears to have originated from
the Upper Mississippi River Basin, above the
confluence with the Ohio River. The Upper
Mississippi, the Illinois and the Missouri Rivers
account for 68 percent of the nitrate. For ortho-
phosphate, the largest sources again appear to
be the Upper Mississippi, Illinois and Missouri
Rivers (52 percent), although a large percentage
also comes from the Lower Mississippi River
below the confluence with the Ohio River
(29 percent) (Figure 80).
Ammonium and nitrite ion loads appear to
originate as point-sources, but disappear within
approximately one hundred miles of their site of
introduction (Figure 81 shows typical data for
ammonium), probably as a result of either con-
version to nitrate ion and/or nitrogen gas or as
a result of sorption to suspended sediment. Any
nitrite or ammonium ions from the Mississippi
River deposited into the Gulf of Mexico, there-
fore, probably originated within one hundred
miles of the Gulf. However, the amount of
nitrite and ammonium ions in the Mississippi
River are always minor (less than 5 percent)
compared with the amount of nitrate ions and
therefore contribute little to the overall quantity
of nutrients arriving at the Gulf of Mexico from
the Mississippi River.
1 Rjpfereriges '; ' - - " '-'-
Antweiler, R.C., Patron, C.J. and Taylor, H.E.
(1994) Automated, colorimetric methods for
the determination of nitrate plus nitrite,
nitrite, ammonium and orthophosphate ions
in natural water samples: U.S. Geological
Survey Open-File Report 93-638, 34 p.
135
-------
Antweiler, R.C., Fatten, C.J. and Taylor, H.E.
(*95a) Chapter 3: Nutrients in Chemical data
for water samples collected during four
upriver cruises on the Mississippi River
between New Orleans, Louisiana and
Minneapolis, Minnesota, May 1990-April
1992, J.A. Moody, ed.: U.S. Geological
Survey Open-File Report 94-523, p.89-125.
Antweiler, R.C., Goolsby, D.A. and Taylor,
H.E, (199Sb) Nutrients in the Mississippi
River. I: Contaminants in the Mississippi
River, 1987-1992, R.H. Meade, ed.: U.S.
Geological Survey Circular 1133, p. 72-85.
Coupe, R.H., Goolsby, D.A., Iverson, J.L.,
Markovchick, D.J. and Zaugg, S.D. (1995)
Pesticide, nutrient, water-discharge and
physical-property data for the Mississippi
River and some of its tributaries, April
1991-September 1992: U.S. Geological
Survey Open File Report 93-657,116 p.
Fishman, M.J. and Friedman, L.C. (1989)
Methods for the determination of inorganic
substances in water and fluvial sediments::
U.S. Geological Survey Techniques of Water
Resources Investigations, book 5, chap. Al,
p. 1-13.
Moody, J.A. (1995) Chapter 1: Introduction in
Chemical data for water samples collected
during four upriver cruises on the
Mississippi River between New Orleans,
Louisiana and Minneapolis, Minnesota, May
1990-April 1992, J.A. Moody, ed.: U.S.
Geological Survey Open-File Report
94-523, p. 1-18.
Use of tradenames is for identification purposes
only and does not imply endorsement by the
U.S. Geological Survey.
136
.
-------
10 F
Mainstem
Cross Channel Variability
Tributaries
2000 1500 1000 500
River miies from Head of Passes, LA
Figure 76.
Concentration of nitrate in milligrams of nitrogen per liter in the
Mississippi River during June 23-Jufy 2, 1991.
MN = Minnesota River
SC = St Croix River
CH = Chippewa River
BL = Black River
Wl = Wisconsin River
RK = Rock River
SK = Skunk River
DM - Des Moines River
IL = Illinois River
MO = Missouri River
OH = Oh/o River
AR = Arkansas River
3X-l3&i£?!!3&a3i3SZS£$!K^^
137
-------
I
3
•5
s
ti
o
o.
V)
DM
MN
~ Instantaneous Load
Tributary Load
J_
2000
1500 1000 500
River Miles from Head of Passes, LA
0
Figure 77.
Transport of nitrate in millions of kilograms of nitrogen per day in the
Mississippi River during June 23-Jufy 2, 1991.
MN = Minnesota River
DM = Des Mo'mes River
IL = Illinois River
MO = Missouri River
OH = Ohio River
AR = Arkansas River
138
-------
Z
cn
c
o
c
t
I
I
~ Clinton, IA
, MO
Baton Rouge, LA
3/1/91 6/1/91 9/1/91 12/1/91 3/1/92
Date
6/1/92
9/1/92
Figure 78.
Transport of nitrate in millions of kilograms of nitrogen per day in the Mississippi River at
Clinton, Iowa; Thebes, Missouri; and Baton Rouge, Louisiana during 1991-1992.
139
-------
DI
O
g
O
•c
p
O.
(A
4
0 -
-J3"~ Ohio River
—^l-- Missouri River
-r^i- Illinois River
J_
3/1/91
6/1/91
9/1/91
12/1/91
3/1/92
6/1/92
9/1/92
Figure 79.
Transport of nitrate in millions of kilograms of nitrogen per day in three tributaries
of the Mississippi River during 1991-1992.
140
•
-------
Missouri River Illinois River
9% 4%
Discharge
(Water)
Upper Mississippi River
46%
Upper Mississippi River
21%
Lower Mississippi River
28%
Upper Mississippi River
31%
Orthophosphate
Ohto River
19%
_„_.»—» Lower Mississippi River
29%
Figure 80.
Sources of water, nitrate and orthophosphate in the Mississippi River
during April 1991-April 1992.
^
141
-------
20
o>
JC
•5
w
TJ
C
OS
W
o
o
CL
«
JO
10
I '
Instantaneous Load
tTributary Load
SC
2000
1500 1000 500
River Miles from Head of Passes, LA
Figure 81.
Transport of ammonium in thousands of kilograms of nitrogen per day in the
Mississippi River during September 25-October 4, 1991.
MO = Missouri River OH = Ohio River
SC = St Cro/x R;Ver
142
;:K2^«SSKiim!ffi32!CTSK^^
-------
Presentation Discussion
Ron Antweiller (U.S. Geological Survey—
Boulder, CO)
Fred Kopfler (Gulf of Mexico Program—
Stennis Space Center, MS) commented that
the scale of the loadings presented for
ammonium nitrogen was thousands of
kilograms per day, while loadings for nitrate
were millions of kilograms per day; even if the
ammonium was converted to nitrate, it is about
three orders of magnitude less. Therefore, it is
lost in the background of the nitrate.
Ron Antweiller confirmed that Fred Kopfler's
understanding was correct. The highest
ammonium loadings in relation to nitrate he has
seen was five percent. Even if all the
ammonium is converted to nitrate, it still is not
significant.
143
-------
m-ff >" -v. 'f-- -: gttnKirs
Effects of EpiSQ<
Nutrients" to the Gu
1 Tr---:i •* '^IM
I ;.•;• ; ;fW{W
D. A. Goolsby ahd W^f^^
US. Geological Surve? -V'Mjj^^
Water Resources Division I P '$
Denver, Colorado Ill j m^ii
LAbstract
Nutrients (nitrogen and phosphorus)
derived from areas of intense
agriculture in the upper Mississippi
River basin have been implicated as the indirect
cause of oxygen depletion (hypoxia) in the Gulf
of Mexico along the Louisiana—Texas coast. The
largest influx of nutrients to the Gulf typically
occurs each year during the spring and early
summer when streamfiow and concentrations of
nutrients, such as nitrate, are highest. During
extreme high flow episodes, such as the 1993
flood in the upper Mississippi River, abnormally
large amounts of nitrate and other nutrients are
transported into the Gulf. During April through
September 1993, for example, the nitrate flux to
the Gulf was more than 900,000 metric tons (as
N). This is 100 percent more nitrate than was
discharged to the Gulf during this same period
in 1992 and 1994, and 50 percent more than in
1991 and 1995. While these episodic events
cause considerable year-to-year variation in flux
of nutrients, there is evidence that annual fluxes
also have increased. An analysis of historical
water chemistry data collected at St. Francisville,
Louisiana and Baton Rouge, Louisiana since
1954 shows that the concentration and flux of
nitrate in water discharged to the Gulf has
increased about threefold, with most of the
increase occurring since 1968. Conversely, the
concentration and flux of total phosphorus has
changed little since 1973 when the first
phosphorus records were collected.
The principal areas contributing nutrients to the
Mississippi River and ultimately the Gulf of
Mexico are streams draining the corn belt states,
particularly Iowa, Illinois, Indiana, Ohio, and
southern Minnesota. About 60 percent of the
nitrate transported by the Mississippi River is
derived from less than 20 percent of the basin.
Current sources of nitrogen for the Mississippi
River basin, in decreasing order of their input
include commercial fertilizers, animal manures,
legumes, municipal and domestic wastes, and
atmospheric deposition. The present use of
nitrogen fertilizer in the basin is estimated to be
about 6.6 million metric tons per year and
accounts for more than one-half the annual
nitrogen input.
No Manuscript Submitted.
____ ' g <
Discussion
Don Goolsby (U.S. Geological Survey —
Lakewood, CO)
Len Bahr (Louisiana Governor's Office-
aton Rouge, LA) congratulated the U.S.
Geological Survey on presenting excellent
information and he added that it was the type of
information that was needed. He continued by
144
-------
wishing the USGS well on receiving their
budget from Congress. He said he was an
ecologist, not a soil scientist or an agricultural
person, and felt that the information presented
was good news. There is an enormous potential
for cost savings in nitrate fertilizer application
and it certainly appeared that changes in
fertilizer use and application could be achieved.
Don Goolsby asked if someone was present
from the Department of Agriculture, because he
wanted a representative to comment on Len
Bahr's statement. He said he knew the
Department of Agriculture conducts many
programs through which they attempt to
account for the nitrogen already in the soil and
give credit for that; applying less nitrogen in
current years. His understanding was that these
programs were working to some degree, and
attributed some of the leveling in contributions
to those efforts.
Eugene Turner (Louisiana State Univers-
ity—Baton Rouge, LA) asked two questions:
• Is some double accounting of manure inputs
and fertilizer application.
• Secondly, he questioned if anything can be
done to improve sampling at the national
water quality stations. He cited that some of
those stations are sampled only six or eight
times a year.
Don Goolsby responded to the first question
by saying that there is some double accounting
of manure inputs because the fertilizer produces
feed that the animals use.
In response to Eugene Turner's second ques-
tion, Don Goolsby said that he had mentioned
that the N AWQA network was being rede-
signed. The sampling frequency will be doubled
or tripled over the past frequency, but the
number of stations will be reduced to 30 or
40 nationwide and will be focused in four large
river basins.
145
-------
•Ml * • V • U 'I.,111 •«* u:i!KilRB TaiMlHBIPWI
Estimating Backgro
&..«• • • • « 1 l^k* "", ?! P '• »* ™*«S«ti31
Mississippi River Basin
CK Walker1
• i i f 'IN IT Til
'National Water Poffc^ Arta/yst
. . _ ' "i ,"' K ;•!'! ^.'•UlliS iii
Service (NRCS) ' .V;i';-t•,!'$!
7emp/e, Texas.76502 i^Mii
' *f"
»W>'W > C ' ' > '
,.„ *-} *
L
sTats?
Abstract
Estimating Background Nutrient
Loads in the Mississippi River Basin
Using Loading Functions and a G/S
The primary source of measured data
available for sediment and nutrient
loading on a national scale is found in the
stream data published by the USGS (USGS,
1991). However, this data does not provide
much information about whether the loadings
come from point, nonpoint, or background
sources or about where the sources are located.
Hydrologic modeling can be used to provide
much additional information about probable
sources of pollutants and how they are
transported through surface water and ground
water systems.
USDA/NRCS and TABS have initiated a
national scale modeling project to estimate the
nonpoint source pollutant loadings (sediment
and organic nutrients) from agricultural areas
and other land uses in the 48 coterminous
United States. We also hope to incorporate
existing data on point source loadings in our
national modeling effort. Some preliminary
results for the Mississippi River Basin from this
study will be presented in this paper.
The primary modeling unit areas for this project
are the USGS' Hydrological Cataloging Units (8-digit
watersheds). A GIS (GRASS) is being used with
national and regional scale topographic, soils,
and land use databases. Interface programs have
been developed to link these GIS databases with
soil and water assessment models. One model
used to estimate sheet and rill erosion rates is
the Universal Soil Loss Equation (USLE). In
this preliminary, generalized approach, the
amount of sediment reaching the water bodies
has been estimated by applying a drainage area
dependent delivery ratio to the USLE based
erosion estimates. Then the amount of organic
nutrients was estimated using a sediment enrich-
ment ratio for each nutrient (N and P) based on
the organic carbon content of the eroded soils.
This simple generalized procedure provides pre-
liminary estimates of background loadings of
sediments, organic nitrogen, and organic phos-
phorous, not accounting for point source load-
ings or contributions from the uses of fertilizers
in the river basin.
