Posted on April 3, 2001

Nitrogen and Phosphorus Limitation in Mill and Green Ponds and the Effects of Nutrient Enrichment

Joel Creswell^1,5, Rebecca Karasack^2,5, Rebecca Johnson^3,5, and Hannah Shayler^4,5

Key Words: nutrient loading, nutrient enrichment study, nutrient limitation, anthropogenic nitrogen loading, phytoplankton dynamics, coastal ponds, Mill Pond, Green Pond

ABSTRACT

Nutrient limitations in two coastal ponds in Falmouth, MA were examined by adding NH₄NO₃ and KH₂PO₄ to water samples taken along a salinity gradient from freshwater to 26 ppt.  Samples were monitored over an eight day incubation at a constant temperature with abundant light.  Chlorophyll, particulate organic carbon, and particulate organic nitrogen concentrations were highest in samples fertilized with a mixture of nitrogen and phosphorus.  Changes in nutrient, chlorophyll, and particulate carbon and nitrogen levels over time suggest that the freshwater sample sites are nitrogen limited, while most of the brackish water sites showed phosphorus limitation.  Both ponds are highly affected by anthropogenic nutrient loading.

INTRODUCTION

Phosphorus has been shown to be the most common limiting nutrient to primary producers in freshwater ecosystems, while nitrogen is most often limiting in marine systems (Tomasky et al., 1999, Vince & Valiela, 1973).  Freshwater phosphorus limitation is due to the fact that the majority of phosphorus inputs to aquatic ecosystems comes from the weathering of primary minerals in marine sediments, not in freshwater systems.  Some reasons why nitrogen is limiting in marine ecosystems is because molybdenum, phosphorus, or energy constraints can be limiting to nitrogen fixers, which makes for lower nitrogen fixation (Vitousek & Howarth, 1991).  There is also significantly higher denitrification in marine sediments than in freshwater sediments, as marine sediments are largely anoxic.

Because they are saltwater systems, Falmouth, MA’s south coast ponds may be threatened by anthropogenic nitrogen loading (Fig. 1).  Because of their shallow bathymetry and large perimeter, these and other coastal salt ponds can accumulate high concentrations of dissolved inorganic nutrients from terrestrial sources.  High anthropogenic nutrient loading to these ponds may cause them to become more eutrophic as it reduces their nitrogen limitation.

Figure 1:  Eutrophication levels and dissolved inorganic nitrogen concentrations in Great, Green, and Bournes Pond.
Source:  Howes et al., 1999

In recent years, the Town of Falmouth has initiated efforts to reduce eutrophication in the ponds.  All the mitigation efforts to date have focused on nitrogen reduction, because nitrogen is expected to be the limiting nutrient, although no study has been conducted on the nutrient limitation in Green Pond.  The Town has also discussed using the natural nitrogen removal capacity of Mill Pond, located north of Green Pond (Fig. 2), to treat water entering Green Pond.  Before devising any strategy to use Mill Pond to control eutrophication through nutrient removal, however, a study must be conducted to determine the limiting nutrients to primary producers in both systems.

Figure 2:  A map of Falmouth, MA showing the three large south coast ponds.  Green Pond is in the center, and Route 28 is represented by the red line crossing the north end of Green Pond.  Mill Pond is just north of Route 28 from Green Pond.
Source:  Horsley and Witten, 1999

Because the water samples in this study were taken along a salinity gradient, it is expected that nutrient limitation will shift gradually along the gradient from P limitation to N limitation, rather than switching abruptly when moving from Mill Pond to Green Pond.  This is consistent with the findings of Tomasky et al. (1999), who conducted a nutrient enrichment study in nearby Waquoit Bay.  They found that nitrogen limitation generally increased with increasing salinity, apart from brief periods of phosphorus limitation.  Algal biomass increased the most in nitrogen enriched saltwater samples and phosphorus enriched freshwater samples.   Oviatt et al. (1995) had similar findings in a mesocosm nutrient enrichment experiment in Narragansett Bay, Rhode Island.  Their study involved only saltwater environments, however, those to which nitrogen was added had a higher primary productivity, respiration, biomass, and net ecosystem production.

