United States
Environmental Protection
Agency
Chesapeake Bay
Program
Annapolis MD 21401
Research and Development
EPA-600/S3-84-015 Feb. 1984
&EPA Project Summary
Submerged Aquatic
Vegetation in Upper Chesapeake
Bay: Studies Related to
Possible Causes of the Recent
Decline in Abundance
W. Michael Kemp, Walter R. Boynton, J. Court Stevenson, Jay C. Means,
Robert R. Twilley, and Thomas W. Jones
This study synthesizes research con-
ducted on possible causes of the decline
in abundance of submerged aquatic
vegetation (SAV) in upper Chesapeake
Bay beginning in the late 1960's. Three
factors potentially were emphasized in
this study: runoff of agricultural herbi-
cides; erosional inputs of fine-grain
sediments; nutrient enrichment and
associated algal growth. Widespread
use of herbicides in the estuarine water-
shed occurred contemporaneous with
the SAV loss, but extensive sampling of
estuarine water and sediments during
1 980-81 revealed that typical bay con-
centrations of herbicides (primarily atra-
zine) rarely exceeded 2 ppb. However,
normal concentrations (< 5 ppb) were
shown experimentally to have little
measurable effect on plants. Increases
in turbidity have been documented for
some bay tributaries since the 1940's.
Light (PAR) attenuation by suspended
fine-grain sediments contributed more
to total turbidity in bay shallows (<
1.5m) than did phytoplankton chloro-
phyll a. Evidence indicated that plant
photosynthesis was light-limited for
much of the day. Effects of the continual
increase in nutrient enrichment of the
bay (documented since 1930) were
tested by experimentally fertilizing pond
mesocosms at levels common to the
upper estuary. Moderate to high nutri-
ent loadings resulted in significant in-
creases in growth of epiphytic and
planktonic algae and decreases in SAV
production.
This Project Summary was developed
by EPA's Chesapeake Bay Program.
Annapolis. MD, to announce key find-
ings of the research project that is fully
documented in a separate report of the
same title (see Project Report ordering
information at back).
Introduction
It is widely recognized that submerged
vascular plants play an important role in
the ecology of littoral regions of lakes,
estuaries and oceans. While a number of
studies have noted the ability of these
plant communities to attenuate variability
of nutrient, sediment and production
cycles, several such communities have
themselves undergone extreme fluctua-
tions in distribution and abundance For
example, in the mid 1930's a widespread
die-off of the seagrass, Zostera marina,
was well documented throughout the
North Atlantic coastal regions The cause
of this occurrence has never been un-
equivocally established, although recent
suggestions have pointed to subtle cli-
matic shifts. Other reports of regional
declines in abundance of submerged
aquatic vegetation (SAV) have indicated
the possible influence of human activities.
Few of the reported SAV declines have
occurred in estuarine environments and
most have involved 1 or 2 plant species
-------
However, in one of the world's largest
estuaries, Chesapeake Bay, a major loss
of SAV has continued from the mid 1960's
to the present. More than 10 species have
experienced significant decreases in
abundance, including Potamogeton per-
foliatus, P pectinatus, Valisneria amen-
cana, Zannichellia palustria, Ruppia
maritima as well as the marine speciesZ
marina. In the upper estuary this decline
in native species was preceded by an
invasion of the exotic Myriophyllum
spicatum, which eventually also died
back. Studies of seed and pollen distribu-
tion m sediment cores from the upper bay
have demonstrated that this diminution
m plant abundance is unprecedented for
at least the last century In general, it
appears that the recent decline occurred
first and with greatest intensity m the
brackish waters of the estuary, with Z.
marina communities m the lower bay
being affected less and somewhat later.
Numerous mechanisms have been
cited as possible causes of this occur-
rence The concept that natural entrained
population cycles or global climatic events
might be responsible seems unlikely m
view of the range of biological and
physiological characteristics for the num-
erous species involved In addition, there
is no parallel trend in plant abundance
apparent in nearby coastal regions. Other
factors including animal foraging and
grazing and major storm events are
probably of occasional and local impor-
tance, but these are part of the normal
milieu to which SAV are exposed and
hence are insufficient to explain this
abnormal decline. The absence of correla-
tions between distribution of SAV and
industrial pollutants renders such an-
thropogenic wastes an unlikely cause;
however, more general changes in water
quality associated with diffuse sources
(e g , runoff) do represent a potential
explanation These include increased
fine-gram sediments from land erosion,
increased algal growth from nutrient
enrichment of estuarine waters, and
aqueous concentrations of herbicides
arising from agricultural runoff
The full report presents results of
research conducted at the University of
Maryland's Centerfor Environmental and
Estuarine Studies concerning factors
potentially involved in the decline of
submerged aquatic vegetation m upper
Chesapeake Bay. The research examines
three mam factors in relation to SAV
growth and production, agricultural herbi-
cides, suspended sediments and associ-
ated light attenuation, nutrient enrich-
ment and resulting algal growth and light
attenuation.