Introduction /> ,, >
The Resources Conservation Act of 1977
(RCA) authorizes the Department of Agricul-
ture (USDA) to appraise the current condition
and trends in the uses and conservation of soil,
water, and related natural resources on non-
146
-------
federal lands in the nation each decade. The
Third RCA Appraisal is due in 1997. The RCA
Appraisals are supposed to provide information
to be used in developing updates to the USDA
National Conservation Program (NCP). The
NCP is a public statement of policy of which
activities will have the highest priorities for
USDA agencies for natural resources conserva-
tion activities on non-federal areas.
The Natural Resources Conservation Service
(NRCS), the Agricultural Research Service
(ARS), the Texas Agricultural Experiment
Station (TAES), and other agencies are cooper-
ating on a Project for Hydrologic Unit Model-
ing of the United States (The HUMUS Project).
This Project is designed to develop a weather-
driven model of soil-plant-water interactions
and to route water flow, erosion, sediment flow,
nitrate flow, phosphate flow, and salt transport
through the major river basins of the 48 conter-
minous States for the RCA Appraisal. We
started the HUMUS Project in 1992 and will
complete it in 1996 or early 1997. We expect,
however, that the technology we are using will
still be in its infancy. We expect to improve this
technology over time and for it to be used by
ourselves and others at a wide range of scales
from small watersheds to international river
basins.
The hydrological model we are using is the Soil
and Water Assessment Tool (SWAT) developed
by the Grasslands, Soil and Water Research
Station of ARS at Temple, Texas. This is a com-
prehensive but somewhat generalized model of
surface water runoff, groundwater return flows,
and streamflow dynamics that integrates
estimates for small subwatersheds into estimates
of flows in major river basins. The SWAT can
operate with historical weather data or with a
series of synthetically generated weather
patterns. The model includes options for
simulating ponds, major reservoirs, and wetland
areas in the system. The SWAT is scale inde-
pendent, but the accuracy of the derived flow
estimates depends directly on the accuracy of
the available data, including data on weather,
topography, channel dimensions, reservoir
dimensions, reservoir operating rules, soils, land
cover, crops, on transpiration rates from crops
and natural vegetation through the dormant and
growing seasons, etc.
We are using the Geographic Resources Analy-
sis Support System (GRASS) geographic infor-
mation system (GIS) as our primary tool to
manage and manipulate the databases we have
assembled from USGS, the Weather Service,
NRCS, and other sources, on weather, topog-
raphy, land cover, soils, crops, stream locations,
watershed boundaries, political boundaries,
water quality, etc.
Dr. Srinivasan has written an interface program
to use GRASS and the databases to develop
input data sets for the SWAT model. He is also
developing a GRASS-based, report-writing
interface program, to help analyze and display
the results of the modeling in both graphical
and tabular formats. We also use the INFOR-
MIX relational data base system for filing and
querying some of our databases, especially those
having to do with soil properties and with agri-
cultural production and practices. Other GIS
from available commercial sources may also be
linked to the SWAT.
Early in the development of the HUMUS
Project, we experimented with a short-cut
approach for estimating background transport
of sediments, organic nitrogen, and organic
phosphorous. This preliminary work was based
on using an earlier version of the SWAT instead
of the more comprehensive water and sediment
routing subroutines included in the SWAT. This
approach uses our basic databases on topog-
raphy, soils, crops, and land cover, but shortcuts
147
-------
the flow routing algorithms and uses generated,
not actual, weather data. Other simplifying
assumptions used were that all trees are ever-
green trees and all crops were minimum tilled.
Figure 82 shows the resulting estimate of
average annual rates of sheet and rill erosion
caused by water runoff. The units on this map
are metric, but 7.5 to 12 metric tonnes per
hectare corresponds roughly to 3 to 5 English
tons per acre, the normal range for so-called
cropland erosion tolerance or "T" levels estab-
lished by NRCS. Figure 83 is the same map
with only four colors. Green represents areas
where sheet and rill erosion is less than
3 tons/acre/ year. Blue represents areas where
erosion rates are 3 to 5 tons per acre per year.
And red areas show where sheet and rill erosion
rates can't be kept to less than 5 tons per year
even with minimum tillage on cropland.Table 8
lists the average sheet and rill erosion rates
derived with our short-cut method for the
6 major river basins in the Mississippi River
Basin. These data suggest that over 1 billion
tons of soil are detached from the land surface
of the Basin by sheet and rill erosion in an aver-
age year. These data also reveal that while the
Tennessee River Region, which has only 11 per-
cent of its area in cropland, has the highest rates
of sheet and rill erosion. Nevertheless, this rela-
tively small Region has the least total tonnage
eroded. Conversely, the Missouri Region,
though having the lowest average erosion rate,
and about 30 percent of its area in cropland,
ranks second in total tonnage eroded. The
Upper Mississippi Region ranks third in total
area and first in percent of its area in cropland,
but fourth in its unit area average rate of erosion.
MM Untvetsity System- Fig t
Potential Averafle AtintlaJ Sheet & Bill Erosion Hates
(RKUSCP) by Hydrotogic Cataloging Units
(where C values correspond to Minimum TiBage on Croplands)
SD-3Qt/ha
30-53 Mm
Preliminary Draft
for Review
The HUMUS Project
148
-------
Preliminary Draft
forftevtew
Q no data
Q i-SVha
12-53 Vhfl
figure 83.
Resulting estimate of average annual rates of sheet and rill erosion caused by water runoff.
t
fceg.
5
6
7
8
10
II
TableS.
Preliminary estimates of sheet and rill erosion.
-""" AVrV-
Hydrotogjc Regions
Ohio River
Tennessee River
Upper Mississippi River
Lower Mississippi River
Missouri River
Arkansas-White-Red Rivers
Totals
Average Annual Erosion Rates
Tonsffear
239,932,000
63,699,000
229,509,000
137,991,000
230,440,000
126,719,000
1,028,290,000
Tons/Acre
2.31
2.44
1.9
2.13
0.71
0.8
1.29
Tons/% W!te
1477
1559
1213
1362
453
513
823
149
-------
Table 9 and Figures 84 and 85 present theses data in terms of the relative proportion each of the regions
contributes to the total Mississippi River Basin load.
Table 9.
Reg.
5
6
7
8
10
11
Hydrologic Regions
Ohio River
Tennessee River
Upper Mississippi River
Lower Mississippi River
Missouri River
Arkansas-White-Red Rivers
Mississippi River Basin
Drainage Areas
Square Miles
162,439
40,864
189,189
101,283
509,172
247,195
1,250,142
% of Area "
12.99%
3.27%
15.13%
8.10%
40.73%
19.77%
100.00%
S&R Eros.
,*•*/ y f
% Erosion
23.33%
6.19%
22.32%
1 3.42%
22.41%
12.32%
100.00%
11 (19.77%)
10(40.73%)
5(12.99%)
(3.27%)
8 (8.10%)
11 (12.32%)
7(15.13%) 10(22.41%)
8(13.42%)
5 (23.33%)
6(6.19%)
7 (22.32%)
Figure 84.
Charts depicting hydrologic region areas in the Mississippi River Basin (left) and
sheet and rill erosion rates by hydrologic regions (right).
150
-------
HComparing to the NRI
The NRCS has another method for estimating
erosion rates, one that has been used since the
late 1950's. It is now called the Natural
Resources Inventory (NRI). The NRI is a
nationwide statistical site sampling procedure
designed to make general estimates of erosion
rates on non-Federal agricultural lands. The
NRI has the advantage of having a detailed set
of data on crop and land management practices
at a large number of sample sites. Its major
disadvantage is that it does not provide com-
plete information about erosion rates on Federal
lands or forest lands. This leaves fairly large
areas unevaluated, as shown in Table lOa.
Nevertheless, the NRI reports include basin-
wide estimates of tons of erosion based on the
samples taken in the inventoried areas. Table
lOb and Figure 86 show how our preliminary
estimates of regional sheet and rill erosion rates
compare to the rates reported in the 1992 NRI.
50%
| 40%
% 30%
1 20%
o 10%
Q_
0%
% of Area
% Erosion
6 7 8
Hydrologic Regions
10 11
Figure 85.
Relative rates of sheet and rill erosion by hydrologic regions.
Table lOa.
Areas reported for erosion in the NRI.
Reg.
5
6
7
8
10
II
*-, ; Hydrologic Regions ^
/ ' ••• t , / '*
\ ' ' ^ x
Ohio River
Tennessee River
Upper Mississippi River
Lower Mississippi River
Missouri River
Arkansas-White-Red Rivers
Totals
, Total Acres #
103,960,758
26,153,159
121,081,008
64,821,123
325,869,927
158,204,925 .
800,090,900
NRtReport Acres
49,927,300
8,437,800
84,967,500
31,058,800
259, 1 50,700
117,138,400
550,680,500
%
Ref^orted,
48%
32%
70%
48%
80%
74%
69%
151
-------
Table I Ob.
Comparing sheet and rill erosion estimates.
Reg.
5
6
7
8
10
11
Hydrologic Regions
Ohio River
Tennessee River
Upper Mississippi River
Lower Mississippi River
Missouri River
Arkansas-White-Red Rivers
Totals
;: HUMUS
Tons/Year
239,932,000
63,699,000
229,509,000
137,991,000
230,440,000
126,719,000
1 ,028,290,000
NR[ I992x
Tons/Year '
254,904,360
26,732,160
283,742,740
110,043,360
491,417,420
176,426,730
1,343,266,770
The NRI erosion estimates are significantly
lower than the HUMUS estimates in the Ten-
nessee and Lower Mississippi Regions. This
could be attributed to the large portions of these
areas under forest cover and thus unreported by
the NRL In the other three regions, especially in
the Missouri Region, the NRI estimates are
significantly higher than the preliminary
HUMUS estimates. The HUMUS estimates are
based on having all croplands in minimum till
and the NRI estimates are based on actual til-
lage conditions in 1992. The differences in the
estimates might imply a policy regarding promo-
tion of minimum tillage. If these data are reas-
onably close to reality, the higher priorities for
promoting minimum tillage to reduce erosion
might better be focused in the Missouri and
Arkansas-White-Regions than in the Corn Belt.
ca
500,000,000
400,000,000
300,000,000
co 200,000,000
100,000,000
0
5 6 7 8 10 11
Hydrologic Regions
Figure 86.
Comparing S&R eras/on rotes by hydrologic region.
HUMUS
NR11992
152
-------
Sedimentation ^ -, \
In the SWAT, there is an algorithm for estimating
edge-of-watershed runoff of sediment. This load
of sediment is called "Wash Load" in this paper,
though this is a plagiarized term here. The
coefficients of sediment delivery are based on the
available literature, the gist of which is that the
delivery ratio is an inverse function of the
logarithm of the drainage areas. The delivery ratios
used in this preliminary set of HUMUS Project
sedimentation estimates are computed from the
drainage areas of each of the contributing
hydrologic cataloging unit (8-digit) areas.
When all of the sediment deliveries from these
areas are lumped to produce sediment runoff
estimates for the Hydrologic Regions, the com-
posite delivery ratio for all of the Regions is about
5 percent.
Table 1 la displays the preliminary HUMUS
Project estimates of sediment wash loads.
For comparison, the regional composite sediment
delivery ratios computed from the HUMUS
Project data were applied to the NRI estimates of
sheet and rill erosion to derive NRI-based
sediment wash load estimates. The resulting
hypothetical estimates are presented in Table lib.
Table IIa.
Estimated average annual sediment "wash" loads.
\ '*''
Reg.";
5
6
7
8
10
II
•','•'-,; "*^ "s •- - ™- '/' ""- * T*Y
.„, '*Hy4r'c|t6igic Region, ^, "':
Ohio River
Tennessee River
Upper Mississippi River
Lower Mississippi River
Missouri River
Arkansas-White-Red Rivers
Total
# '" - ^ HUM^Preliftiinary 'Estimates-"- ; ^
foris/Year,
12,185,000
3,362,000
11,760,000
7,220,000
1 1 ,483,000
6,474,000
52,484,000
T©ns/Acrest,,
0.117
0.129
0.097
O.I II
0.035
0.041
0.066
'™Tons'/Sq.'-Mife -
75
82
62
71
23
26
42
Table lib.
Imputed NRI estimates of average annual sediment "wash" loads
\\
#.
5
6
7
8
10
II
s, r **' ' r
--."•' . "* ' ' N
Hydrologic Region ' ;
Ohio River
Tennessee River
Upper Mississippi River
Lower Mississippi River
Missouri River
Arkansas-White-Red Rivers
Totals
,/ :- HUMAS Estimates f.
- JErbsfon ..
tons/Year ;
239,932,000
63,699,000
229,509,000
1 37,99 1 ,000
230,440,000
126,719,000
1 ,028,290,000
Sediment
Tons/Year
12,185,000
3,362,000
11,760,000
7,220,000
11,483,000
6,474,000
52,484,000
'•w.
