Groundwater Discharge to Long
Island Marshes:
Theory, Inferences, and Data
David
J. Tonjes
Elyse
O’Brien
Kim
Somers
Cashin
Associates, PC
Hauppauge,
NY
Groundwater discharges in an unconfined aquifer in locations
where the head elevation for the aquifer is greater than its corresponding
sediment surface elevation (Freeze and Cherry, 1979). On Long Island,
where the Upper Glacial aquifer is intersected by the ground surface, streams
and ponds occur.
USGS modeling suggests 28 percent of annual recharge
discharges to stream systems in Suffolk County (Buxton and Smolensky,
1999). This discharge to streams impacts the water table, causing lower
head pressures in the near vicinity of streams. The measurable effect,
especially for smaller streams, is often extremely local. At Connetquot
Brook, the difference in heads was detectable only 30 vertical feet below the
creek and approximately the same distance from each bank (Prince et al.,
1989). However, other modeling has suggested that even modest streams
can drain large portions of individual watersheds (the freshwater portion of
Meetinghouse Creek collected 25 percent of the recharge of that area)
(Schubert, 1999). Research on coastal plain streams fed by groundwater
indicates that the upper stretches are often receiving recently recharged
groundwater (which thus is from the immediate vicinity of the stream).
This changes for downstream reaches, where discharge from the banks or
stream bottom close to the banks may have been recharged locally, but
discharge into the central portions of the stream bottom often have long
aquifer residence time (and thus may be from areas of the watershed that are
not particularly close to the stream) (Modica et al., 1998).
At the shoreline, the elevation of the aquifer is greater
than the surface of the sediment. Groundwater discharges through the
saltwater interface. USGS modeling suggests that 43 percent of recharge
discharges at or near the shoreline in Suffolk County (Buxton and Smolensky,
1999). This phenomenon has begun to receive attention.
Bokuniewicz (1980) was one of the first to quantify aquifer
discharge to the nearshore environment. His studies suggested that most
of this submarine discharge from the Upper Glacial aquifer occurred within a
hundred feet or so of the shore. Follow-up work by others, especially
Paulsen, has shown that the discharge rates are highly variable. They
are a function of tidal cycles and sediment characteristics. Generally,
higher tides impede discharge, and low tides allow for greater discharge
rates, as would be expected. This pulsing of flow creates a mixing zone
between the fresh aquifer and the saline marine waters in the sediments
(Paulsen et al., 2001). It has been noted that the mixing of the waters
often results in altered characteristics of the discharging groundwater
compared to the nature of the groundwater measured just onshore (Sanudo ref).
Salt marshes are found on the periphery of the shoreline
(Chapman, 1960). It is not clear whether they should be classified as
“on-shore” or “off-shore” in terms of aquifer discharge. The surface of
the marsh tends to lie above mean sea level (Teal and Teal, 1969). The
marsh surface can be incised by natural marsh creeks or man-made mosquito
ditches. The bottom of natural creeks may or may not lie above mean low
water, and so some creeks retain salt water in them throughout the tidal
cycle, and some drain completely (Pomeroy and Imberger, 1981). Most
mosquito ditches were designed to drain through tidal cycles, meaning the
elevation of their bottoms is above mean low water (Richard, 1938), but this
is not necessarily the case on the south shore of Long Island, where
micro-tidal ranges mean the typical three foot depth of mosquito ditches can
leave their bottoms well below mean low water.
The marsh sediments are often saturated with saline
groundwater, as a result of flooding tides. The salty groundwater lies
above the fresh groundwater aquifer (Pomeroy and Imberger, 1981). The
elevation of the freshwater aquifer has not been published for any salt marsh
on Long Island known to us, but presumably is near to mean sea level.
Competing Theories
There are several theories regarding fresh groundwater
discharge in salt marshes. One, presented by Howarth and Teal (1980),
showed that fresh groundwater discharges occur at the base of the marshes in
the bottoms of marsh creeks. The salinity of the salt groundwater
system is controlled in these marshes by incidents of tidal flooding and dilution
by rain (Teal, 1986). Evaporation may affect summer salinities, leading
to elevated salinities in high marsh areas that are not flooded each tidal
cycle. Teal’s work has been primarily conducted in what are defined as
New England salt marshes – the kind of salt marshes found from Maine through
Long Island, whose histories were affected by glaciation and subsequent sea
level rise.
Pennings
and Bertness (1999) described salinity in the marsh soils and aquifer in New
England-type marshes as decreasing with distance from the seaward edge of the
marsh. This relates to distance from the salt water source, so that
fewer inundations by tidal waters means that rainwater constitutes
proportionally more of the perched salt marsh aquifer. Alternately, the
source of fresh water in the upland edges of the marsh could be groundwater
discharges. Harvey and Odum (1990), working in a fringing marsh in
Maryland with a “hillslope” aquifer (the hills were six to 20 m. tall), found
that maximal discharge into the wetlands peats was at the upland fringe, and
decreased with distance towards the open estuary. Overall, the pore
water flows in the marsh were dominated by tidal flows, meaning that
groundwater had long residence time in the marsh peats and thoroughly mixed
with saline waters prior to discharge through the marsh. The marsh
peats, because the base of them is located lower (in relation to mean sea
level) than the head of the freshwater aquifer, especially close to the toe
of the hill slope, receive discharges from the aquifer.