This study investigated the nutrient limitation in Mill Pond and Green Pond though nutrient addition to water samples.  By monitoring chlorophyll levels, particulate organic carbon and nitrogen, and dissolved inorganic nutrient levels over the course of an incubation, phytoplankton dynamics were used to determine nutrient limitation.

Nutrient limitation data will also be useful in determining the ponds’ nitrogen removal capacity.  This is because, if the phytoplankton in Mill Pond and Green Pond have ample nutrients available to them in the Redfield stoichiometry of 106C:16N:1P weight ratio (Redfield, 1934), they will become highly productive.  When this happens, they will transport large quantities of nutrients to the bottom as they die and settle out of the water column.  This process can facilitate the removal of nitrogen from the water, as it leaves nutrients in organic form at the water-sediment interface, where they can be remineralized, nitrified, and denitrified.  It has been shown that there is an average amount of denitrification occurring in the sediments of Mill Pond, when compared with other freshwater ponds (Johnson, R., 2000, unpublished data).  Thus it is feasible to assume that some of the nitrogen transported to the sediment-water interface by phytoplankton will be denitrified.

METHODS

Site Description

Mill Pond is a small, shallow freshwater pond in Falmouth, MA (Fig. 2) with an area of 50117 m² and a maximum depth of 1.5 m (Karasack, R., 2000, unpublished data).  It is fed by the Backus River, which enters at the north end of the pond and has a discharge of 6480 m³/day (Ramsey et al., 1999).  The Backus River passes through several cranberry bogs before entering Mill Pond, which makes the water level in the pond highly subject to the water needs of the cranberry farmers.  During the course of this study, the water level was raised 45 cm above its initial stage, due to the draining of a cranberry bog north of the pond.

Green Pond is south of Mill Pond (Fig. 2) and is much larger, 53 ha (Howes et al., 1999).  It receives most of its surface freshwater input from Mill Pond via a culvert which passes under Route 28, the highway that separates the ponds.  The long-term average freshwater input from this culvert is 269 m³/day (Ramsey et al., 1999).  The long-term average groundwater input to the pond is 348 m³/day.  Green Pond is brackish, ranging in salinity from 8.5 ppt at the north end, closest to the freshwater input, to 29.3 ppt at the south end, near the outlet to Vineyard Sound (Ramsey et al., 1999).  It is a true estuary, as it acts as the mixing zone for freshwater and seawater.  It is also shallow, with a maximum depth of roughly three meters, although the water level varies with the tide.

Green Pond ranges from mesotrophic to eutrophic in its trophic status (Ramsey et al., 1999) (Fig. 1).  The ponds’ primary source of nitrogen is the septic systems of nearby residents, which account for 65 percent of the total loading (Horsley and Witten, 2000).  The secondary source is lawn fertilizers, which account for 26 percent.

In addition to the present nitrogen inputs to the ponds, a nutrient plume from the nearby Massachusetts Military Reservation’s (MMR) wastewater treatment facility is migrating through the groundwater toward the ponds (Fig. 3).  If the plume reaches Green Pond, it is estimated that it will increase the total nitrogen loading to the pond from the upper watershed by 10 percent, and to the entire pond by 3 percent above the current conditions (Howes et al., 1999).

Figure 3:  The water table in the area surrounding the Massachusetts Military Reservation (MMR).  The MMR sewage treatment plant is at the top of the map in the center.  The three south coast ponds are in the rectangle at the bottom of the map in the center.
Source:  Howes et al., 1999

Both ponds are situated on the southern edge of the Mashpee Outwash Plain, a glacial deposit consisting mainly of highly permeable sand and gravel.  Green Pond was formed in a drowned valley in the outwash plain, Mill Pond is a shallow basin at its head.  Because of the high permeability of the soil, groundwater flow accounts for more than half of the freshwater input to Green Pond, and is likely to make a similar contribution to Mill Pond.

There is housing around the perimeter of both ponds, and all of the houses have septic systems.  Green Pond also experiences a considerable amount of boat traffic, particularly in the lower pond, where the Green Pond Marina is located.