Deiiv/
Ratio
0.051
0.053
0.051
0.052
0.050
0.05 1
0.051
, * ' x -rNRI" *-
Es,t. Erosion
Tons/Year
254,904,360
26,732,160
283,742,740
1 1 0,043,360
491,417,420
176,426,730
1 ,343,266,770
, Hyp. „
Se4imsnt
t
-------
Table lie and Figure 87 present a comparison
of HUMUS Project and imputed NRI-based
sediment wash load estimates with USGS data
on suspended sediments as reported for the
Hydrologic Regions in the Mississippi River
Basin in the National Water Summary of 1990/
1991 by Smith, Alexander, and Lanfear.
The USGS suspended sediment data do not
include estimates of reservoir sediment entrap-
ment or of bedload sediment transport. That
the USGS suspended sediment loads are gener-
ally higher than the HUMUS and NRI estimates
of sediment wash may be attributed to the fact
that a significant part of the suspended sediment
load is derived from instream erosion, mass
wasting, and deposition of airborne sediments
not accounted for by the estimates of sheet and
rill erosion.
On the other hand, the available literature on
sediment delivery ratios is largely based on his-
torical comparisons of estimates of erosion rates
with recorded suspended sediment data from
stream gages. A slight change in the selection of
sediment delivery ratios could have markedly
changed Tables 11 a, lib and lie and Figure 87.
In the ongoing phase of the HUMUS Project,
we intend to use a stream power function to
derive new estimates of sediment transport
processes. We hope this approach will provide
new insights into comparisons between sedi-
ment source and delivery estimates.
The reasons for differences between the NRI
and HUMUS sediment delivery estimates are
probably the same as for the differences in
estimated erosion rates. Nevertheless the close
correlations between the current condition NRI
erosion rates for the Ohio, Missouri, and
Arkansas-White-Red Regions are remarkable,
given the short-cut method used herein to
derive the NRI-based estimates. The low cor-
relations for the Tennessee and Lower Missis-
sippi Regions suggest the need for further
research. For example, we need to check with
the authors of the referenced USGS report as to
whether the USGS estimate of suspended loads
in the Lower Mississippi Region include some
sediment loading from upstream regions.
Figure 88 is a map showing the preliminary
HUMUS Project estimates of sediment wash
loads for the 8-digit watersheds in the Missis-
sippi River Basin. Notice that the sediment
delivery estimates are particularly high for the
Tennessee and Lower Mississippi Regions, areas
where the NRI reports erosion for significantly
less than half of the contributing drainage areas.
Table lie.
Est/moted average annual sediment "wash" loads
Reg
5
6
7
8
10
II
Hydrologic Region
Ohio River
Tennessee River
Upper Mississippi River
Lower Mississippi River
Missouri River
Arkansas-White-Red Rivers
Total
Humus Preliminary Estimates
HUMUS
"wash" load
t/mi2/year
75
82
62
71
23
26
26
NRI Imputed
"Wash
Load
t/mi2/year
80
35
77
57
48
36
36
' / USGS;"''
Suspended
t/mi2/year
85
85
102
III
45
31
31
154
-------
and Phosphorous
The preliminary HUMUS Project estimates for
movements of nitrogen and phosphorous are
based entirely on computations of contributions
of organic matter contained in and transported
with topsoils detached by erosion. They include
no estimates of contributions by fertilizers, ani-
mal wastes, urban wastes, industrial discharges,
atmospheric exchanges, leaf fall, or any other
sources. They are also based on the assumption
that all cropsoils are minimum tilled. Thus, they
are underestimates of background loads. They
represent low levels of nutrient contributions
that society probably cannot hope to achieve by
any conceivably adoptable set of pollution
reduction policies.
Table 12 provides an insight into the implica-
tions of our assumptions.
The reason that the percentages for Organic N
and Organic P are so similar is that the data are
in percentages of total contribution, not actual
loads. Though Phosphorous delivery tonnages
are much lower than are Nitrogen loads, the
percentages of total loads by river basins are so
nearly identical that they show as only one line
on Figure 89.
Table 12.
Comparing sediment load rates with organic N&P rotes.
'/
. No.V
5
6
7
8
10
II
s ^ £
- ' / ' ^
' N ^
' '/
^ •<• f-tydrologic Region
Ohio River
Tennessee River
Upper Mississippi River
Lower Mississippi River
Missouri River
Arkansas- White-Red Rivers
Mississippi River Basin
; -, Percent Total
^Sed VYash %
23.22
6.41
22.41
13:76
21.88
12.34
100.00
Org. N'l
21.14
4.64
30.69
3.11
26.41
9.01
100.00
- Pr§-p
21.17
4.73
30.63
8.11
26.35
9.01
1 00.00
These data show that soils in the Upper Mississipppi
and Missouri Regions have significantly higher levels
of organic content than do the soils in the other
regions. The soils in the humid warm Lower
Mississippi Region have the lowest levels of organic
content Thus, on a ton per ton basis, they provide
the lowest share of associated organic matter
contributions to the Gulf of Mexico.
This information is illustrated in the maps in Figure
90 for organic nitrogen and Figure 91 for organic
phosphorous.
Although deliveries of organic nitrogen and
phosphorous are not directly related to streamflow
estimates of nitrates and total phosphorous, some
insights can be inferred from correlations between
these data.
Table 13 and Figure 92 display comparisons
between HUMUS Project estimates of delivery of
organic nitrogen under low erosion rate conditions
to data on nitrate deliveries as published in the
USGS report described above.
Table 14 and Figure 93 show similar comparisons
between estimates of organic phosphorous runoff
and total phosphorous deliveries reported in the
USGS report.
155
-------
.g 35%
s
2 3O%
25%
»*»
CD
--. 2O%
•x
|2 15%
CO
I
o
o>
Q_
5%
0%
5 6 7 8 1O 11
Hydrologic Regions
Sed. Wash
•*
Org. N.
Org. P.
Figure 89.
Comparing sediment, Org. N, & Org. P by hydrologic regions.
Atmosplxric *rt disposition rate o/NADP nitrate (N03) in
1988 (top) and 1993 (bottom) in mollm2. NADP sites are
located as solid circles; Mississippi River watershed outlined by
ne.
Atmospheric wet disposition rate of NADP ammonium
(NH4) in 1988 (top) and 1993 (bottom) in mollm2.
NADP sites are located as solid circles; Mississippi River
watershed outlined by heavy line.
'130 120
110 100 90
Figure 90.
Organic nitrogen map
80
70
130 120 110 100 90 80
Figure 91.
Organic phosphorous map
70
156
-------
Table 13.
••;\.
' N<^>
5
6
7
8
10
II
'« .,> ":^ ;' Sv" v,
?. ., /*+. "-/ •!*--. '••>-' :.• , - •••,.
',-'*•• '•- ^ »;;• . "- -' ^ %i.
,'--^ :„-. -»« '•"Cs- ', ' -5.. *r ^
s ;i Hydfologjc Region t ^
Ohio River
Tennessee River
Upper Mississippi River
Lower Mississippi River
Missouri River
Arkansas-White-Red Rivers
Mississippi River Basin
:. '"' "•- Percent Total "••*- _,,-
Avg. Org!>4';
- Tdi^/year^-
23.22
6.41
22.41
13.76
21.88
12.34
100.00
v,A%;,6rg,N\
'?t;/mi2/x'ear -
21.14
4.64
30.69
8.11
26.41
9.01
100.00
^u's-gS^Esfc.*
"v:;-.Njtrates^
- t/rn{2/vear:
21.17
4.73
30.63
8.11
26.35
9.01
! 00.00
4
(0
a>
CO
a-
CO
1 1
o
0
-. __J*CX _
Ave. Org. N
Nitrates
5 6 7 8 10 11
Hydrologic Regions
Figure 92.
Comparing organic N with nitrate loads by hydrologic regions.
157
-------
Table 14.
No.
5
6
7
8
10
II
Hydrologic Region
Ohio River
Tennessee River
Upper Mississippi River
Lower Mississippi River
Missouri River
Arkansas-White-Red Rivers
Totals
HUMAS Preliminary
Avg. Org. N
Tons/year
94,000
21,000
136,000
36,000
117,000
40,000
444,000
Av. Org. N
T/mi2/year
0.58
0.51
0.72
0.36
0.23
0.16
0.36
USGSEst. :
TotalP
t/mi2/year
0.125
0.125
0.157
0.103
0.028
0.039
0.072
3
CD
or
CO
CO
u.o
0.7
0.6
0.5
0.4
0.3
0.2
0.1
n
A
' \
/ \
**" ~"-~. x x
~^>S \
\
\
. "X.
•s.
"~~ "-"•--><
"^* ***'**a<^^^ ""
^^^^M^^n^MH
,
x;
Organic P
Total P
5 6 7 8 10 11
Hydrologic Regions
Figure 93.
Organic P runoff vs. Total phosphorous by hydrologic region.
Even though the erosion-related sources of
nutrients are only a part of the total load of
nutrients to the streams in the Mississippi River
Basin, the preliminary estimates of runoff loads
of organic N and P are significantly higher than
are the USGS streamflow-based records of
nitrates and total phosphorous.
This is not at all surprising. This is an indication
of the vital importance of instream deposition,
reduction, and volatilization of nutrients and of
the effects of living organisms in aquatic
ecosystems.
^P re sen tation Discussion/
Clive Walker (NRCS—Texas A&M University)
No questions after Clive Walker's presentation.
158
-------
Abstract , ' '
Daily water samples have been collected
since February 1992 immediately
below the pool 19 dam at Hamilton,
Illinois. The frozen samples were analyzed for
orthophosphate, dissolved silicon, nitrate plus
nitrite, nitrite and ammonium using standard
methods (Whitledge et al, 1986) on a Technicon
AutoAnalyzer II segmented flow analyzer.
Water samples were filtered for chlorophyll a
determinations starting in February 1993 to
provide estimates of phytoplankton biomass
and "potential phytoplankton production."
A strong annual cycle of dissolved nitrogen was
observed over the three year period ranging
from near depletion in the summer to greater
than 370 umole/liter during the winter. The
rapid changes in ambient concentrations of
dissolved inorganic nitrogen over short time
periods Implies that local events can induce
relatively large changes.
No Manuscript Submitted.
Mr. Whitledge's presentation was canceled.
159
-------
mm A • .
Estimates
Scott P. Dinnel
Center for Marine Science
University of Southern Mississippi,
Abstract
Annual values of atmospheric deposition
of nitrogen to the Mississippi River
System drainage basin were computed
for 15 years (1979-1993) using National
Atmospheric Deposition Program wet
deposition data and 3 years (1990- 1992) of
National Dry Deposition Network dry
deposition data. Wet deposition of nitrogen was
measured as nitrate (NO3) and ammonium
(NH4), dry deposition was determined as nitrate,
the sum of gaseous nitric acid (HNO3) and
participate nitrate. Fifteen year average wet
deposition of nitrate and ammonium were
44 and 42 X 109 mol/yr, respectively. Three
year average dry deposition of nitrate was
33 X 09 mol/yr, approximately 75 percent the
like year wet deposition of nitrate. Total annual
nitrogen deposition was estimated using NADP
data and literature factors for nitrite and organic
nitrogen. Average atmospheric deposition of
total nitrogen was estimated as 200 x 109 mol/yr.
Annual atmospheric deposition of total nitrogen
was compared to total Mississippi River nitro-
gen for the same time period. U.S. Geological
Survey water quality data and U.S. Corps of
Engineer water discharge data from the Missis-
sippi and Atchafalaya Rivers were used to esti-
mate the annual riverine flux of nitrogen. Aver-
age annual riverine nitrogen flux was deter-
mined to be 115 x 109 mol/yr. Average atmos-
pheric deposition of total nitrogen accounts for
approximately 174 percent the average total
riverine nitrogen flux.
Introduction
The Mississippi River has a vast watershed;
including the upper Mississippi, the Missouri
and the Ohio Rivers, it drains 41 percent of the
continental U.S. (Figure 94). The Mississippi
River also transports high nutrient loadings from
the watershed to the northern Gulf of Mexico.
These nutrient loads from the Mississippi River
have changed over the last four decades. There
has been a doubling in nitrate, from a low in the
mid-1960's to a high in the mid-1980's (Dinnel
and Bratkovich, 1993), creating an enrichment
of the coastal waters of the Gulf of Mexico, and
contributing to the summer depletion of dis-
solved oxygen in the bottom waters (Turner
et al., 1987). In order to manage the riverine
nitrogen, an understanding of the nitrogen types
and sources must be made. One useful method
would be to account for all the nitrogen in the
Mississippi River watershed. By quantifying the
various inputs and outputs of nitrogen, a budget
or mass balance, and the relative importance of
each nitrogen source can be made. The input
and output sources of nitrogen, especially in the
vast and diverse watershed, is complex. Using
the budget terms following Jaworski et al.
(1992), one can appreciate the task. Input terms
160
-------
for a watershed nitrogen budget would be, but
not limited to, waste water effluent, animal
waste, soil fertilizers, atmospheric deposition,
biological fixation and adsorption, importation
as commodities and also in ground water. These
inputs would have to balance the outputs and
whatever storage that would take place. These
outputs would be crop harvest, river discharge,
volatilization, export as commodities and into
the ground water, and denitrification. Storage of
nitrogen would take place in the soil, ground
water and in the biomass.