Approach and Methodology
In 1 978 we initiated a 4-year study to
investigate various aspects of the ecology
of SAV communities in Chesapeake Bay.
While intensive research was conducted
at several locations along the estuarine
salinity gradient, our work focussed on
communities located in the low salinity
(5-15°/oo) region. The research considers
factors potentially responsible for the
observed decline m SAV distribution and
abundance. It was organized in a hier-
archical fashion with both mechanistic
and holistic experiments combined in a
sequence of systems and subsystems to
deal with the complexity of the ecosys-
tems studied in addressing these ques-
tions.
Herbicide concentrations were meas-
ured in the field to describe both long-
term mean levels and short-term re-
sponses to storm/runoff events Phyto-
toxic effects of these compounds on SAV
photosynthesis, growth and vegetative
reproduction were examined in various
experimental systems ranging m size from
1-500 I and m duration from hours to
months. Photosynthesis was estimated
as Os production, and )4C incorporation
and growth as increases in number and
biomass of shoots and other plant mate-
rial Degradation, sorption, and plant
uptake of atrazme were measured using
14C ring-labelled compounds.
Nutrient enrichment studies were done
using 500 m3 experimental ponds filled
and flushed with estuarine water and
planted with SAV from the Choptank
River estuary Plankton, epiphyte, and
SAV biomass were measured at 1 -4 wk
intervals throughout the growing season
Light attenuation due to epiphytes was
measured as reduction in transmittance
through clear acrylic slides covered with
various levels of algal growth. Nutrient
levels in water, sediments, and plant
material were also analyzed periodically
using standard techniques. Parallel exper-
iments were done m 75 I laboratory
microcosms having short (30 cm) water
columns to minimize the effect of phyto-
plankton.
Data from these and other experiments
were analyzed for inclusion in numerical
models to simulate ecosystem behavior.
These models were calibrated and verified
with separate data sets to produce models
that behaved consistently with nature
Model computations were done using
digital computers; experiments were per-
formed by changing one or more external
factors to simulate various spatial and
temporal conditions.
Results and Conclusions
Herbicides
Considerable effort was expended to
investigate the potential importance of
herbicides m contributing to the overall
stress of the estuary's SAV populations.
This research emphasized two specific
compounds. The first of these, atrazme,
which is closely associated with corn
crops, has been the most widely used
herbicide m the region, and the second
compound, linuron, iscommonlyemploy-
ed in weed control for soybeans. Concen-
trations of these two herbicides were
monitored in water and sediments
throughout the upper bay over the period
1980-81. A hierarchically designed strati-
fied sampling scheme revealed typical
aqueous concentrations of both com-
pounds to be about 0-3 ppb in the main
bay, 0-5 ppb in a major eastern shore
tributary, and 0-40 ppb in a creek con-
necting a small estuarine cove to sur-
rounding agricultural fields. Concentra-
tions in the creek and small cove were
measured at 1-4 h intervals before,
during, and after all runoff events, and
values above 5 ppb never persisted for
more than 6-8 h. Atrazine concentrations
associated with suspended or deposited
sediments were less than 5 ppb for > 95%
of samples and never exceeded 20 ppb
Initial studies indicated a wide range of
physiological and morphological respons-
es of one common SAV species, Pota-
mogeton perfo/iatus, in response to herbi-
cide treatment, including photosynthetic
depression, stem elongation, reduction in
stem weight per unit length, and in-
creased chlorophyll a per unit leaf area.
Several of these effects are analogous to
observed adaptations of this and related
species to reduced light intensity.