The primary producer community in Mill Pond is dominated by yellow-green algae, although pinnate diatoms and blue-green algae are present in noticeable quantities.  Green Pond is dominated by flagellates, but also has large numbers of pinnate and centric diatoms, ciliates, and blue-green algae.

Sampling and Incubation Procedure

Initial samples were taken in both Mill and Green Ponds to determine nutrient concentrations, salinity, chlorophyll levels, and several other characteristics.  Based on those data, six sample sites were selected (Figs. 4 & 5)  Surface water only was sampled at four of the sites, and surface water and water from a depth of 0.5 m was taken at the fifth site, site GP2d.

Figure 4:  Sample sites on Mill and Green Ponds where water samples were taken for bottle incubations.

Figure 5:  Photos of sample sites in Mill Pond and Green Pond

A 20 L Nalgene Carboy was filled at each site with water that was filtered through a 153 µm mesh screen to remove phytoplankton grazers.  A Black & Decker Jackrabbit Pump was used to obtain water from a 0.5 m depth at site 24.

In the lab, 72 two liter, clear, acid washed soft drink bottles with the labels removed were filled with 1.5 L each of sample water, 12 bottles for water from each sample site.  The bottles with water from each site were then separated into four groups of three: a control group, a nitrogen enriched treatment, a phosphorus enriched treatment, and a nitrogen and phosphorus enriched treatment.  The nitrogen enriched  bottles received 3.75 ml of 20,000 µM ammonium nitrate (NH₄NO₃) to give them a final concentration of 50 µM of NH₄NO₃, or a 100 µM concentration of dissolved inorganic nitrogen.  The phosphorus enriched bottles received 3.75 ml of 40,000 µM potassium phosphate (KH₂PO₄) to give them a final concentration of 100 µM of KH₂PO₄.  The remaining water in the carboys was analyzed to provide ambient water column data.  The above concentrations were used in order to provide equal concentrations of biologically available N and P in the bottles.

After nutrient additions were made, the bottles were laid on their sides and placed on a slow-moving gyratory shaker apparatus in a Conviron Growth Chamber.  The chamber was set to provide the bottles with 14 hours of full light (measured to be roughly 2400 µEinstein) per day and to maintain a temperature of 10° C.  Actual water temperature in the bottles ranged from 9° C to 37° C, due to heat produced by the lights.  The bottles were rotated throughout the incubation, however, to ensure that the average temperature in all the bottles over the course of the incubation was the same.

The incubation ran for eight days, and roughly 300 ml of sample was taken from each bottle roughly every 48 hours.  This procedure maintained equal water levels in all the incubating bottles.

Water Sample Analysis

All samples were analyzed for chlorophyll a, ammonium, nitrate, phosphate, particulate organic carbon (POC), and particulate organic nitrogen (PON) concentrations.  Chlorophyll a was measured using the Lorenzen spectrophotometric method (Lorenzen, 1967).  Ammonium was analyzed on a spectrophotometer, using the Phenolhypochlorite Method (Solarzano, 1969).  Phosphate was also analyzed on a spectrophotometer (Murphy & Riley, 1962).  Nitrate concentrations were determined using a Lachat Flow Injection Analyzer following a modification of the copperized-cadmium reduction method of Wood et al. (1967).  POC and PON were measured using a Perkin-Elmer CHN 2400 Elemental Analyzer.

RESULTS

Nutrient levels measured in initial water samples taken before the bottle incubation began were significantly higher in Backus River water than in Mill Pond and Green Pond water (Fig. 6).  Ammonium (NH₄⁺) and phosphate (PO₄^3-) concentrations were highest in Backus River samples taken closest to the discharge into Mill Pond.  Phosphate levels were relatively high throughout the Backus River.  The nitrogen to phosphorus ratios (N:P) in the Backus River were lower than Redfield stoichiometry in all samples (Fig. 7), suggesting that the system is nitrogen-limited.  Nutrient concentrations in Mill Pond are roughly an order of magnitude less than those in the Backus River, but remain lower than Redfield stoichiometry.  Nitrate (NO₃^-) concentrations at the northernmost sample sites in Green Pond (1-3) were high relative to the other samples from Green Pond, and were significantly higher than the Redfield ratio.  Ammonium levels were low relative to the concentrations in the Backus River, but are significantly higher than the concentrations at most sample sites in Mill Pond.  Phosphate levels were low in all Green Pond samples.  N:P ratios in the Green Pond samples are higher than Redfield stoichiometry at all but two sites.