A budget such as this has been accomplished
for numerous smaller watersheds, but not for
one of the scope of the Mississippi River water-
shed. Some of these nitrogen budget terms are
readably quantifiable. One is the nitrogen con-
tained in the river discharge, an assumed major
output term, it is of primary importance to the
problem of the nitrogen enrichment of the
northern Gulf of Mexico coastal waters. Using
sampled riverine nitrogen concentrations and
water discharge, one can estimate the annual
flux from the watershed.
Of the suggested inputs, soil fertilizer has been
cited as the leading source of nitrogen in the
river discharge (Turner and Rabalais, 1991). It is
conceivable that other input terms are quanti-
fiable, and could also contribute to the nitrogen
river discharge. Specifically the waste water
effluent, the animal waste and the atmospheric
deposition terms. It is important to determine if
these terms, if any, are of sufficient magnitude
as to contribute to the budget. The remaining
terms are arguably more difficult to estimate. A
direct comparison of any input term to a major
output term would be one technique, albeit
simple, in determining the relative importance
of that input.
The atmospheric deposition term is one that is
quantifiable, but for which no quantity has been
made. Using sampled precipitation volumes and
nitrogen concentrations of the precipitation,
combined with estimates of non-precipitation
atmospheric deposition, the annual atmospheric
deposition term can be determined and com-
pared to the magnitude of the river discharge
term.
JfethocU. f /L", ,__,"/_
Annual Mississippi River nitrogen flux was
determined as a combination of the flux down
the Atchafalaya River, the major distributary of
the Mississippi River, and the Mississippi River
proper. Riverine nitrogen flux included total
nitrate, nitrite, ammonium and organic fluxes.
Riverine nitrogen flux was determined from
U.S. Department of Interior (1978-1993)
National Stream Quality Accounting Network
(NASQAN) concentrations at St. Francisville,
Louisiana and Melville, Louisiana and U.S.
Army (1978- 1993) Corps of Engineers
(USACOE) water discharge from Tarbert
Landing, Mississippi and Simmesport,
Louisiana, for the Mississippi and Atchafalaya
Rivers, respectively (Figure 94). The lower
Mississippi River distributes a portion of the
water discharge down the Atchafalaya River.
Since 1978 this annual portion has been con-
trolled by the USACOE at 30 percent of the
total discharge.
The total atmospheric deposition of nitrogen to
the Mississippi River watershed was determined
as the sum of wet and dry deposition of the
various forms of nitrogen. These were inorganic
forms such as nitrate, nitrite, and ammonium,
and organic nitrogen. Wet deposition was by
way of any form of precipitation, and dry depo-
sition was by way of gaseous and particle depo-
sition. In order to estimate annual total deposi-
tion of nitrogen various data and relationships
were used.
161
-------
Wet deposition of nitrate and ammonium have
been sampled since 1979 by the National
Atmospheric Deposition Program (NADP,
1995), sponsored by the U.S. Department of
Agriculture and the U.S. Geological Survey.
These are weekly concentrations and precipi-
tation volumes for over 200 sites currently in
the network. Annual mass of wet deposition of
nitrogen as nitrate and ammonium were deter-
mined by summing weekly products of preci-
pitation volume and concentrations over each
year from 1979 through 1993, for each NADP
site. These annual values were contoured using
PLOT88, a software library of PLOTWORKS,
Inc., on a 10 X 10 grid. The gridded data was
then summed over the Mississippi River water-
shed to get the annual mass deposited by each
nitrogen form. Annual wet deposition of nitrite
was not measured and so was estimated as
3 percent the nitrate deposition using conserva-
tive literature relationships (Meybeck, 1982).
Information on the dry deposition of nitrogen
was limited. Although the National Dry Deposi-
tion Network (NDDN) sponsored by the
Environmental Protection Agency (EPA) was
begun in 1986, and combined into the Clean Air
Status and Trends Network (CASTNet) in 1990,
the data coverage was sparse compared to the
NADP data. These sites did not represent the
Mississippi River watershed adequately, being
predominantly located in the eastern U.S. Three
years, 1990-1992, of annual NDDN dry nitrate
deposition (ESE, 1995), was determined using a
similar procedure as in the determination of the
annual wet nitrate deposition. These watershed
annual dry nitrate deposition estimates were
compared to the same three years of wet nitrate
deposition values. The NDDN determined dry
nitrate and ammonium deposition by particulate
counts of nitrate and ammonium and by
sampled nitric acid gaseous concentrations
combined with modeled deposition velocities.
Mississippi River watershed dry nitrate
deposition was estimated as 0.75 the wet nitrate
deposition, an average of 1990—1992
comparisons. This was somewhat near the
middle of the various dry to wet nitrate relation-
ships reported in the literature (Table 15). Dry
to wet ammonium relationships also were quite
varied and poorly represent the Mississippi
River watershed (Table 15). Dry ammonium
was estimated as 0.25 that of wet ammonium
using the average of the literature values in
Table 15. Dry nitrite deposition was estimated
as equal to the wet nitrite deposition. This
assumption was likely to underestimate dry
nitrite deposition when compared to data from
the northeast U.S. where dry nitrite deposition
was high (Barrie and Sirois, 1986). The total
inorganic atmospheric deposition of nitrogen
was estimated as the sum of measured wet
nitrate and ammonium values that were
increased to include estimates of wet and dry
nitrite, and dry nitrate and ammonium. Wet and
dry organic deposition was combined and esti-
mated as a 1:2 organic to inorganic ratio
(Hendry et al, 1981; Correll and Ford, 1982,
Meybeck, 1983; Jaworski et al., 1992).
Using the stated relationships between wet and
dry nitrogen forms and among different forms
one could compute a total atmospheric deposi-
tion on nitrogen for the Mississippi River water-
shed. The relationships were all based upon the
measured wet nitrate and ammonium deposi-
tions. Given the annual NADP wet depositions
of nitrate and ammonium, the total wet deposi-
tion of inorganic nitrogen was the moles of
ammonium plus 1.03 nitrate (0.03 as wet
nitrite). Dry deposition of inorganic nitrogen
was determined as 0.78 the wet nitrate (0.75 as
dry nitrate, 0.03 as dry nitrite) plus 0.25 the wet
ammonium. The total inorganic nitrogen was
then 1.81 the wet nitrate plus 1.25 the wet
ammonium. To include in an estimate of
162
-------
organic deposition a factor of 1.5 the inorganic
nitrogen was used.
Results ' /; V,
The average annual total riverine nitrogen flux
was 115 x 109 mol (Table 16). Total annual
riverine nitrogen flux varied from <70 x 109
mol to > 150 X 109 mol during the study period
(Figure 95). The average major components to
riverine nitrogen flux were nitrate (59 percent)
and organic nitrogen (37 percent); ammonium
(3 percent) and nitrite (1 percent) were minor
components. Annual nitrate and organic nitro-
gen fluxes were fairly well correlated, with
higher nitrate to organic nitrogen ratios in low
discharge years.
Using a 29.3 percent average runoff or precipi-
tation retention factor for the continental U.S.
(U.S. Department of the Interior, 1984), the
gauged annual Mississippi River discharge was
of the same magnitude and varied in a reason-
able fashion as the total annual runoff from the
watershed precipitation. This supported the
computational technique used here to sum
parameters over the watershed, and thus in
determining the total annual nitrogen deposition
to the watershed as wet deposition. Spatial
distribution of annual precipitation varied from
a southeastern U.S. high to a Rocky Mountain
low, which is in good agreement with 30 year
means of the National Weather Service.
Average annual NADP atmospheric wet depos-
itions of nitrate and ammonium were almost
equal, with 44 and 42 x 109 mol, respectively
(Table 16). Interannual variation was small,
values were within ±10 x 109 mol of the means
(Figure 96).
The annual distribution of NADP atmospheric
deposition rates of nitrate (Figure 97) and
ammonium (Figure 98) for 1988, a low preci-
pitation year and 1993, a high precipitation year
show a similar pattern. Highest wet deposition
rates of nitrate were centered around the
southern Great Lakes and extended into New
England; deposition rates were lower towards
the western portion of the watershed. Although
nitrate wet deposition magnitude was consis-
tent, during years of higher precipitation, the
total deposition was greater and the region of
highest deposition extended farther into the
Midwest. Highest wet deposition rates of
ammonium were also centered near the south-
ern Great Lakes and decreased towards both
coasts. During years of high precipitation
ammonium magnitudes substantially increased
and high deposition regions were centered in
the middle of the watershed.
The NDDN dry nitrate deposition had lower
annual values, but with a similar spatial distribu-
tion as the NADP wet deposition (Figure 99).
The NDDN dry deposition was determined
from fewer station locations and for only three
years of data. The sparse spatial coverage,
relative to the NADP coverage, was likely to
contribute to some uncertainty in annual
NDDN deposition distribution, but the magni-
tudes of the three years of NDDN dry nitrate
deposition were considered reasonable. The
three computed annual values of NDDN dry
nitrate deposition were approximately 75 per-
cent the corresponding annual values of NADP
wet nitrate deposition. The limited number of
years of NDDN coverage forced the use of this
factor to relate annual dry nitrate deposition to
the longer series of annual NADP wet nitrate
deposition.
The average annual total atmospheric deposi-
tion, the, sum of all measured and estimated
components, was 200 x 109 mol (Table 16).
This was 174 percent the average annual total
riverine nitrogen flux.
_ 163
-------
Total atmospheric deposition does not exhibit
the same interannual variation as the total
riverine flux (Figure 100). Although both
values did decrease during years of low precipi-
tation, the magnitude of the atmospheric depo-
sition of nitrogen was less variable.
i" , , , • ' --'''"+,
EConcIusions ^ _' ' , ' ' ._ .„
It is clear that the annual total atmospheric
deposition of nitrogen to the Mississippi River
watershed is of the same order of magnitude, if
not larger, as the annual total riverine flux of
nitrogen. In a watershed nitrogen budget, one
of the previously unquantified input terms, the
atmospheric deposition, is found to be of
comparable magnitude to one of the presumed
major output terms, the riverine nutrient flux. It
is therefore essential that the atmospheric depo-
sition of nitrogen be included into any nitrogen
budget of the Mississippi River watershed.
Although purposely simple, this analysis does
have uncertainties in the annual nitrogen
deposition quantities. The vast spatial scale of
the watershed creates a number of accuracy
problems. The Mississippi River watershed is
composed of a number of smaller watersheds,
each having different precipitation, different
spatial deposition of nitrogen forms, and
different depositional relationships between
nitrogen forms. Wet and dry temporal vari-
ations, as well as spatial deposition variations,
were optimistically accounted for with the use
of annual quantities in this study.
There are still many points to clarify. First is the
question of quantifying the deposition quantities
of the various nitrogen forms and the modes of
deposition. In this study nitrite and organic
nitrogen, as well as dry nitrate and ammonium,
were related to the wet deposition of two
nitrogen forms, nitrate and ammonium. A
number of assumed relationships were used in
this study for lack of direct measurements.
Additional analysis of existing dry nitrate and
ammonium measurements are in order to more
accurately determine the dry to wet nitrate
relationship or to replace them altogether with
information derived from dry deposition meas-
urements. More information is necessary for
accurate determination of both wet and dry
organic nitrogen deposition. This includes both
spatial watershed and multi-year temporal
differences. Where decade length measurements
are not available, clarification of relationships
between wet to dry and between nitrogen forms
are necessary.
A major component in determining the role of
atmospherically deposited nitrogen is a reten-
tion factor. Even a spatial averaged retention
factor, such as the continental U.S. value used in
the comparison of annual precipitation volumes
to river discharge, would enhance our under-
standing of the relative importance of the
atmospheric input of nitrogen.
Certainly a more in-depth accounting is neces-
sary in both time and space. A temporal analysis
of the sub-basin watersheds using atmospheric
deposition information would improve our
understanding of the relationships between the
atmospherically derived nitrogen and the trans-
port by the rivers.
Remembering that the ultimate goal is the com-
prehensive nitrogen budget of the Mississippi
River watershed, only with this level of under-
standing can reasonable management plans be
created to address the problem of the excess
riverine flux of nutrients to the northern Gulf
of Mexico.
164
-------
f & c kn o vne1I e cf g m e fits
This work was the resulrof research sponsored
by the NOAA Nutrient Enhanced Coastal
Ocean Productivity program, Department of
Commerce, under grant [NA90AA-D-SG688
(R/ER-20)], the University of Southern Missis-
sippi and The Mississippi-Alabama Sea Grant
Consortium.
-. ,
References
Baker, L.A., 1991. Regional estimates of
atmospheric dry deposition. In: Acidic
deposition and aquatic esosystems. Charles,
D.F. and S. Christie (eds.), Springer-Verlag,
New York. p. 645-652.
Barrie, L.A. and A. Sirois, 1986. Wet and dry
deposition of sulfates and nitrates in eastern
Canada, 1979-1982. Water, Air, and Soil
Pollution, 30:303-310.