In
southern marshes, where evapo-transpiration rates are much greater, the
saltwater aquifers away from the estuary can have elevated salinities above
those near the estuary. The salinity of the creek waters is usually the
same as the water found in the bankside levees, but the marsh water table
water is usually higher in salinity, according to work done in Georgia by
Pomeroy and Imberger (1981). They suggested this showed natural creeks
drain little water from the marsh, resulting in a consistent-head, perched
water table. Hemond and Fifield (1982) also thought that seepage in the
marsh peat is negligible except near creeks, and theorized that
evapo-transpiration is the primary means for removing water from marsh peat
away from creeks. Then, due to the loss of head, groundwater inflows
would ensue to maintain the perched water table. Nuttle and Harvey
(1995) expanded this argument by constructing a water balance for a marsh
controlled by these kinds of flows. Using an assumption no loss of
water to the creek from the interior of the marsh, they determined that
groundwater upflow volumes were twice as great as tidal inflow volumes, for
an irregularly flooded high marsh, because of large evapo-transpiration
losses.
However, it is not clear that all peats do not transmit
groundwater to creeks. A model by Harvey et al. (1987) found that, if
the head in the marsh peat layers was great enough, horizontal flows to the
creek bank occurred as the tide retreated off the marsh surface. The
water balance indicated that two-thirds of the water infiltrating the marsh
surface during any particular tide will drain out of the marsh during that
same tidal cycle (note the study was made in a shallow, 20 m. wide, S.
alterniflora marsh that was completely flooded each tidal cycle).
Frey and Basan (1985) noted that the greater the height of the tide on the
marsh surface, the more infiltration into the sediments would occur, due to
greater head. And, generally, infiltration during a high tide is matched
by discharge from the sides of tidal creeks during the following low tide;
furthermore, it was found that the amount of water infiltrating into the
marsh surface decreased with distance from the marsh creek (Burke et al.,
1980), probably relating to reductions in inundation depths. These
studies focused on regularly flooded low marshes. Williams et al.
(1994), while focusing on high marshes, also suggested that water tables were
more variable than consistent. The amount of variation in the water table
height would depend on the frequency and duration of flooding, marsh
elevation, proximity to and the number of creeks, depressions, and pannes,
and the underlying sediment type.
Thus, in different settings, different forces may be at
work, meaning that it is probably not possible to define one general theory
regarding groundwater discharge in or near marshes.
Implications from Other Work
Bertness
et al. (2202) found that increases in nitrogen concentrations, measured in
plants, correlate with destabilized marsh vegetative regimes, especially
resulting in Phragmites australis (Phragmites)
expansions. These “excessive” nitrogen concentrations further
correlated with the degree of development measured on the upland border of
the studied marshes. The implication is that the development is
delivering the nitrogen to the marsh. Generally, on a coastal plain as
was the case here, the nitrogen impacts from local development are found in
groundwater. Although not explicitly stated by Bertness et al., it seems
to be understood that the local groundwater flow is the source of nitrogen
additions to the marsh.
Valiela
et al. (1978) determined that groundwater was an important source of nitrogen
to the total nitrogen budget for a Cape Cod marsh. The marsh had springs
at its upland reaches; however, the finding depended on groundwater inputs to
the marsh through the marsh creek bottoms. This was estimated by
comparing incoming tidal salinities with the least salinities measured at ebb
flow from the marsh, and determining how much inflow would have been required
to dilute the inflow to this level. This seems to greatly overestimate
the groundwater contribution, as it is not clear groundwater diluted the
entire inflow. Nor is it clear that groundwater inflow occurs as
rapidly at high tides (when the marsh surface is flooded, allowing for access
to any dissolved nutrients) as the estuarine head is greater, and so is
likely to restrict inflows from groundwater.
Wertheim National Wildlife Refuge Data
Particular Long Island marshes do and do not fit the
particulars discussed above. Phragmites invasions on Long Island
began on the East End at the turn of the last century (Lamont, 1997), and
those marshes were not especially impacted by development (at least, not on
the scale seen today). Marshes in parts of the Peconic Bay system and
along the North Shore do have hilly uplands, and are therefore likely to have
steeper groundwater tables in their immediate vicinity. The steeper
slope to the water table suggests a greater chance that the underlying marsh
peat will intercept the water table. The South Shore of Long Island,
which generally has a microtidal regime, tends to host marshes with larger
high marsh expanses. In such a setting, it may be possible to determine
if variations in water table salinity are due to inundations,
evapo-transpiration, or rainfall.