Figure 6:  Dissolved inorganic nutrient concentrations in the water column in the Backus River, Mill Pond, and Green Pond prior to the beginning of the nutrient enrichment experiment.

Figure 7:  Nitrogen to phosphorus (N:P) ratios of dissolved inorganic nutrients in the water column in the Backus River, Mill Pond, and Green Pond prior to the beginning of the nutrient enrichment experiment.

Water column seston levels, including phytoplankton and detritus, were quantified using chlorophyll a, particulate organic carbon (POC) and particulate organic nitrogen (PON) data.  Chlorophyll a levels are higher in Mill Pond and the Backus River than in Green Pond, in general (Fig. 8).  The highest chlorophyll concentrations are in the Backus River.  All three systems had a wide range of chlorophyll levels.  POC and PON varied widely in all three systems (Fig. 9), but each system’s C:N ratios were more consistent (Fig. 10).  The C:N ratios in the Backus river were all roughly 18, which is much higher than the Redfield ratio of 6.63.  In Mill and Green Ponds, the ratios were much closer to Redfield, ranging from approximately 6 to 9.

Figure 8:  Chlorophyll a concentrations in the water column in the Backus River, Mill Pond, and Green Pond, prior to the beginning of the nutrient enrichment experiment.

Figure 9:  Particulate Organic Carbon and Nitrogen (POC and PON) concentrations in the water column in the Backus River, Mill Pond, and Green Pond prior to the beginning of the nutrient enrichment experiment.

Figure 10:  Water column particulate C:N ratios in the Backus River, Mill Pond, and Green Pond prior to the beginning of the nutrient enrichment experiment.

Dissolved nutrient levels in the freshwater sample bottles used in the incubation generally declined quickly (Fig. 11).  Ammonium was taken up at the highest rate in N enriched bottles in both freshwater samples, although nitrate was drawn down relatively rapidly as well.  Phosphate declined slowly over the course of the incubation in both samples after a sharp drop in its concentration on day 2 and a subsequent return to a high level on day 4.

Figure 11:  Dissolved inorganic nutrient concentrations in the freshwater sample bottles along the time course of the incubation. The legend indicates control samples, samples enriched with nitrogen, samples enriched with phosphorus, and samples enriched with both.

In saltwater samples, there was also a rapid depletion of ammonium in N-enriched bottles (Figs. 12 & 13).  Ammonium levels rise sharply between days 4 and 6 in the P  and N+P enriched bottles from sites GP2d and GP3.  Nitrate levels declined quickly in all but the N enriched bottles in the saltwater samples.  It declined much more gradually in the N enriched bottles.  Phosphate in the saltwater samples showed a similar pattern to that shown in the freshwater samples: It dropped sharply in the P and N+P enriched bottles on day 2, then returned to a high level on day 4 and declined more slowly for the rest of the incubation.

Figure 12:  Dissolved inorganic nutrient concentrations in saltwater sample bottles along the time course of the incubation.

Figure 13:  Dissolved inorganic nutrient concentrations in saltwater sample bottles along the time course of the incubation.

The chlorophyll levels were highest in the N+P enriched bottles for every site but GP2d and GP3 (Fig. 14).  The chlorophyll peak in the N enriched bottles was higher than that in the N+P enriched bottles at sites GP2d and GP3.  When comparing the chlorophyll peaks in the experimental bottles with the peaks in the controls, the N+P enriched bottles had the highest peaks in every sample but GP2d, in which the N enriched bottle had the highest peak (Fig. 15).

Figure 14:  Chlorophyll a concentrations in each sample and treatment as they change over the time course of the incubation.

Figure 15:  Chlorophyll a peaks in each experimental treatment. These are calculated using the value of the highest peak in each treatment and subtracting from it the chlorophyll concentration in the control bottles at the same time point.

