THE ONGOING SALT MARSH RESTORATION AT STONINGTON, CONNECTICUT
Jim Myers
INTRODUCTION
The Atlantic coast of the northeastern United States is dotted
with innumerable salt marshes, all of which are quite similar in structure and
species composition. Salinity, frequency of tidal inundation and nutrient
availability vary significantly across these salt marshes, leading to complex
patchworks of distinct vegetation zones. Each zone can be characterized by one
or two dominant species which have adapted to tolerate the localized conditions
(Niering and Warren 1980). Where a salt marsh meets open water, unbroken
stands of tall salt marsh cordgrass, Spartina alterniflora, dominate the area
between the mean daily low and high tide lines. Above mean daily high tide line
but below the mean monthly high tide line, Spartina patens, salt meadow hay,
intersperses with stands of spikegrass, Distichlis spicata. Tidewater from the
monthly high tides collects in shallow depressions in this area, providing areas
of open water above the mean daily high tide line. Evaporation of this water
creates extraordinarily saline conditions, often twice as high as that of the
open salt water. A stunted form of Spartina alterniflora is one of the few
plants that will grow in these hyper-saline pannes. Juncus gerardi, black rush,
dominates the areas that are only flooded by extremely high tides. At the
brackish edges of the marsh, Typha angustifolia and Phragmites australis,
narrow-leafed cattail and reedgrass, respectively, are the dominant species.
From the brief description of typical northeastern United States salt marsh
vegetation given above, it is evident that the spatial distribution of salt
marsh plants within a marsh depends heavily on the localized physical
conditions. Any disturbance or human intervention which alters these physical
conditions can significantly alter the spatial distribution of the vegetation.
Most salt marshes in this area have been subjected to a variety of human caused
disturbances (Broome et al. 1988). Nearly all of the salt marshes in this area
have been ditched for mosquito control, simplifying the structure of the marsh
and altering the ratio of open water to emergent plants. The topography of some
marshes has been altered by the deposition of material dredged from nearby
channels. Many others have had tidal flushing interrupted by diking for
agriculture or road building, shifting the balance between fresh water input
from the uplands and salt water carried by the tides.
BARN ISLAND MARSH #1:
Site History The five salt marshes at Barn Island Wildlife management area in
southeastern Connecticut are no exceptions to the general rule of human
disturbance in marshes. For over 300 years, salt hay farmers have been removing
Spartina patens from these marshes every 3 or 4 years by mowing (Peck et al
1994). Although never dredged for navigation, the topography of these marshes
was altered by the mosquito control ditching that occured in the 1930's. The
peat removed during ditching was usually deposited next to the ditch, creating a
levee with a surface higher than the interior marsh. This impedes water flow to
the interior marsh and can create hypersaline pannes. Although this phenomenon
is not explictly mentioned in the literature, it was a common ditching practice
and likely to have been a factor in the degradation of the marshes. Mosquito
control ditching also reduced available wildlife habitat by removing open water.
In response, 4 of the 5 marshes were impounded in the late 1940's in an attempt
to increase open water (Barret and Niering 1993). Although this impoundment was
successful in increasing open water, it also deprived the marshes of tidal
flooding, greatly altering salinity throughout the marsh and changing the
composition of the plant community as well. Use of the impoundments by
waterfowl peaked in 1960 and then declined rapidly as the composition of the
plant community continued to change. Barn Island Marsh #1 is the largest (20
ha) and easternmost impounded marsh at the refuge. Like all of the other marshes
at Barn Island, it lies in a north-south valley that constrains the possibility
of marsh expansion. The watershed above this valley is the source of
freshwater input. The marsh is bordered on the north by an elevated dirt track
that completely eliminates tidal flushing on its north side. The impoundment,
built in 1946, lies 1 km south of the dirt track. A relatively intact salt marsh
exists south of the impoundment, buffering the it against the direct influence
of Little Narragansett Bay and Long Island Sound (Sinicrope et al. 1990). The
vegetation studies and restoration efforts involve the area between the
impoundment and the dirt track. Several studies have chronicled the vegetation
present at various times before and after impoundment. Taken together, they
provide a reasonable picture of the vegetation changes wrought by the cessation
of tidal flushing. Miller and Egler (1950) compiled a map of the
pre-impoundment vegetation and found it to have the typical salt marsh
vegetation composition and patterns outlined above. They also found that 1 year
after impoundment Spartina spp. still dominated the marsh, but salinities had
already begun to drop. In 1970, the Connecticut Department of Environmental
Protection evidenced concern about the prevalence of Phragmites australis in the
impoundment area (Sinicrope et al. 1990). In the mid-1970's, when the marsh was
at its freshest state, Hebard's (1976) vegetation map showed the site to be a
Typha angustifolia wetland, with significant amounts of Phragmites australis in
the more brackish areas.