Cashin Associates, as part of a larger monitoring effort at
Wertheim National Wildlife Refuge has analyzed pore water salinity data from
a variety of high marsh sampling points. These data were compared to
precipitation and tide records.
Data analysis to be presented at the conference, and
inserted into the abstract.
References
Bertness, MD, PJ Ewanchuk, and BR Silliman.
2002. Anthropogenic modification of New England salt marsh
landscapes. Proceedings of the National Academy of Sciences
99(3):1395-1398.
Bokuniewicz, H. 1980. Groundwater
seepage into Great South Bay, New York. Estuarine and Coastal Marine
Science 10:437-444.
Burke, R., H. Hemond, and K. Stolzenbach.
1980. An infiltrometer to measure seepage in salt marsh soils.
pp. 413-423. In: Hamilton, P., and KB MacDonald (eds). Estuarine
and Wetland Processes with Emphasis on Modeling. Plenum Press, New
York, NY. 653 pp.
Buxton, HT, and DA Smolensky. 1999. Simulation
of the Effects of Development of the Ground-water Flow System of Long Island,
New York. USGS Water-Resources Investigations Report 98-4069.
US Geological Survey, Coram, NY. 57 pp.
Chapman, VJ. 1960. Salt Marshes and
Salt Deserts of the World. L Hill, London, UK. 392 pp.
Freeze, RA, and JA Cherry. 1979. Groundwater.
Prentice-Hall, Englewood, NJ.
Frey, RW, and PB Basan. 1985. Coastal
salt marshes. pp. 225-301. In: Davis, RA (ed.). Coastal
Sedimentary Environments. 2nd Ed. Springer-Verlag,
New York, NY. 716 pp.
Harvey, JW, and WE Odum. 1990. The
influence of tidal marshes on upland groundwater discharge to
estuaries. Biogeochemistry 10:217-236.
Hemond, HF, and JL Fifield. 1982.
Subsurface flow in salt marsh peat: a model and field study. Limnology
and Oceanography 27(1):126-136.
Howarth, RW, and JM Teal. 1980. Energy
flow in a salt marsh ecosystem: the role of reduced inorganic sulfur
compounds. The American Naturalist 116(6):862-872.
Lamont, E. 1997. Early collections of
Phragmites from Long Island, New York. Long Island Botanical Society
Newsletter 7(2):9-10.
Modica, E., HT Buxton, and LN Plummer.
1998. Evaluating the source and residence time of groundwater seepage
to streams, New Jersey coastal plain. Water Resources Research
34(11):2797-2810.
Nuttle, WK, and JW Harvey. 1995. Fluxes
of water and solute in a coastal wetland sediment. 1. The contribution
of regional groundwater discharge. Journal of Hydrology
164:89-107.
Paulsen, RJ, CF Smith, D. O’Rourke, and T-F
Wong. 2001. Development and evaluation of an ultrasonic ground
water seepage meter. Ground Water 39(6):904-911.
Pennings, SC, and MD Bertness. 1999.
Using latitudinal variation to examine effects of climate on coastal salt
marsh patterns and process. Wetlands Biogeochemistry 3:100-111.
Pomeroy, LR, and J. Imberger. 1981. The
physical and chemical environment. pp. 21-36. In: Pomeroy, LR,
and RG Wiegert (eds.). The Ecology of a Salt Marsh.
Springer-Verlag, New York, NY. 271 pp.
Prince, KR, TE Reilly, and OL Franke.
1989. Analysis of the shallow groundwater flow system near Connetquot
Brook, Long Island, New York. Journal of Hydrology 107:223-250.
Richards, AG, Jr. 1938. Mosquitoes and
mosquito control on Long Island, New York, with particular reference to the
salt marsh problem. New York State Museum Bulletin
316:85-172.
Schubert, CE. 1999. Ground-water Flow
Paths and Traveltime to Three Small Embayments within the Peconic Estuary,
Eastern Suffolk County, New York. USGS Water-Resources
Investigations Report 98-4181. US Geological Survey, Coram, NY.
41 pp.
Teal, JM. 1986. The Ecology of
Regularly Flooded Salt Marshes of New England: A Community Profile.
Biological Report 85(7.4), US Fish and Wildlife Service, Washington,
DC. 61 pp.
Teal, J., and M. Teal. 1969. Life
and Death of the Salt Marsh. Little, Brown, Boston, MA. 278
pp.
Valiela, I., JM Teal, S, Volkman, D. Shafer, and EJ
Carpenter. 1978. Nutrient and particulate fluxes in a salt marsh
ecosystem: tidal exchanges and input by precipitation and groundwater. Limnology
and Oceanography 23(4):798-812.
Williams, TP, JM Bubb, and JN Lester.
1994. Metal accumulation within salt marshes: a review. Marine
Pollution Bulletin 38:277-290.
|