POC accumulated to the highest concentrations in N+P enriched bottles from all sites save GP1, in which the P enriched bottles reached the highest POC level (Fig. 16).  In the freshwater samples, the second highest POC levels were in the N enrichment bottles.  In the saltwater samples, the POC levels varied widely across treatments and over time.

Figure 16: Particulate Organic Carbon (POC) concentrations in each sample and treatment as they change over the time course of the incubation.

PON levels were highest in the N+P enriched bottles of every sample (Fig. 17).  The second highest PON levels were in the N enriched treatment for most bottles.  In sites GP1 and GP3, however, the PON levels in the P enriched treatments peaked higher than those in the N enriched treatments.

Figure 17:  Particulate Organic Nitrogen (PON) in each sample and treatment as they change over the time course of the incubation.

The average particulate C:N ratio across all sample sites was highest in the P enriched treatment at most time points (Fig. 18).  N+P enriched bottles showed the lowest average particulate C:N ratio throughout the incubation.  C:N ratios in every treatment but the N+P enriched treatment increased over the time course of the incubation.  The N:P ratio in every sample was highest throughout the incubation in the N enriched treatments (Fig. 19).  The lowest N:P ratio was in the P enriched treatments in every sample.

Figure 18:  Average C:N ratios for each treatment averaged across all sample sites over the time course of the incubation.

Figure 19:  N:P ratios for each sample and treatment as they change over the time course of the incubation.

DISCUSSION

The N:P ratios in the Backus River and Mill Pond indicate that both systems are N limited (Fig. 7), as they are lower than Redfield stoichiometry.  The N:P ratios in Green Pond suggest that it is P limited, as most of them are much higher than Redfield.  Rebecca Karasack (2000, unpublished data) also determined that Mill Pond has a low N:P ratio, 1.01, which suggests N limitation.  The POC and PON data from the incubation support the idea that Mill Pond and the Backus River are N limited.  The highest POC and PON concentrations in the Mill Pond samples are in the N+P and N enriched bottles (Figs. 16 & 17, samples MP1 and BR1).  This shows that phytoplankton are taking the most carbon and nitrogen into their biomass in N rich environments, further indicating that these sites are N limited.  The POC and PON data from the incubation show that Green Pond is P limited at three of the four sites; GP2d does not seem to be P limited (Figs. 16 & 17).  Site GP2d is the only Green Pond sample that was taken at depth, which may be an explanation as to why it displayed a different nutrient limitation: it came from underneath the halocline, and is therefore part of a different water mass than that from which the other samples were taken.  In the three P limited sites, POC and PON levels in the P enriched treatments peak earlier in the incubation and usually higher than in the N enriched treatments.  This indicates that the primary producers in the P enriched treatments accumulated more biomass than those in the N enriched treatments up until their population crashed.  These data are consistent with the fact that the N:P ratio is significantly higher than Redfield throughout most of Green Pond.

The chlorophyll data (Fig. 14) lend some support to my conclusions about the nutrient limitations of each system, however, many of them do not show clear enough trends to draw meaningful conclusions.  We expect the N+P treatment’s chlorophyll levels to be the highest in all samples, as the phytoplankton in those bottles are provided with ample nutrients to grow.  We look at the N enriched and P enriched treatments to help determine limitation.  At site MP1, the chlorophyll in the N enriched samples increased to a much higher level than it did in any other treatments, indicating an N limitation at that site.  In the other samples, however, there is not a clear enough difference in chlorophyll concentrations between treatments to determine limitation.  There are also few conclusions that can be drawn from the pre-incubation chlorophyll data, other than that it is highly variable by site (Fig. 8)

Chlorophyll peaks can also help to determine limitation.  They provide a convenient way to compare chlorophyll levels across treatments and sites.  Generally, chlorophyll within each treatment at each site increased to a peak, then declined again during the incubation, so measuring the peaks quantifies the maximum chlorophyll accumulation of each treatment at each site.  At sample sites MP1, GP2d, and GP1, the peaks support my hypotheses on which nutrient is limiting in each place (Fig. 15).  A high N enriched treatment peak indicates that that sample is N limited, and a high P enriched treatment peak indicates that P is limiting.  At other sites besides the three above, the peaks are too ambiguous to be used in determining limitation.