At atrazine or linuron concentrations
between 5-100 ppb, significant photosyn-
thetic inhibition was observed for both P.
perfo/iatus and Myriophyllum spicatum
in microcosms, followed by strong recov-
ery (toward untreated control plants)
within 1-3 wk, even though herbicide
levels remained within 5-10% of initial
values throughout Plant biomass de-
creased significantly after 5 wk of treat-
ment at herbicide concentrations > 50
ppb for P. perfoliatusand > 500 ppb for M.
spicatum. Overall, the effects of the two
herbicides were statistically identical,
while some differences between plant
-------
species were observed (M. spicatum being
more tolerant) Estimates of h 0(herbicide
concentration at which 1 % loss of photo-
synthesis [Pa] is predicted) were 2-4 ppb
for P perfol/atus and 8-1 1 ppb for M.
spicatum, and values of I50(concentration
for 50% loss of Pa) ranged from 45-55 ppb
and 80-117 ppb, respectively Similar
phytotoxicities were observed for Zanni-
chellia palustna a nd Ruppia mantima.
Rapid uptake of 14C-labelled atrazine
was demonstrated for P. perfoliatus, with
equilibrium between internal and external
concentrations being achieved within
about 1 h A direct relation between
atrazine uptake and photosynthetic de-
pression was observed for this plant,
however, disproportionately high appar-
ent uptake at low external herbicide
concentrationssuggesta two-step uptake
process with simple sorption (without
inhibition of photosynthesis) dominating
at low concentrations Root uptake of
atrazine appears to be of little importance
for these plants. Initial photosynthetic
recovery of atrazme-treated plants was
affected by release of sorbed herbicide
within 2 h after rinsing in atrazme-free
water Some loss of photosynthesis (~
5%) was evident after 3 d of wash,
however, this difference was not statisti-
cally significant Short-term (2 h) experi-
mental exposures to atrazine revealed
reductions in P. perfoliatus photosynthe-
sis which were similar to those observed
over 2-6 wk in microcosms, with values of
I50 being about 80 ppb.
Atrazine is readily sorbed to soil and
sediment particles, with a partition coef-
ficient (sorbed'aqueous) greater than 1.0
However, the potential importance for
plant uptake of atrazine sorbed to over-
lying sediments (resting on SAV leaves)
seems to be remote Experiments with
14C-labelled atrazine showed negligible
plant uptake of herbicide sorbed to soils at
concentrations of about 1 20 ppb In addi-
tion, the presence of epiphytic sediments
significantly retarded leaf uptake of
aqueous atrazine, although such sedi-
ments themselves inhibited photosynthe-
sis presumably by attenuation of light and
reduction of COa uptake
The degradation of 14C-labelled atrazine
was observed under simulated field condi-
tions for upper and middle bay sediment-
water systems and for two common
agricultural soils in the Maryland coastal
plain. The distributions of atrazine and
two categories of metabolites or degrada-
tion products (hydroxyatrazme and deal-
kylated atrazine) were followed over an
80 d period The half-life (time for 50%
degradation to metabolites) for atrazine
was markedly shorter for estuarme sys-
tems (1 5-20 d) than for soils (330-385 d).
The accumulation of hydroxyatrazme in
experimental estuarme water and sedi-
ments raised questions concerning the
potential phytotoxicity of these com-
pounds Bioassay experiments were per-
formed with 4 species of SAV to examine
uptake and photosynthetic depression for
14C-labelled atrazine and 3 metabolites.
Overall, the inhibitory effect of the metab-
olite, hydroxyatrazine, on plant photosyn-
thesis was negligible compared to that for
atrazine, with no significant inhibition
even at 1500 ppb. Some significant loss
of Pa was observed for deethylated atra-
zine at 500 ppb ; however, this metabolite
has a short half-life in the estuary, being
similar to that for atrazine.
Nutrients, Sediments and Light
The effects of nutrient enrichment on
algal (planktonic and epiphytic) growth
and SAV production and abundance were
investigated by fertilizing 8 (duplicates at
4 levels) experimental ponds (500 m3)
during June-August 1981. These ponds,
which were seeded with sediment, water
and plants from the Choptank River
estuary, were maintained m batch mode
for sequential periods of 7-10 d punctu-
ated by complete exchange of water
followed by retreatment prior to the next
batch period Maximum fertilization rates
were typical of nutrient loading in areas
of upper Chesapeake Bay receiving direct
agricultural runoff. Nutrient concentra-
tions m treated ponds were reduced
rapidly to control levels within 1 -3 d, and
plant tissue nutrient contents were direct-
ly related to treatment. Initial growth of
the two dominant SAV species (P. perfoli-
atus and R. mantima} was enhanced m
fertilized ponds, however, plant abun-
dance in August was inversely related to
treatment, with SAV virtually eliminated
at the highest dosage
Planktonic and epiphytic algal biomass
(as chlorophyll a) increased significantly
with treatment Light (PAR) attenuation
by microalgae was sufficient to account
for the reduction in SAV production and
abundance m August Epiphytic growth
accounted for most of the light reduction,
although attenuation in the water column
was also necessary to reduce PAR below
plant compensation levels Field observa-
tions indicated that inorganic sediments
could comprise as much as 80% of the
total mass of material accumulated on
SAV leaves, but these inorganic particu-
lates appear to be directly associated with
growth of epiphytic organisms. Direct
measurements of epiphyte effects on both
PAR attenuation (by leaf scrapings m
petri dishes) and plant photosynthesis
(with 14C-labelled bicarbonate)confirmed
this relationship.