Correll, D.L. and D. Ford, 1982. Comparison of
precipitation and land runoff sources of
estuarine nitrogen. Estuarine, Coastal and
Shelf Science, 15:45-56.
Dinnel, S.P. and A. Bratkovich, 1993. Water
discharge, nitrate concentration and nitrate
flux in the lower Mississippi River. Journal
of Marine Systems, 4:315-326.
Environmental Science & Engineering, Inc.,
1995. CASTNet National dry Deposition
Network 1990-1992 status report.
Gainesville, FL. EPA-600/R-95/086.
Hendry, C.D., P.L. Brezonik, and E.S. Edger-
ton, 1981. Atmospheric deposition of
nitrogen and phosphorus in Florida,
Chapter 11, p. 199-215. In: Eisenreich, S.J.,
(ed.) Atmospheric Pollutants in Natural
Waters. Ann Arbor Press, Ann Arbor, MI.
p512.
Jaworski, N.A., P.M., Groffman, A.A Keller,
and J.C. Prager, 1992. A watershed nitrogen
and phosphorus balance: the upper
Potomac River Basin. Estuaries 15:83-95
Meybeck, M., 1982. Carbon, nitrogen, and
phosphorus transport by world rivers.
American Journal of Science 282:401-450.
Meybeck, M., 1983. Atmospheric inputs and
river transport of dissolved substances. In:
Dissolved loads of rivers and surface water
quantity/quality relationships. Proceedings
of the hamburg Symposium, August 1983.
lAHSPubl. 141.pl73-192.
Shepard, J.P., M.J. Mitchell, T.L. Scott, Y.M.
Zhang, and D.J. Raynall, 1989. Measure-
ments of wet and dry deposition in a north-
ern hardwood forest. Water, Air, and Soil
Pollution, 48:225-238.
National Atmospheric Deposition Program
(NRSP-3)/National Trends Network
(NADP/NTN), 1995. NADP/NTN
Coordination Office, Natural Resource
Ecology Laboratory, Colorado State
University, Fort Collins, CO 80523.
Turner, R.E., R. Kaswadji, N.N. Rabalais, and
D.F. Boesch, 1987. Long-term changes in
the Mississippi River water quality and its
relationship to hypoxic continental shelf
waters. In: Estuarine and coastal manage-
ment—Tools of the trade, Proceedings of
the 10th National Conference of the Coastal
Society, 12-15 October, 1986, New Orleans,
LA. .
165
-------
Turner, R.E. and N.N Rabahis, 1991. Changes
in Mississippi River water quality this
century. Bioscience, 41 (3): 140-147.
U.S. Army, 1979-1993. Stages and Discharges
of the Mississippi River and tributaries.
Corps of Engineers, New Orleans District.
New Orleans, LA.
U.S. Department of the Interior, 1979-1993.
Surface water Quality Reports, U.S.
Geological Survey. Washington D.C.
U.S. Department of the Interior, 1984. National
Water Summary, 1983: Hydraulic Events
and Issues. U.S. Geological Survey, Water
Supply Paper 2250. Washington D.C.
Young, J.R., E.G. Ellis, and G.M. Hidy, 1988.
Deposition of air-borne acidifiers in the
western environment. Journal of
Environmental Quality, 17(l):l-26.
Table IS.
Dry to wet deposition relationships for nitrate (NO3) and ammonium (NH4).
DRY:WET
LOCATION
REFERENCE
NO3
0.25 eastern Canada
0.3 midwest U.S.
0.4 northeast U.S.
0.96 Florida
1 no. New York
1 western U.S.
1.5 so. Blue Ridge
1.6 Tennessee
4 Florida
Barrie and Sirois (1986)
Baker (1991)
Baker (1991)
Baker (1991)
Shepard et al. (1989)
Young et al. (1988)
Baker (1991)
Shepard et al. (1989)
Hendry et al. (1981)
NH4
0.14 no. New York
0.19 midwest U.S.
0.25 Tennessee
0.34 northeast U.S.
0.39 Florida
0.38 so. Blue Ridge
0.5 Florida
Shepard et al. (1989)
Baker (1991)
Shepard et al. (1989)
Baker (1991)
Baker (1991)
Baker (1991)
Hendry et al. (1981)
166
-------
Table 16.
Average annual Mississippi River nitrogen flux, as riverine total, nitrate (N03), organic nitrogen OR,
ammonium (NH4), and nitrite (NO 2); total average atmospheric deposition of nitrogen to the Mississippi River
watershed, and as nitrate (NO^ and ammonium (NH4); and the
equivalent deposition rates for the Mississippi River watershed, 1979-1993.
NITROGEN
FLUX
TOTAL RIVER
NO3
NOR
NH4
N02
TOTAL
ATMOSPHERIC
NO3
NH4
MEAN
(xlO9 rool)
115
68
42
3.2
1.6
200
44
42
DEPOSITION
RATE (mol/m2)
0.036
0.021
0.013
0.001
0.0005
0.062
0.014
0.013
167
-------
UNITED
STATES
Mississippi
River
Drainage
Basin
Mississippi,
~River
ATLANTIC
OCEAN
PACIFIC
OCEAN
DETAIL MAP
GULF OF MEXICO
Simmesport
Melville
LOUISIANA
St. Francisville
o 10 20 so 40
Baton Rouge
GULF OF MEXICO
Figure 94.
Genera/ limit of the Mississippi River watershed (top). Lower Mississippi River with Atchafalaya
River, USACOE discharge gauging sites at Tarbert Landing, MS and Simmesport LA, and USGS
NASQAN sampling sites at St Francisville, LA and Melville, LA (bottom). Redrawn from Dinnel
andBratkovich(!993).
168
-------
o
E
CO
o
LJ
O
O
o:
200
150 -
100 -
IT? 7 f TTi ' T~ i
0
1980
1984
1988
1992
YEAR
Figure 95.
Annual total Mississippi River nitrogen flux, with equivalent average deposition
rate to Mississippi River watershed, 1979-1993. Total nitrogen (TN solid
square), nitrate (NO3 solid circle), organic nitrogen (Nor open circle),
ammonium (NH4 open square), and nitrite (solid triangle).
100
o
E
en
O
o
o
80 -
60 -
40 -
20 -
0
- 0.02
E
\
o
UJ
a:
- 0.01 o
0.00
o
QL
Id
1980 1984 1988
YEAR
1992
Figure 96.
Annual atmospheric wet deposition ofNADP nitrate (NO3 solid circle) and ammonium (NH4
open square), with equivalent average deposition rate to Mississippi River watershed,
1979-1993.
169
-------
130 120
130 120
70
Figure 97.
Atmospheric wet deposition rate ofNADP nitrate (N03) in 1988 (top)
and 1993 (bottom) in mollml. NADP sites are located as solid circles;
Mississippi River watershed outlined by heavy line.
170
-------
130 120
110
100
80
70
Figure 98.
Atmospheric wet deposition rate ofNADP ammonium (NH4) in 1988 (top) and 1993
(bottom) in mol/m2. NADP sites are located as solid circles;
Mississippi R/ver watershed outlined by heavy line.
171
-------
130
120
110
90
80
70
100
Figure 99.
Atmospheric dry deposition rate ofNDDN nitrate (NO3) in 1991 in mol/m2. NDDN sites are
located as solid circles; Mississippi River watershed outlined by heavy line.
o
E
en
o
z:
UJ
o
o
300 -
200 -
100 -
0
T - 1 - 1 - 1 - 1 - 1 - 1 - 1 - 1 - 1 - 1 - 1 I I t
1980 1984 1988 1992
YEAR
0.10°!
•x.
o
LJ
- 0.05
in
o
Q_
LJ
0.00°
Figure 100.
Annual atmospheric deposition of total nitrogen (NA solid triangle), annual atmospheric wet
deposition of nitrogen as nitrate plus ammonium (NW open triangle) and annual total
Mississippi River nitrogen flux (NR solid circle), with equivalent average deposition rates to
Mississippi River watershed, 1979-1 993.
172
-------
Presf ntot/on Discussion
Scott Dinnell (University of Southern
Mississippi—Center for Marine Sciences)
Don Boesch (University of Maryland-
Cambridge, MD): asked Scott Dinnell if he
could quantify the export of atmospheric
deposition from the landscape (ground-
water) because a lot of the deposition is
taking place in the northeastern part of the
Basin, which tends to have higher forest
cover than the rest of the Basin. This would
presume to be more retentive of that source.
He asked him if he has calculated some
hypothetical estimates of exports.
Scott Dinnell responded that he has not
quantified retention based on different
landscapes in sub-basins, even between the
Ohio River to the upper Missouri River, or in
the plain states where land cover and soil types
would cause some kind of variation of the
quantities atmospherically deposited versus the
amounts found in the river. He said that this
type of study was another step that could be
conducted. He would like to look at the spatial
and temporal differences, for at least the wet
deposition information.This weekly data
collected over 15 years could be used to look at
phasing between the deposition, the heavy
deposition times, and the local river signals in
the drainage basins. He felt it was important to
at least look at the major drainage basins from
that point of view. There is a relationship
among spatial and temporal distribution and the
amounts and locations of atmospheric deposi-
tion and different retention factors.
173
-------
!? ' ,'.•:. !•„ ™!;r:f? jiwiii^^ ; JiiiiiiiS
Abstract
U.S. Geological Survey (USGS) is
beginning a study to compute annual
mean concentrations and annual loads of
total nitrogen and total phosphorus, and
temporal trends in annual mean concentrations
and annual loads of total nitrogen and total
phosphorous, at 41 USGS National Stream
Quality Accounting Network (NASQAN)
gaging stations on die Mississippi River and its
major tributaries (Figure 101). The study is in
cooperation with the Nutrient Enrichment
Committee of the Gulf of Mexico Program. Of
the 41 stations, 9 are on the main stem of the
Mississippi River and 32 are on 25 major tribu-
taries of the river. Only data collected during
the period October 1967-September 1994 will
be used because, before October 1967, methods
for nitrogen analysis were different from current
(1995) methods. Annual constituent loads will
be computed by summing daily loads obtained
from equations developed by regressing log-
transformed daily loads computed from consti-
tuent concentrations on log-transformed daily
streamflow. Temporal trends in constituent
concentrations and loads will be computed
using Kendall's Tau test or shown graphically
using the smoothing technique LOWESS
(LOcally WEighted Scatterplot Smoothing).
No Manuscript Submitted.
Presentation Discussion
j V •» f
Dee Lurry (U.S. Geological Survey-
Austin, TX)
An unidentified audience member com-
mented that Dee Lurry said she intended to
correct the concentrations for flow and asked if,
both the flow and concentration are seasonal
and correlated, but are independent in some
aspect, what changes in data could occur by
doing a flow adjustment.
Dee Lurry responded that she expects to have
flow adjusted residuals. She told the audience
member that she had just begun the study, but
would like to discuss the results when she
obtains the adjusted residuals. She asked the
audience member to provide his name and tele-
phone number so that she could contact him
when the results are concluded.
She said that her approach was very similar to
the approach used in a comparable study on
trends along the Gulf Coast. One of her col-
leagues, David Dunn from Austin, Texas,
actually conducted that work, which is currently
being reviewed by the committee. She suggested
that the audience member, David Dunn, and
she, could discuss the topic further.
174
-------
Figure 101.
The Mississippi River and its major tributaries.
175
-------
An Assessment of Wat
Frederick C KoMM*".-^jjjK.
Gulf of Mexico Program Oftf^jijjjl
Stennis Space Cent^ri ^^sll^Cl
Abstract
Some inventories of ongoing watershed
based projects in the United States
contain as many as 700 entries. It is
estimated that 200-250 of these projects are
within the Mississippi River Drainage Basin.
The goals of the projects are varied: control of
nonpoint sources of nutrients or pesticides or
both; control of suspended solids from soil ero-
sion; education of the public about the value of
the watershed; monitoring of various pollutants;
protection of various fisheries; etc. On a
national level the projects range in size from
5 acres to over 150 million acres in size. Funds
have been obtained from federal, state, and
private sectors on many of the projects.
In a recently released report, the General
Accounting Office concluded that two factors
are necessary for success in watershed manage-
ment projects:
1. Flexibility in the kinds of financial and
technical assistance provided by the federal
agencies
2. Local tailoring of approaches to watershed
management that allows for differences in
the type and source of pollutants, local
agricultural practices and the community's
attitudes.
176
Overview of Existing Water-
sWeci Management Programs
From the beginning of the modern age people
have recognized that natural systems of varying
scale are related. For example, Jonathan Swift's
familiar little poem:
So naturalists observe a flea
Hath smaller fleas that on him prey;
And these have smaller still to bite 'em
And so proceed ad infinitum. (Swift, 1733)
It is not surprising then that participants in the
Gulf of Mexico Program understood from its
beginning that the Gulf of Mexico ecosystem is
connected to that of its watershed. At the first
meeting of the Public Health Committee in
March, 1989 members of the Public Health
Committee were overwhelmed by the enormous
size of the area from which pollutants that
could potentially affect the Gulf could arise.