The C:N ratio data from the incubation show N depletion over the course of the incubation for every treatment but the N+P enrichment (Fig. 18).  The most nitrogen is depleted in the P enriched treatment.  The abundance of phosphorus leads phytoplankton to take up more carbon and nitrogen than they otherwise would, which causes them to exhaust most of their nitrogen supply and take up carbon and nitrogen in a ratio higher than that given by Redfield.  The particulate C:N ratios in the ponds before the incubation began also support the hypothesis that the Backus River and Mill Pond are nitrogen limited.  The ratios in both places are higher than the Redfield ratio (Fig. 10).  The ratios in Green Pond are generally closer to the Redfield ratio, suggesting that phytoplankton in that system are not depleting their supply of dissolved inorganic nitrogen.

The dissolved inorganic nutrient levels in the incubated samples from Mill Pond and the Backus River also suggest that these systems are N limited.  In both samples, ammonium declines rapidly and nitrate declines quickly initially, then gradually over the rest of the incubation in the N enriched treatments (Fig. 11).  There is a small increase in the nitrate in the N enriched treatment from sample BR1, but the concentration at the end of the incubation is still significantly lower than at the beginning.

There are few consistent patterns in the nutrient levels in the saltwater samples along the time course of the incubation, other than a dip in phosphate concentration on day 2 (Figs. 12 & 13), which is probably the result of error in the analysis procedure.  It is also apparent that phosphate concentrations decline gradually (excluding the drop on day 2) in the P enriched treatments of samples GP1, GP2s, and GP3, which are P limited, but that they stay roughly the same throughout the incubation in sample GP2d, which is not P limited.  It is expected that phosphate will decline more slowly than nitrate and ammonium in the incubated samples because phytoplankton only require 1/16 as much of it to grow, according to Redfield stoichiometry.

It is uncommon to find a freshwater system that is N limited and a salt or brackish water system that is P limited, as was previously discussed.  It seems especially odd that Mill Pond is nitrogen limited, as it receives large N inputs from the septic systems of the surrounding houses.  One explanation for the N limitation in Mill Pond is that the nutrient concentrations in the water entering from the cranberry bog change the relative nutrient availability in the pond.  This is due mainly to high levels of phosphate and low DIN levels in the water entering from the bog.  Prior to the drainage of the cranberry bog, the N:P ratio in Mill Pond was 23.80 (Karasack R., 2000, unpublished data).  After the drainage, the ratio was 1.01.  This relationship is reflected in the pre- and post-drainage dissolved inorganic nutrient levels (Fig. 20).

Figure 20:  DIN and phosphate levels in Mill Pond before and after the cranberry bog north of the pond was drained.
Source:  Rebecca Karasack, 2000, unpublished data

An explanation as to why Green Pond is P limited in places, rather than N limited throughout the entire pond, as we would expect, probably lies in the high anthropogenic nitrogen inputs to the system.  There are many more houses along the shore of Green Pond than Mill Pond, which provide the system with nitrogen from their septic systems and lawn fertilizers.  Green Pond also has a much lower freshwater input than Mill Pond.  The total freshwater input to Green Pond, including surface and groundwater, is 617 m³/day.  The surface freshwater input alone to Mill Pond is 6480 m³/day.  This means that there is much more dilution in Mill Pond than in Green Pond.  In addition, Green Pond does not have the same high phosphorus inputs that Mill Pond has, which further explains why it is phosphorus limited at most sites.  Most of Green Pond’s freshwater input is groundwater, which contains little phosphorus, as phosphorus is strongly immobilized in soils (Weiskel & Howes, 1992).  Green Pond’s overland freshwater input can also be cut off by blocking the culvert from Mill Pond, which happens at various times of the year.  This reduces the phosphorus input to the pond even further.

The nutrients found to be limiting in each pond in this study are probably not limiting in these systems at all times of the year.  Since the nitrogen limitation in Mill Pond and the Backus River seem to be attributable to the draining of the cranberry bog, it is likely that the system would become phosphorus limited during times when the bog is not being drained.  Green Pond may remain phosphorus limited in the upper pond year-round, as its nitrogen and freshwater inputs change little throughout the year.  The system is probably nitrogen limited in the lower pond, however where there is more mixing with ocean water from Vineyard Sound.  This hypothesis is supported by the N:P ratios in water samples taken from the furthest south points in the pond, below the bridge (Fig. 2).  The ratios at these points are lower than Redfield stoichiometry, which is common in seawater. 