A second year (1982) of fertilization in
the experimental ponds provided a more
detailed examination of the nutrient-algal-
SAV relationships Problems encountered
in the batch-mode approach in 1981
were alleviated with a continuous flow
system and more frequent treatment In
this 1 982 study only 4 ponds were used,
and SAV communities in these were
essentially mono-specific stands of P.
perfoliatus, thus eliminating the compli-
cating problems of differential epiphytic
colonization on 2 SAV species The
general patterns observed m 1 982 were
more pronounced and consistent than in
the 1981 research. Preliminary evidence
suggests that a shortening of SAV grow-
ing season, as observed here m response
to fertilization, may ultimately lead to
decimation of these plant populations by
disrupting plant reproduction Light atten-
uation by microalgae and suspendable
sediments may affect the normal balance
between SAVproduction and respiration,
leading to premature flowering and/or
msufficienttranslocation to underground
propagules, both of which would reduce
the viability of regrowth m the following
spring. It is concluded that further re-
search is needed to understand the
reproductive capacities and strategies for
these plants
In nature, due to sediment resuspen-
sion by tides and storms, turbidity levels
can increase rapidly by factors of 3 and
10, respectively. Therefore, detailed
studies of the responses and adaptations
of P perfoliatus to direct treatments of
various light levels (high 100%, medium
34%, low 6%) were also done in 1982
Numerous morphological and physiologi-
cal changes m this plant were observed m
response to reduced (moderate and low)
light, including stem elongation, in-
creased pigmentation, increased specific
leaf area, as well as increased initial
slope of photosynthesis versus irradiance
relations Most of these adjustments
appear to confer adaptive advantage on
shaded plants, however, after 2 wk of
exposure to low light, significant reduc-
tions in stem density, flowering, and
underground reproductive propagules
were observed
Conclusions
The relative contributions of herbicide
runoff, sediment loading and nutrient
enrichment to the environmental stress
-------
experienced by SAV in upper Chesapeake
Bay were considered with integrative
approaches. Combining these research
findings in a conceptual framework as
well as a numerical simulation model
suggested that the relative importance of
effects on SAV associated with these 3
inputs is as follows' nutrients > sedi-
ments » herbicides Historical data
suggest that SAV declines over the past
several decades occurred earliest in the
upper bay and tributary rivers and pro-
gressed downstream towards the main
bay Such a pattern is consistent with
factors found to stress SAV, as it is in the
upper reaches of these estuaries, where
water quality declined earliest and waste
loadings are most intense. In terms of
SAV rehabilitation, it appears that reduc-
tions in nutrient loading would be the
most advisable strategy, since excessive
nutrients severely stress these communi-
ties and because nutrient control meas-
ures are available and effective. It may
also be advisable to consider transplant-
ing programs to accelerate recovery in
areas where water quality is adequate,
but such activities should be coordinated
with efforts to better understand the
reproductive biology of SAV to increase
the probability of success.
W. Michael Kemp, J. Court Sevenson, Robert R. Twilley, and Thomas W. Jones are
with Horn Point Environmental Laboratories. Cambridge, MD 21613; Walter R.
Boynton and Jay C. Means are with Chesapeake Biological Laboratory,
Solomons. MD 20688.
David Flemer is the EPA Project Officer (see below).
The complete report, entitled "Submerged Aquatic Vegetation in Upper
Chesapeake Bay: Studies Related to Possible Causes of the Recent Decline in
Abundance," (Order No. PB 84-140 292; Cost: $26.50, subject to change) will
be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Chesapeake Bay Program
U.S. Environmental Protection Agency
2083 West Street. Suite 5G
Annapolis. MD21401
irUS GOVERNMENT PRINTING OFFICE 1984-759-015/7298
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
------- |