Dr. Merrill McPhearson of the Food and Drug
Administration acknowledged that the Gulf is
influenced by its entire watershed, but by using
a deliberate, systematic approach the issues
could be dealt with. He suggested that we begin
at the bottom of the watershed, identify and
rank the public health problems, determine the
causes wherever they were located in the water-
shed and finally develop solutions. This is the
-------
path of action that all of the issue committees
have been following in developing characteriza-
tion reports and action agendas.
The members of the Gulf of Mexico Program's
Nutrient Enrichment Committee members also
recognized the potential impact of nutrients
getting into the Rivers far upstream on the Gulf.
Based on this understanding of the problem the
committee recommended that the Program
undertake several activities. One of the first
efforts was an attempt to develop a compen-
dium of sources and quantities of nutrients in
the rivers of the entire Gulf of Mexico water-
shed. This report was produced by staff at
Purdue University from USGS data taken from
the EPA STORET database (LJSEPA, 1992a).
While the report was not sufficiently detailed to
determine the sources of nitrogen to the Missis-
sippi River, it was clear that most of the nitro-
gen in the River at St. Francisville, Louisiana
was already in the River at Cairo, Illinois. A
companion report described the information
available on the effects of nutrient
overenrichment in Gulf waters; the hypoxic
zone on the Louisiana-Texas shelf was
described as an area where there was evidence
for excess nutrient input (U.S. EPA, 1992b).
In the spring of 1993 the Gulf of Mexico
Program partners organized and conducted a
Mississippi River Project to educate students
about the issue of nonpoint source pollution,
sources of nutrients, effects of nutrient over-
enrichment, and to make them aware of the fact
that their actions far up the Mississippi River
could affect the Gulf of Mexico. Vice President
Al Gore participated in the project. The Project
was successful in reaching students and the
public at large through the media coverage that
resulted. The project report states that, "Beyond
any doubt, students enthusiastically and over-
whelmingly responded to the project, its
activities and issues. Their letters, comments to
Vice-President Gore, and to reporters under-
score their desire to actively make a difference
in the quality of the environment." (Mote, 1993)
The effort to raise public awareness upstream
was continued this year when the Program
awarded a project (GMPO, 1995a) to develop
educational information and displays that will
illustrate the effects that excess nutrients in the
River can have on the Gulf and the importance
of controlling nonpoint sources of nutrients to
the River in the upper reaches of the watershed.
The Gulf Program participants have also recog-
nized that one of the most effective ways to
reduce nutrient input into the river upstream
will be to forge partnerships with similar groups
in the Gulf of Mexico watershed to control
impacts on the Gulf ecosystem. To this end we
have begun to identify these groups and have
now a preliminary inventory of 48 activities and
projects. (GMPO, 1995b) We intend to com-
plete and maintain this inventory of watershed
projects and to build a network for action and
education with these groups. This paper
describes what we have learned so far about
them.
We know from an inventory completed in June
of this year by the General Accounting Office
(GAO, 1995) that there are considerably more
projects than 48 in the Mississippi River Basin.
The national GAO inventory reported the iden-
tification of 618 watershed based projects aimed
at agricultural sources of pollution that have
received federal funding. I have estimated that
between 200 and 250 of these projects are
within the Mississippi River watershed. The
GAO reviewed nine of these projects in detail;
four of these projects were in the Mississippi
River watershed. The conclusions of the GAO
report will be summarized later in this paper.
Funding and initiative for many of the projects
in the Gulf Program's preliminary inventory are
provided by the federal government thru EPA,
NRCS, USGS, etc. The funding is provided
177
-------
either solely by the government or there is a
cost sharing between federal and state govern-
ment. Funding does not always refer to direct
monetary contribution; "in-kind" funding such
as technical expertise and cooperation were
provided for many projects by the "funding"
agency. At least five of the projects are being
conducted using no federal funds; they are
funded by private landowners and local
property taxes. The projects represent a wide
spectrum of management activities at a variety
of technical and nontechnical levels. All of the
projects are relatively new; the oldest project
began in 1986.
Of the 48 projects identified, 12 were basin
specific projects of the USGS National Water-
Quality Assessment Program. The purpose of
this Program is to identify factors that affect
water quality and monitoring to determine levels
of pollutants. At this conference USGS person-
nel have presented information based on these
projects. This information will be necessary to
determine changes in nutrient loading in the
rivers and consequently will allow for a measure
of the effectiveness of remedial actions taken in
the watershed. The remainder of the projects
focus on activities to manage and control pol-
lution problems.
The objectives of the management projects
ranged from increasing public awareness to
multiple objectives such as improving water
quality, developing public outreach documents,
and implementing best management practices.
Most projects have the general objective to
improve water quality.
The objective of the four projects occurring in
Pennsylvania is the control of drainage from
abandoned mines. However, one of the
treatment options being considered or actually
implemented in each of these projects is the
use of a passive wetland treatment system which
will be effective for control of nutrients and
sediments as well as acid mine drainage and
heavy metals which are the primary focus of the
projects. About 20 projects focus on three
specific objectives:
1. farm animal waste management
2. fertilizer use reduction
3. erosion control;
all of which will help reduce nutrient flux to the
rivers.
Three sets of the projects in the preliminary
inventory will be reviewed in some detail as
examples of the types of activities: Table 17
describes those projects in Arkansas that
received funding and support from NRCS,
Consolidated Farm Service Agency and the
Extension Service; Table 18 provides informa-
tion on the TVA River Action Team Projects
and Table 19 summarizes those projects from
the GAO report that received no federal fund-
ing. Many of these projects have as one of the
goals the reduction of nonpoint source pollu-
tion of the river by nutrients. The NRCS and
the TVA are active partners in the Gulf of
Mexico Program, so working with these
upstream projects will be easily accomplished.
Although many projects have resulted in
successes such as decreases in erosion or
reduction in fertilizer application, few of the
projects have been able to quantify their
successes (such as: reduction of soil erosion by
7 ton/acre or nitrogen application reduced by
70 pounds/acre). Participants in almost all of
the projects reported "unquantifiable" successes,
including reports from farmers that they were
satisfied with best management practices and
new technologies introduced by the program
and they would continue to use them;
establishment of citizen action and interstate
cooperative groups; distribution of public
outreach material. These "unquantifiable"
178
-------
successes are the first step to implementing
actions that will lead to quantifiable results.
Because environmental problems cannot be
corrected instantly, a series of indicators of
progress toward the ultimate environmental
goal is needed. The Gulf of Mexico has adopted
a hierarchy of indicators to measure success in
achieving the many steps toward our goals. This
hierarchy is shown in Figure 102. A suite of
indicators to indicate progress toward reducing
extent, severity and duration of the hypoxic
zone in the Gulf will be developed as part of the
strategic assessment and planning process.
The General Accounting Office inventory of
watershed based projects was limited to those
that have received federal funding and are
aimed at agricultural sources of pollution. If it is
assumed that the projects in states that are only
partly in the Mississippi River watershed are
distributed uniformly across each state, it can be
estimated that between 200 and 250 of these
projects are within the Mississippi River water-
shed. Nationally the projects ranged from as
small as five acres to over 150 million acres in
size; they involved both surface and ground
water resources; and they addressed such
agricultural pollutants as animal waste, fertilizer
runoff, pesticides and soil sediment. Through
early 1995 these projects had received an
estimated $514 million in federal funds.
They reviewed nine of these projects in detail;
four of these were in the Mississippi River
watershed. Table 20 presents a few facts about
these projects.
The Project participants pointed out that even
given rigorous monitoring, demonstrating a link
between changes in land use and diminished
chemical pollution is difficult, if not impossible,
especially within a short time frame. Participants
in several projects noted that current science
can demonstrate only a tenuous link between
land use practices and water quality, and it may
take years for their projects to produce chemical
improvements in water quality. Participants in
the Big Spring Basin project said that climatic
variations, such as droughts followed by years
of heavy rainfall, and other factors have made it
difficult to establish a link between changes in
farming practices and groundwater quality,
despite more than 10 years of monitoring and
analysis.
The GAO reported that while their conclusions
from a thorough study of 9 watershed projects
cannot be projected to the entire inventory of
618 projects, participants in all nine agreed on
two key factors for success that have been
learned during the course of the projects:
1. Flexibility in the kinds of financial and
technical assistance provided by federal
agencies
2. Local tailoring of approaches to watershed
management.
Because watershed projects differ in character-
istics such as the type and source of pollutants,
local agricultural practices, and the community's
attitudes, participants believed that a prescrip-
tive, one-size-fits-all approach would be
inappropriate. At the local level, the projects'
participants emphasized to the GAO that the
keys to reducing agricultural pollution include
1. Building citizens' cooperation through edu-
cation
2. Getting stakeholders to participate in
developing the project's goals
3. Tailoring the project's strategies, water
quality monitoring, and regulatory
enforcement efforts to local conditions.
179
-------
Conclusion
•*-
-Tlefer
All of the watershed protection activities were
begun to protect the water quality of a particular
section of a creek or river in this great water-
shed; none were undertaken specifically to
reduce the severity or extent of the hypoxic area
in the Gulf. However, the combined effort of all
of these management projects and the support
of citizens living in the Mississippi River Basin
should result in measurable improvements in
the Mississippi River water quality that will be
detected by monitoring projects such as the
USGS National Water-Quality Assessment
Program and ultimately reduce the nutrients
reaching the Gulf via the River. In fact, while it
is too early to say with certainty, the summary of
existing data as presented by Turner and
Rabalais (1991) indicates that nitrate-nitrogen
concentrations in the Mississippi River at St.
Francisville and New Orleans, Louisiana may
have begun decreasing in the late 1980's which
would be commensurate with the formation of
these pollution management projects within the
basin.
As the Gulf Program works to develop viable
solutions through a strategic assessment proc-
ess, we will rely to a great extent on these exis-
ting programs to provide existing data, to
participate in the assessment process, to identify
and prioritize the areas of greatest need and to
undertake demonstration projects to initiate
implementation of the strategic plan.
ences
GAO, 1995. Agriculture and the
Environment—Information on and
Characteristics of Selected Watershed
Projects. United States General Accounting
Office. GAO/RCED-95-218,June 1995.
GMPO, 1995a. U.S. EPA Interagency
Agreement #DW-12-94-5693 with USDA,
Natural Resources Conservation Service.
Mississippi River Watershed Nutrient
Exhibit. August 1995.
GMPO, 1995b. Gulf of Mexico Program
Office, U.S. EPA. Collection and Analysis
of Information on Watershed Management-
Activities in the Mississippi River Basin.
Draft Report, EPA Contract No. 68-C2-
0134, Work Assignment 2-202, Task 6.
September 30, 1995.
Mote, 1993. Mississippi River Project for
America's Sea. Mote Marine Laboratory,
1600 Thompson Parkway, Sarasota, FL
34236,1993.
Swift, 1733. Swift, Jonathan. On Poetry—A
Rhiapsody. 1733
Turner and Rabalais, 1991. Turner, R. E. and
Rabalais, N. N., Changes in Mississippi
River Water Quality this Century.
Bioscience, 41,140-147, March 1991.
USEPA, 1992a. Sources and Quantities of
Nutrients Entering the Gulf of Mexico from
Surface Waters of the United States. U.S.
EPA, EPA 800-R-92-002, September 1992.
USEPA, 1992b. An Updated Summary of Status
and Trends in Indicators of Nutrient
Enrichment in the Gulf of Mexico. U.S.
EPA, EPA 800-R-92-004, September 1992.
180
-------
Title ,.
Long Creek
Ag. NFS HUA1
Millwood Lake
Watershed
Demo. Project
Muddy Fork of
the Illinois R.
Ag. NFS HUA
Table 17.
Watershed projects identified in Arkansas.
Began
1991
1991
1990
Acres
96,574
1,325,000
47,122
Budget/Year
-$1.6 million •
(proposed 5-yr budget)
~$l.5 million
(proposed 5-yr budget)
~$2.5 million
(proposed 6-yr budget)
j, >;-« Purpose/Goal '•;
Reduction of nutrients and
pathogens from animal waste
Provide for disposal of manure
from poultry and swine operations
using BMPs2
BMPs include waste management
for confined animal operations and
nutrient management for pasture
lands
'Agricultural Nonpoint Source Hydrologic Unit Area
2Best Management Practices
Rat Title '
Flint Creek
Wheeler Elk
Chickamauga
Hiwassee
Holston
Watts Bar...
Clinch-Powell
Began
1992
!993
1995
1992
1993
1994
1993
Table 18.
TVA Sponsored River Action Teams (RAD.
-'Area -"-,„
290,000 acres
S.ISOsq. miles
1 ,865 sq. miles
2,700 sq miles
3,776 sq miles
1 ,370 sq miles
2,954 sq miles
Fiinds/Yr
$2.15 million
FY92-FY94
Not available
Not available
~$l million
Not available
Not available
Not available
Goals/Actions
Animal waste lagoons; Wastewater
irrigation systems; composters for
poultry operations; no-till agriculture
Install oil/water separator on parking lot;
encourage agricultural BMPs
Acid mine drainage remediation
Stream bank stabilization; animal waste
management; erosion control; public
education; 20-30 projects/ year.