ACKNOWLEDGEMENTS

I would like to thank Ken Foreman for mentoring this project.  I would also like to thank Kris Tholke, Marcus Gay, Sarah Morrisseau, Bart DeStasio, and Anne Giblin for their indispensable assistance.  Thanks also to The Ecosystems Center for the use of their growth chamber, to the Green Pond Marina for the use of their moorings, and to the Town of Falmouth for providing us with the GIS data set of the town. 

WORKS CITED

Horsley and Witten, Inc.  1999.  Ashumet Plume Nitrogen Offset Program: Identification of All Feasible Measures to Improve Water Quality – Initial Screening of Alternatives.  Report prepared for the Town of Falmouth, 59 Town Hall Square, Falmouth, MA 02540.

Horsley and Witten, Inc.  2000.  Ashumet Plume Nitrogen Offset Program Tasks 5&6 Report: Detailed Analysis of All Feasible Corrective Measures and Identification of Alternative Strategies That Can Be Accomplished Within a Capital Budget of $8.5 Million.  Report prepared for the Town of Falmouth, 59 Town Hall Square, Falmouth, MA 02540.

Howes, B. L., Li, F. and Horsley and Witten, Inc.  1999.  Ashumet Plume Nitrogen Offset Program: Evaluation of Nutrient Loadings to Great, Green, and Bournes Ponds, Falmouth, MA.  Center for Marine Science and Technology, New Bedford, Massachusetts, U.S.A.

Lorenzen, C.J.  1967.  Determination of chlorophyll and phaeo-pigments: spectrophotometric equations.  Limnology and Oceanography 12: 343-346.

Murphy, J., and J. P. Riley.  1962.  A modified single solution method for the determination of phosphate in natural waters.  Analytica Chimica Acta 27: 31-36.

Oviatt, C., Doering, P. Nowicki, B., Reed, R., Cole, J., and J. Frithsen.  1995.  An ecosystem level experiment on nutrient limitation in temperate coastal marine environments.  Marine Ecology Progress Series 116: 171-179.

Ramsey, J. S., Kelley, S. W., and B. L. Howes.  1999.  Water Quality Analysis of Great, Green, and Bournes Ponds, Falmouth MA.  Report prepared for the Town of Falmouth, MA.  Applied Coastal Research and Engineering, Inc., Mashpee, Massachusetts, U.S.A.

Redfield, A. C.  1934.  On the proportion of organic derivatives in sea water and their relation to the composition of plankton.  Pages 176-192 in James Johnston Memorial Volume.  University Press of Liverpool, Liverpool, U.K.

Solarzano, L.  1969.  Determination of Ammonia in Natural Waters by the Phenolhypochlorite Method.  Limnology and Oceanography 14: 799-801.

Tomasky, G., Barak, J., Valiela, I., Behr, P., Soucy, L., and K. Foreman.  1999.  Nutrient limitation of phytoplankton growth in Waquoit Bay, Massachusetts, USA: a nutrient enrichment study.  Aquatic Ecology 33: 147-155.

Vince, S. and I. Valiela.  1973.  The Effects of Ammonium and Phosphate Enrichments on Chlorophyll a, Pigment Ratio and Species Composition of Phytoplankton of Vineyard Sound.  Marine Biology 19: 69-73.

Vitousek, P. M. and R. W. Howarth.  1991.  Nitrogen limitation on land and in the sea: How can it occur?.  Biogeochemistry 13: 87-115.

Weiskel, P. K. and B. L. Howes.  1992.  Differential transport of sewage-derived nitrogen and phosphorus through a coastal watershed.  Environmental Science and Technology 26: 352-360.

Wood, E. D., Armstrong, F. A. J., and F. A. Richards.  1967.  Determination of Nitrate in Seawater by Cd-Copper Reduction to Nitrite.  Journal of the Marine Biological Association of the United Kingdom 47: 23-31.