RESTORATION EFFORTS
Since it became a state wildlife refuge in the 1940's, the
entire Barn Island area has been managed to maximize waterfowl hunting and other
recreational opportunities. When use of the impoundments by waterfowl declined
seriously after 1960, it became obvious that a change in management schemes was
necessary in order for the refuge to meet its management goals. The first
attempts at what might now be termed restoration were narrowly focused on
reducing or eliminating Phragmites australis. From 1965 to 1971, various
herbicides were used to control the plant, but none achieved much lasting
success (Barrett and Neiring 1993). In 1970, an old culvert in the center of
impoundment, located at the old tidal stream channel, was retrofitted with a
flapper gate to allow tidal inflow (Sinicrope et al. 1990). The DEP hoped that
this would increase the salinity enough to control Phragmites australis.
Unfortunately, the gate did not allow in enough salt water to change the
salinity appreciably; consequently, the marsh remained dominated by Typha
angustifolia and Phragmites australis. By the late 1970's it became obvious
that such small efforts concentrated on the control of a single species would
not be effective in reestablishing conditions favorable to waterfowl. The DEP
decided to restore the functional integrity between the marsh and Little
Narragansett Bay (Sinicrope et al. 1990). This, of course, meant the
restoration of significant tidal flushing. In order to reach this goal, a 1.5 m
diameter culvert was installed in the impoundment in 1978. The flapper gate
installed on the original 0.6 m diameter culvert in 1970 was removed in 1979,
slightly increasing tidal exchange volume. In response to an act by the
Connecticut State Legislature mandating restoration of the impounded marshes at
Barn Island, a 2.1 m diameter culvert was installed in 1982. This addition
brought the combined cross-sectional area of the culverts to 20% of the tidal
creek's original cross section.
EVALUATION
Despite the discrepancy between the current and original
cross-sectional area of inflow, several studies indicate that the restoration
attempts have been at least partially successful in restoring salt marsh
vegetation to the site. Two major studies focusing on changes in vegetation
since the restoration of tidal influx have been undertaken. Sinicrope et al.
(1990) compared the distribution of vegetation present in the marsh in 1988 to
the vegetation distribution in 1976 as mapped by Hebard (1976). After surveying
the vegetation by following Hebard's methods as closely as possible, they found
dramatic changes in the composition and distribution of species throughout the
area of the marsh where tidal flushing had been restored. The area covered by
Typha angustifolia declined from 74% to 16% of the marsh, and the remaining
stands were mostly stunted. Spartina alterniflora increased its coverage from
<1% to 45% of the marsh. Coverage of the marsh by high marsh plants approached
20%. Ironically, the one species that the DEP set out to control, Phragmites
australis, increased in coverage from 6% to 17%. However, most the stands of
Phragmites australis evidenced stunted growth, and the authors believed that the
plant would not continue to increase in coverage. Other freshwater and brackish
plants declined significantly. These findings led the authors to conclude that
the marsh had been restored to a significant degree. A perhaps more useful
evaluation of the success of the project was undertaken by Barrett and Niering
(1993). Using a geographical information system, they compared preimpounded,
impounded and postrestoration vegetation maps to determine the number and
direction of trends that vegetation changes have taken. The use of
preimpoundment vegetation maps allowed them to evaluate the success of the
restoration in the strictest sense. They determined that only 28% of the marsh
had been restored to its preimpoundment vegetation. However, if the definition
of restoration was broadened to accommodate the establishment of salt marsh
vegetation regardless of pre-impoundment spatial position, then 63% of the marsh
can be considered restored. Typha angustifolia and Phragmites australis
accounted for 37% of the marsh. Of course, plants are not the only components
of the salt marsh ecosystem. In fact, the genesis of this restoration project
lay in regaining habitat for important waterfowl. Aside from casual
observations (Sinicrope et al. 1990), no determination of waterfowl usage at the
restored site has been performed. Peck et al. (1994) have examined the marsh
for evidence of macroinvertebrates, however. They compared the populations of
two macroinvertebrates, the high marsh snail (Melampus bidentatus) and the
ribbed mussel (Guekensia demissa), in the restored marsh with the one
unimpounded marsh at Barn Island. Although there were differences in the
density and biomass of these invertebrates between the two marshes, they were
small enough to allow the authors to conclude that the marsh was in an advanced
stage of restoration.