Identification of NFS sources; Installation
of 6 Agricultural BMPs; Elimination of
unpermitted discharges; Monitoring
Goal is to solve pollution problems in the
watershed. Public education and
outreach.
Constructed animal waste treatment
system and live stock exclusion fences,
revegetated riparian zones
181
-------
Table 19.
Projects receiving no federal funding.
Title
Piasa Creek Watershed
Partnership
Pontiac/Skeator Watershed
Area
Chippewa River
Stewardship Partnership
S. Washington Watershed
District
Tributaries of Stillwater,
Rock Creek
State
Illinois
Illinois
Minnesota
Minnesota
Montana
Project Sponsors ,
Am. Farmland Trust
Piasa Creek Conservancy
Great Rivers Land Trust
Northern Illinois Water
Company
Am. Farmland Trust
Chippewa R. Stewardship
Partnership
City of Woodbury
MN Board of Water and
Soil Resources
Land and Water Services
and private landowners
Goals/Activities
Establish water retention
basins. Create field
buffers and filter strips.
Develop whole farm
nutrient plans.
Reduce fertilizer use.
Involve and educate
citizens. Comply with
nitrate/nitrite DW
Limits.
Improve Water Quality.
Restore wetlands and
riparian areas. Reduce
Flood Damages.
Management of urban
runoff to control
nutrients, sediment,
salinity. Control of
flooding.
Reduce soil erosion and
nitrogen by moving
corrals and stockyards
away from the river and
revegetating banks.
Selected watershed broi
Watershed
Otter Lake, IL
Big Darby Creek, OH
Black Earth Creek, Wl
Big Spring Basin, IA
Table 20.
ects in the Mississib t>i River drainage basin from GAO report.
Area (Acres)
12,255
371,000
64,000
66,000
Total Funds
$292K
$5,I45K
$3,245K
$7,II9K
Duration
1992-1995
1990-1995
1986-1994
1982-1993
182
-------
Presentation Discussion
Fred Kopfler (Gulf of Mexico Program—
Stennis Space Center, MS)
Daniel Ray (The McKnight Foundation-
Minneapolis, MN) gave an example of the
scale of the community-based watershed
management infrastructure in place in the upper
Midwest, saying that the McKnight Foundation
is involved with supporting a network of
community-based activities that span through-
out the Mississippi watershed and above the
quad-cities. He believed this area was approxi-
mately 12 percent of the overall Mississippi
River Watershed. He said they have identified
about 110 community-based, initiatives that
cover about one third of the watershed, not
including the involvement of federal land. He
finalized his comments saying that there is a
huge infrastructure available and ready for some
direction to pursue a strategy. This infrastruc-
ture should be effectively linked into the Gulf
Coast comprehensive program.
Fred Kopfler told the audience that Daniel Ray
was one of the individuals contacted by Battelle
who provided a lot of information. Much of the
material was received by Battelle too late in the
government fiscal year to be incorporated into
the report. The Gulf of Mexico Program has
some revisions and changes in format that they
will give to Battelle for incorporation into the
final report and these changes will include all of
the updated information. He agreed to follow
up with Daniel Ray to make sure all those
organizations are included.
Administrative Environmental
i....Chatigb&m
Discharge/
emission
Quantities
in Ambient
Conditions
^
Figure 102.
An illustration of a hierarchy of indicators that can be used to track progress toward an
environmental goal before results are measurable in the field.
183
-------
What is B$iHg
• %'^a™ iipiiii
Nutrient LioaM&iMff
Charles Spoony ' j^'
Environmental Protect/on Agency;
Office of Water ,. ;;:, j,£
Washington, DC 20WL^
Abstract
t~ t ~. ' •*•
Presentation Discussion
Nutrients have been identified as
problems in many places in the
Mississippi River Basin making them
one of the most ubiquitous and complex
categories of water pollution. Control efforts
tend to be localized and tend to focus on
controlling phosphorus, the most important
nutrient in freshwater systems. Concerns for
nitrogen are found where nitrate's toxicity in
drinking water and ammonia's toxicity to fish
are noted.
Nutrient concerns are always expressed in their
relationship to specific water bodies, and these
concerns are seldom seen as issues that affect
more than one water body. Examples of
projects in the watershed and examples drawn
from successful programs in other areas will be
eked.
No Manuscript Submitted.
Charles Spooner (Environmental Protection
Agency/Office of Water)
Beverly Ethridge (U.S. Environmental Pro-
tection Agency, Water Quality Division-
Baton Rouge, LA) commented that the
Biennial State Water Quality Reports attribute
about 60 or 70 percent of the Nation's water
pollution to nonpoint sources. She asked
Charles Spooner if that percentage is correct,
and if the nonpoint source budget is about
$100 million, how it compares to the remainder
of the water budget which is presumably
targeted for point source control.
Charles Spooner stated that Beverly Ethridge
was making the point that the Section 319 Pro-
gram is relatively small. He said that it was
explained early in the conference that one of the
reasons they reacted negatively to the Section
319g petition was because the remedy for a
Section 319 petition is to invoke controls on the
Section 319 Program which would, ultimately,
lead to the EPA leveraging a small program
within the agency. The major capital funding for
the state revolving fund is large. Therefore, it is
not difficult to list two states that have con-
struction loan subsidies that would equal the
nonpoint source program nationwide. It is
encouraging that the rules are allowing greater
flexibility to access those funds to support non-
point source control capital costs.
184
-------
MWMUpH-^ltt • \,\;*4mMl«'i-i~:v'tieftS-* '<
113 ¥tl rf'
^.j4v-«n. ^ ^.^s^^Si™- *" ^ „« ^v , s^gs *v" VT-^t.^** J
&^x- j/~-*!-"*j5= — s.^: - v * J *~ * - --> %g- — - ,
T Abstract;; , / ..... \ _ -„,-''
The 1985 and 1990 Farm Bill has pro-
duced a conservation revolution across
the country. Farmers are using conser-
vation systems to reduce erosion, almost elimi-
nating wetland drainage, restoring wetlands, and
using space technology to refine nutrient and
herbicide application for maximum efficiency.
These efforts have reduced erosion on highly
credible land by 70 percent and on all cropland
by 33 percent. In addition, the combination of
drastically reduced drainage and wetland
restoration and creation has begun to show a
net gain in wetland acreage in agriculture areas.
The Mississippi River Basin should be yielding
less nutrients and sediments from agricultural
nonpoint sources.
Agricultural sources of nonpoint source pollu-
tion are a major nutrient load to the Gulf of
Mexico as indicated in various water quality
reports. While the storm water runoff carries
natural nutrient loads and a mix of nonpoint
source loads, it is generally agreed that agri-
cultural sources are significant contributors.
Agriculture is by far the dominant land use in
the Mississippi River Basin. Before 1985, the
basin contained approximately 56 percent of the
Nation's land in farms and 80 percent of the
Nation's cropland. After the Conservation
Reserve Program (CRP) contracts converted
cropland to permanent cover, the area still
contained 65 percent of the Nation's cropland.
This area produces 84 percent of the Nation's
corn, 81 percent of the soybeans, 59 percent of
the wheat and 57 percent of the hay. Often, the
upper Mississippi River Basin is referred to as
the Nation's "Bread Basket," relating to the
large grain production.
Mississippi River Basin Production
• 84 percent of national corn production.
• 81 pecent of national soybean production.
• 59 percent of national wheat production.
• 65 percent of the Nation's cropland.
For agriculture to remain profitable and
efficient, agricultural chemicals—both pesticides
and fertilizers—are necessary. These chemicals
vary from pre- and post-herbicides, to insecti-
cide and fungicides, to nutrients from animal
manure and fertilizer. Since nutrients are the
major concern in the Gulf of Mexico, this paper
will concentrate on nutrient availability and soil
erosion, which is an indicator of nutrient
movement from agricultural land.
Bask Transport Mechanisms
While reviewing the following information, the
basic nonpoint source transport mechanisms
should be kept in mind. These mechanisms are:
availability, transport, and in-stream integration.
185
-------
The availability of agricultural nutrients is
related to soil erosion and the presence of
nutrients on the surface or in the soil. If soil
erosion and excess nutrients are reduced, a
similar reduction in downstream nutrient load
should occur.
The transport is the movement of the available
material —soil, chemicals, organic matter,
manure, and so forth—from the field to the
stream. The actual amount of nutrients and soil
transported depends upon the transporting
forces and interceptions in the transport path. If
the runoff flow is interrupted due to terraces in
the field, filter strips, or wetlands, some eroded
soil and nutrients will be removed.
The in-stream integration is the assimilation of
the transported material into the stream, lake, or
estuary environment, creating some impact—
good or bad depending upon many aquatic
factors.
'Basic
* Availability — Soil erosion, chemical use,
concentrated manure production.
• Transport — Energy of sheet and
concentrated flow from precipitation.
" Instream integration — Effects on the aquatic
environment.
Once the flow enters a water body, the potential
for managing the nonpoint source loads is lost.
However, the effects on the water body must be
understood to determine if additional load
reductions are necessary.
With the 1985 and 1990 farm bills, improved
technology, and farmers better-educated on
environmental issues, the nutrient loads to the
Gulf from agriculture should be declining.
tl985 and 1990 Farm Bills
The 1985 Farm Bill, known as the Food
Security Act (FSA), set in motion some major
conservation policy that has generated a
revolution across the agriculture community.
These policies are embedded in the Conserva-
tion Title of the Farm Bill.
•arm Bill, Conservation Title
• Conservation Compliance Provisions
• Conservation Reserve Program
• Swampbuster Provisions
• Wetland Reserve Program
Common names are applied to these provisions
and programs rather than the actual Farm Bill
terms. Briefly, these programs and provisions
had the following conditions.
HTgeryation Compliance
If farmers were participating in any USDA
program — commodity deficiency payments,
loans, and so forth — they had to provide soil
conservation treatment on their highly erodible
Conservation compliance provisions were
implemented in two stages. From 1985 until
1990, conservation compliance plans were
developed. This plan scheduled the cultivating
systems, structures, and crop management
systems that would be necessary to adequately
reduce erosion on the highly erodible land. The
implementation schedule must have all the plan
installed by December 31, 1994.
Visual
1985—
-»- 1990-
-*• 1994
Develop Conservation
Compliance Plan
Implement Conservation
Compliance Plan
186
-------
The Natural Resources Conservation Service
(NRCS) (formerly the Soil Conservation
Service) has a soil loss tolerance, commonly
called "T" value, assigned to each soil in the
United States. The "T" value is expressed in
tons/acre allowed for sustainable agricultural
production. In the development of the con-
servation compliance plans, the agency had to
develop conservation systems that meet four
primary criteria as described below.
Criteria for^Conservation
Compliance Systems _ -_/
1. 50 percent or more soil erosion reduction
2. Economically feasible
3. Locally practical
4. Socially acceptable
Produced Alternative Conservation Systems
These criteria often required conservation
systems with soil loss higher than the "T" value
for the soil. However, the system, called
alternative conservation systems, had to
achieve at least 50 percent or greater soil
erosion reduction.
By the deadline of Dec. 31, 1994, over
90 percent of the farmers had implemented
their compliance plan. This has been a major
conservation revolution in the country-side and
the farm community is very proud of its accom-
plishment. Everyone worked together to get the
job done. Not only farmers but also equipment
dealers, farm organizations, chemical dealers,
and, above all, the farm press had a major
impact on reaching the conservation goals.
To meet the goals of Conservation Compliance,
1.7 million conservation compliance plans were
developed on 143 million acres of the Nation's
most highly credible cropland. This is about
one-third of all cropland in the United States. A
major portion of this land is located in the
Mississippi River Basin.
1 - /
Conservation Compliance
• 1.7 conservation compliance plans.
• 143 million acres of highly credible land.
Conservation; Reserve Program /
• 36.4 million acres
• Cost—1.8 billion per year
As part of treating highly credible land, the
Conservation Reserve Program was offered to
retire land for 10 years. Through 12 sign-ups,
36.4 million acres of cropland was placed under
contract and planted into permanent vegetation,
which essentially eliminates herbicide and
fertilizer use on those acres (Figure 103).
» v A ~ " "* / " ^ ">
_ f-r -s- TU, j
Results of Conservation Compliance
and Consjeryatfon Reserve Program
The combination of conservation compliance
along with the conservation reserve program
reduced erosion rates by 70 percent on the
highly credible land based on 1994 status
reviews and by one-third on all cropland based
on the 1992 National Resources Inventory
(NRI). The NRI is a sample of the Nation's
resources from 800,000 points across the
country (Figures 104, 105, 106).
The reduction in nutrient availability must be
having a significant effect eventually on the
Gulf. However, immediate response is not
anticipated when the complexity of the
transport mechanism, the entire Mississippi
River system, is considered. It takes time for
the streams to purge the current loads and make
adjustments.
187
-------
Sy/ampbuster
Another program that should be helping to
reduce nutrient loads is the swampbuster
provisions. Simply stated farmers cannot drain
or fill wetlands and still receive USDA program
benefits. This provision has drastically reduced
wetland drainage. The remaining wetlands will
continue to serve as a major nutrient and sedi-
ment trapping system in the transport process.