CRITIQUE
Judging by the evidence provided by the three evaluations of the Barn
Island salt marsh restoration, the efforts seemed to have achieved a signficant
amount of success in a relatively short period of time. Although by no means
complete, the restoration has proceeded significantly towards establishing the
original salt marsh community. Several factors account for this rapid
restoration. Tidal flushing is obviously the single-most important factor in
the establishment of a salt marsh. By restoring tidal flushing to the marsh,
salinities were increased to the point at which Typha angustifolia and
Phragmites australis began to suffer. This allowed establishment of Spartina
alterniflora and other salt marsh plants. The relatively intact salt marsh on
the south side of the impoundment proved to be a handy propagule source.
Finally, although eliminating the tidal flushing had a dramatic impact on the
plant community, it apparently did not impact the system in such a way as to
seriously decrease the ability of the salt marsh plants to recolonize. Had the
original community been lost through a more serious disturbance, the restoration
probably would not have proceeded as easily or as successfully. Conversely,
had the original DEP plan for Phragmites australis control been executed in a
more effective manner that allowed for greater tidal inflow, it could have had
consequences reaching far beyond the control of one problem plant by giving
later restoration efforts the benefit of a head start on natural recolonization.
Restoration can encompass a variety of outcomes, and which of those outcomes
are the actual goals of a project are not always made clear. In the case of the
Barn Island marsh, restoration began as an attempt to control Phragmites
australis and gradually evolved into a full scale marsh restoration. Even then
the exact goals of the restoration were not quite explicit. Given the variety
of goals possible in restoration, it seems to make sense to adopt a evaluation
procedure that measures success in several different ways, as did Barrett and
Niering (1993). Of course, scientifically gathered predisturbance data is not
always available, in which case the method chosen by Sinicrope et al. (1990),
comparing pre- and postrestoration conditions, must be adequate. Peck et al.
(1994) show that evaluators still have options when even prerestoration data is
unavailable. They resolve this problem by comparing the restored system with an
appropriately similar system. Although the Barn Island marsh restoration
effort seems to have engendered considerable success, the impulse to put another
notch on the restoration success belt and move on should be held in check. The
restoration was by no means a complete success; 37% of the marsh was still
covered by Typha angustifolia and Phragmites australis as of 1988. Sinicrope et
al (1990) expressed optimism that re-establishment of the salt marsh species
would continue unimpeded. Barrett and Niering (1993), on the other hand, were
concerned that low rates of peat accumulation during the freshwater years will
hamper the ability of the system to match peat accumulation rates with relative
sea level rise, something natural marshes appear to be doing. The eastern
United States has been experiencing relative sea level rise for several thousand
years, with little apparent loss of salt marshes. Global warming has the
potential to greatly accelerate this trend. If the surface of the restored marsh
falls relative to the water level, many of the restoration gains of the past
decade and a half could be lost. This reason alone is enough to suggest
judgment be reserved on the absolute success of the project. If the restored
marsh can compensate for relative sea level rise as well as relatively intact
marshes, then perhaps pronouncements of success can be less qualified.
REFERENCES
Barrett, N.E. and W.A. Niering. 1993. Tidal marsh restoration: trends in
vegetation change using a geographical information system (GIS). Restoration
Ecology 1(1): 18-28.
Broome, S.W., E.D. Seneca and W.W. Woodhouse, Jr. 1988. Tidal salt marsh
restoration. Aquatic Botany 32: 1-22.
Hebard, G. 1976. Vegetation pattern and changes in the impounded salt marshes
of the Barn Island Wildlife Management Area. M.S. thesis. Connecticut
College, New London, Connecticut.
Miller, W.R. and F.E. Egler. 1950. Vegetation of the Wequetequock-Pawcatuck
tidal marshes, Connecticut. Ecological Monographs 20: 143-172.
Niering, W.A. and R.S. Warren. 1980. Vegetation patterns and processes in New
England salt marshes. Bioscience 30(5): 301-307.
Peck, M.A., P.E. Fell, E.A. Allen, J.A Gieg, C.R. Guthke and M.D. Newkirk.
1994. Evaluation of tidal marsh restoration: comparison of selected
macroinvertebrates populations on a restored impounded valley marsh and
unimpounded valley marsh within the same salt marsh system in Connecticut, USA.
Environmental Management 18(2): 283-293.
Sinicrope, T.L., P.G. Hine, R.S. Warren and W.A. Niering. 1990. Restoration of
an impounded marsh in New England. Estuaries 13(1): 25-30.
Return to the
Restoration & Reclamation Review Home Page