Based on the 1992 NRI, the rate of wetland
drainage due to agriculture has been reduced to
31,000 acres per year (Figure 107).
Wetland Reserve Prograrri
The wetland reserve program has restored
almost 134,000 acres of cropland back to
wetlands under perpetual easements. The
landowners have offered seven times as many
acres for the wetland reserve program as funds
are available to accept.
The restored wetland acres are often in a
position to receive some drainage from
agricultural land and serve as a trap in the
transport mechanism.
The combination of Swampbuster and Wetland
Reserve Program has probably turned the
wetland acreage in agricultural areas to a net
gain rather than a loss.
Impact on Water Quality in the
Mississippi River
Some of the papers presented at this conference
may estimate change in water quality within the
Gulf or Mississippi River. The U.S. Geological
Survey (1) data for 1980 to 1989 shows nitrate
level dropped 0.4 percent per year in the Upper
Mississippi River Region, and the Lower
Mississippi River Region dropped 1.6 percent
per year. The phosphorus levels decreased
from 1 to 1.7 percent per year in the northern
parts of the Mississippi River Basin and
decreased yearly from 3.1 to 3.8 percent in the
lower Mississippi River Basin. Clearly, this
reduction is not due solely from nonpoint
sources. Reductions in point source discharges
and different weather patterns could be the
major reasons for the nutrient level reductions.
However, a significant reduction from
agricultural nonpoint sources should continue
this trend.
piewTecKnofbgy „
New technology that is being implemented will
offer an additional reduction in nutrient move-
ment from the fields. Some refer to this tech-
nology as "Prescription Farming" while others
refer to it as "Precision Farming." The technol-
ogy is still in its infancy, but it is operational,
proving profitable, expanding rapidly, and
optimizes nutrient consumption by the plants.
With the use of Geo-Positioning Systems
(GPS), a small four-wheeler can move across
the field and take soil samples on a grid while
recording the exact position in a compute file
using GPS.
While the harvesting equipment moves across
the field, it is collecting data continuously on
crop yields and recording the data with GPS
markers. The farmer has a color coded record
of the crop yield variability across the field and
the file is stored for computer use. The crop
yield can be overlaid with soil types to
determine potential crop yields.
With this information, the data are loaded into a
computer mounted on a fertilizer and herbicide
application truck, which applies fertilizers based
on potential yield and nutrient content of the
soils and applies herbicides to weedy spots in
the field rather than the entire field. When the
seed is planted, the seed density per area is also
188
-------
varied based on soil potential yield and the
nutrients applied.
It is space-age technology applied to farming.
In most cases, the more efficient use of
chemicals and chemical savings will off set the
cost of using the equipment on a custom fee
basis. Now, rather than applying the herbicides
and fertilizers across the field based on average
field conditions, the chemicals can be precisely
applied for the most efficient use (Figures 108,
109, 110).
New Technology
• Using GPS technology and computers.
• Grid sampling for nutrient analyses.
• Yield monitoring with GPS markers.
• Pest weed problems identified with GPS
markers.
• Computer controlled herbicide and fertilizer
application truck linked to GPS.
• Seeding based on soil potential yield and
linked to GPS.
This technology is not available everywhere due
to lack of equipment and trained staff, but it is
rapidly expanding. On the horizon, this
technology has the potential of being the best
method for reducing nutrients from the field
and still maintaining a productive agriculture.
Farmers, in general, have been striving for
better nutrient utilization, and the trend of
lower nutrient consumption per unit of
production for some commodities supports
their success.
When animal manure is used for part or all the
nutrient needs of the plants, farmers are still
having some difficulty estimating nutrient
availability during the growing season. This is
especially a problem with nitrogen. Annual soil
and manure testing helps build confidence on
accuracy of estimates.
.What Does the Future Hold?
Much of the progress described above is
dependent upon the 1995 Farm Bill and Budget
Reconciliation. Although neither bill has been
passed into law, it is obvious some major
changes will occur. Commodity payments may
be reduced significantly in Budget
Reconciliation. Due to commodity payments
and loan reduction, Congress feels that some
regulator relief is warranted. Therefore, some
of the farming constraints that were a condition
for receiving commodity support, loans, and so
forth, will be reduced. In addition, farmers will
be released to plant any crop and not be
constrained by the Farm Bill. In addition, set-
aside land will be released for planting. The
entire process is to move agriculture to a more
free market-based economy.
Based on the current debate, the following
conditions may occur.
• Conservation Compliance—Farmers out of
compliance will have reduced penalties.
Plus the penalty will be limited to the field
out of compliance.
• Conservation Reserve—Acreage enrollment will
probably drop to 50-70 percent of the
current enrolled acres. Enrollments in
Conservation Reserve started in 1986 with
the big enrollments in 1987 and 1988.
These 10-year contracts will soon expire.
• Swampbuster—Some wetlands will be exempt
from, swampbuster. The wetlands farmed 6
out of 10 years and 1 acre or less may be
exempt from swampbuster. This will
exempt 6 to 10 million acres of wetlands
189
-------
could reduce the commodity prices, but at some
additional cost to the environment.
The program will still be supported but at a
lower funding level. Perpetual easements may
not be allowed. Instead, WRP contracts will be
used and they may vary from 15 to 30 years
depending upon who wins during Budget
Reconciliation conference.
The bottom line is a much weaker Conservation
Tide in the farm bill.
r™ - • - = - -™ ........ »""" ~ ........ «i ......... ="- ........................ i,-ii:--^y4.^»™-.T-.|
|f995 Parm fell" Potential Impacte ""
9f_[\ - - _ w*l - u^Jwiu — Ih JU^^->ojii^^L^.j..JIE^-jLJrfu^^-. j^uta-iJattfAut-J-a-ai-—*. --- _— . --- »
Conservation Compliance — Penalties reduced.
Penalties limited to field not in compliance.
Conservation "Reserve — Funding reduced.
Fewer acres in reserve.
Swampbuster — Frequently cropped wetlands
and small wetlands exempt. 90 percent of
pothole wetlands are vulnerable.
Wetland Reserve Program — No perpetual
easements — only contracts. Funding
reduced. Acreage capped at 975,000 acres.
rpmQfjjty Prices
..... i
American agricultural products are and will
continue to be major contributors to world
trade. Bad weather in some regions of the world
and increased demand from countries such as
China, have increased the prices of commodities
in recent months. Commodity payments for
wheat and loans for cotton will probably not be
necessary for the 1995 crop year due to the
increased prices. It is anticipated that upward
pressure will continue on commodity prices
through 1996 and possibly beyond. However, if
the 1995 Farm Bill completely releases the
farmers from all restrictions and weather is
good, a big crop yield from the United States
Increased demand and prices for certain crops
combined with a weaker Conservation Tide in
the 1995 farm bill presents the possibility for a
setback to the environment. One major hope,
and a valid hope, is that the farmers realize their
accomplishments, retain their conservation ethic
and continue conservation farming.
It would help to reinforce the conservation
ethic if the farmers had a clear understanding of
the potential impact of their actions on the
Gulf. They are witnessing local improvement in
water quality and wildlife but connecting that
improvement to the Gulf is difficult to recog-
nize from a farm in the upper Mississippi River
Basin. Communications between the Gulf and
the farmers in the Mississippi River Basin needs
improvement.
However, to get major reductions in nonpoint
sources over a large area like the Mississippi
River Basin, Farm Bill policy must be
compatible with the environmental objectives
for the basin.
ummary
Agricultural nonpoint source loads in the Mis-
sissippi River Basin have been reduced due to
the programs and provisions of the 1985 and
1990 Farm Bills. The future of these accom-
plishments rests in the 1995 Farm Bill and
Budget Reconciliation. These two bills will
determine if this country is willing to maintain
the present accomplishments, possibly make
some improvements, or allow for some back-
sliding. Widi the decline in current world grain
supply reserves, the opportunity is present for
some reversal from the current accomplishments.
190
-------
Reference ^ ^^ ^ -__'"/
Smith, R.A., Alexander, R.B., and Lanfear, K.J.,
1993, National Water Summary 1990-1991,
Page 129.
Discussion
John Burt (U.S. Department of
Agriculture/Natural Resources Conservation
Service- — Washington, D.C.)
No questions/discussion after John Burt.
191
-------
Conservation Reserve Program Acres
(Signups i-18)
Figure 103.
50
40
fc 30
20
-------
Figure 105.
Summary 1994 Status Review Results—Distribution of erosion rates after full
implementation with respect to the soil loss tolerance values preliminary data as
ofFebruary 9', 1995
Change in Ave
by Wind and Vateir on
Figure 106.
193
-------
Agriculture Wetlands Loss Down
Wetland losses caused by agricMltural
activity have slowed considerably
between 1954 and 1992.
Thousands of acres
lost annually
Source: USDA Natural Resources Conservation Serylic^; NgUowal Re^ur^slw^ntory, NR1
NRi data cover the 43 contiguous stales, Hawaii. Puerto j=ScoT and the U.S. VirgpiniMgnas, but not AfasKa.
Source for "54 - 74 and T4 • '63 data: U.S. Department of the Inferior,
Figure 107.
194
-------
Nutrient Rate per Acre Planted
s.
O.14-
0.12-
0.1 -
0.08-
o.oe -
O.O4-
O.02 -
0
1964 1969 1974
Sources: AER 717; Cropping Practices Survey Data
1979
Year
1984
1989
1994
Figure 108.
Nutrient Rate for SB & Wheat
S.1.S-
O.5 -
1904 1B6»
Souroes: AER 717; Cropping Practices Survey Data
1S7S
Year
1984
Figure 109.
Nutrient Rate for Cotton
Nubtanto pw Bmb,
1984
19S9
1874
187D
Year
1984
1989
1994
Source*: AER 717; Cropping Practices Survey Date
Figure 110.
195
-------
. . . . ,. ... ' . W'^ranHHi
Louisiana Activities and Progra
Control and Management?! j
Dugan S. Sabins and Jan R. Boydstun .-Jp^liH^^
Louisiana Department of Environmental Quality < I ••?, :§\ vf||||
Baton Rouge, Louisiana - _ ';:ltlsllii!i|
* I
Abstract
Louisiana has implemented many activities and
programs to address the problems of nutrient
enrichment For many years, most activity was
associated with point source control programs
for municipal and industrial discharges, many of
which had significant nutrient discharges. These
point source control activities have led to sub-
stantial reductions in the nutrient loadings to
state water bodies and are continuing. Recently,
however, it has been recognized that diffused
rainfall runoff from a variety of "nonpoint"
sources are now contributing to the majority of
man-induced nutrient loadings to Louisiana
water bodies. To address the nonpoint sources
of nutrients, Louisiana, as have most states, has
initiated an aggressive nonpoint source program
designed to work cooperatively with the
Environmental Protection Agency under
Section 319 of the Clean Water Act. Louisiana's
Nonpoint Source Program has sought to sup-
port and incorporate existing local, state, and
federal agency programs and enter into a broad
cooperative "interagency" approach to the
problem.
One of the first water bodies in the state that
was targeted for nonpoint source implemen-
tation activities, which included nutrient
controls, was Bayou Queue de Tortue in the
Mermentau River Basin (Figure 111). Over a
five year period, the program has seen the
development of best management practices for
rice cultivation that has benefited water quality.
The state believes the Bayou Queue de Tortue
experience shows that, working cooperatively,
nutrient and water quality goals can be achieved
(Figures 112 and 113). Other nonpoint source
program activities addressing forestry practices
and urban runoff, although not as far along in
the development, are also showing promise in
reaching water quality goals. The state is
committed to pursuing whatever point and
nonpoint source controls are necessary to
address nutrient enrichment problems and
believes its existing programs are achieving this
goal.
No Manuscript Submitted.
Presentation Discussion
Dugan Sabins (Louisiana Department of
Environmental Quality—Baton Rouge, LA)
There were no questions/discussion following
Mr. Sabins' presentation.
196
-------
Bayou Queue de Tortue
Gueydan, Louisiana
58010046
Post BMP: 1990-1995
Annual Average DO = 3.512 mg/L
Winter Average DO
Spring Average DO
Summer Average DO
Fall Average DO
Figure III.
5.639 mg/L
3.316 mg/L
2.144 mg/L
2.938 mg/L
Bayou Queue de Tortue
Gueydan, Louisiana
58010046
Pre BMP : 1982 - 1989
Annual Average DO = 2.434 mg/L
Winter Average DO
Spring Average DO
Summer Average DO
Fall Average DO
Figure 112.
4.917 mg/L
2.083 mg/L
1.251 mg/L
1.483 mg/L
197
-------
MERMENTAU RIVER BASIN (05)
0308 \ osozot
CROWLEY
_ Ambient Water
Quality Sample Site
•—— Basin Boundary
Subsegment
Boundary
State Border
— Parish Boundary
O City > 10,000 pop.
• Parish Seat
\
N
i
75 mi.
Figure 113.
198
------- |