The intent of this report is to document the different treatment methods and the first year post treatment response of Phragmites  in a 535-acre impoundment (East Pool) at the Edwin B. Forsythe National Wildlife Refuge (EBFNWR). Ultimately we wish to identify the most cost effective method for controlling Phragmites growth, while maximizing habitat benefits.  This will be achieved by monitoring the plant and hydrogeochemical conditions for 3 -5 post treatment years.  The pre-treatment measurements of baseline conditions in the East Pool and adjacent reference sites in the Northwest Pool and Leeds Trail salt marsh were made in the winter of 1999.  These baseline conditions included height and density of Phragmites, cation concentrations, pH and salinity.  A report describing the results of this baseline study was completed and submitted to EBFNWR in May of 1999, and was also posted on the web. Multilevel experimental treatments were initiated last year (1999) in the East Pool. The treatments imposed included 1) Tidal enhancement alone (TE, initiated July 1999); 2) TE + Glyphosate (G, September 1999); 3) TE + G + Burn (B, March 1999).  The post treatment data were collected in the winter of 2000.  In comparison with the pretreatment baseline data, the post-treatment data indicated that tidal enhancement (TE) resulted in an increase in salinity from 7 to 27 ppt and corresponding changes in the concentration of potassium, sodium and calcium ions.  Surprisingly, there was not a significant increase in the average pH measured on site (from 4.6 to 4.8). For Phragmites, 1), in the TE treatment site, there was a significant increase in mean plant density without a concomitant change in mean plant height; 2), in the TE+ G site, plant density increased significantly, coupled with a significant decrease in plant height; 3), for the TE + G + B treatment, there was no significant change in plant density, but a significant decrease in plant height was observed. Overall, the first year post-treatment data for winter are somewhat inconclusive in that the geochemistry and plant conditions were in a transitional stage.  Additional measurements should be made this year (summer 2000), and again in next several years (2001-2002).

The pre-treatment baseline data has been established and posted on the web (http://genius.rider.edu/~hsun/phrag.html). We intend to update more post-treatment information on our web site as it becomes available.

INTRODUCTION

Rapid spread of Phragmites australis to the exclusion of other preferred wetland plant species at the Edwin B. Forsythe National Wildlife Refuge poses a serious threat to resident and migratory waterfowl.  In the past thirty years Phragmites has invaded the impoundments and now covers more than 50% of the vegetated area. The problem is most evident in the East Pool where visual comparison with an early photo of the site (1963) show the dramatic increase in Phragmites. Thus far Phragmites has been managed with aerial applications of glyphosate in late August or early September (4 pints per acre), controlled burns in winter followed by intermittent flooding of fresh water in the summer months, and some mowing along the margins of the levee roads.  The effort to control the plant has been very costly with few benefits realized, as the plant seems to continue to grow and spread. Careful documentation and monitoring of treatment effects are impeded by limited funding and site access constraints (the impoundments are bounded on all sides by deep, mucky channels that are at times unnavigable by boat or foot).  The 535-acre East Pool was selected for the initial phases of this project due to the proposed plans by EBFNWR managers to restore tidal flow into this site for Phragmites control.  We hope to expand the project to include additional study sites in the other impoundments.

In the fall of 1998, we discussed the options that the EBFNWR might have for controlling Phragmites with Paul Steblein and Kelly Hogen. We agreed on a combination treatment experiment and a monitoring plan. The first treatment method was to introduce tidal water into the East Pool so that the fundamental growing conditions for Phragmites could be altered. We predicted that the increased salinity above a threshold of 19 ppt would suppress the growth of Phragmites and favor the growth of Spartina and other species (Sun et al., 1999, Figure 2). In some locations, another treatment method---burning was also planned to quickly reduce the above ground biomass of Phragmites for a quick visual improvement at the site. The third treatment method, applications of glyophosate in the fall was planned to minimize the potential of Phragmites returning the following year. The glyphosate has a half-life of 60 days, and is thought to have minimum impact on surrounding vegetation and wildlife if applied correctly. The disadvantage is that the Phragmites can return once the glyphosate decays since below ground reserves are often unaffected. Treatment with glyphosate is also expensive in comparison with other treatment methods. Based on the above consideration, treatment combinations were proposed, and implemented by refuge personnel in 1999 in East Pool. The treatment combinations that we consider in this report include: 1) Tidal enhancement alone (TE, initiated July 1999); 2) TE + Glyphosate (G, September 1999); 3) TE + G + Burn (B, March 1999). The team from Rider University collected the post treatment Phragmites growth and geochemical data.  The field data collection and measurements were made in January and February of 2000. The analyses of cations were conducted in February and March.  The data collected include plant height and density, cation concentrations, porewater pH, salinity at the time of collection. Our analysis of the data on the relationship of Phragmites growth and the geochemical conditions were included in the report as well.

THE EBFNWR SITE
TREATMENTS ON PHRAGMITES  AT EBFNWR

a. Glyphosate Application: EBFNWR commissions aerial applications of glyphosate at 4 pints per acre in late August or early September. Normally, glyphosate is applied at the end of the growing season when a large percent of the leave’s nutrients are translocated back into the rhizomes facilitating maximum die-back. Refuge personnel have used glyphosate over the past 20 years during periods when funding was available.  Normally, these treatments are followed by a prescribed burn the following February to remove dead standing biomass and litter. Preserve managers have reported some initial success with this procedure but when the treatments lapsed, Phragmites returned (Steblein, personal communication).  In 1998, glyphosate was administered to over 150 acres along the internal margins of the channels near the levee roads.  These areas were chosen for glyphosate applications since access constraints made them impossible to mow.

b. Prescribed Burning: Approximately 150 acres of Phragmites—dominated sites are burned yearly in the impoundments at EBFNWR.  In the State of New Jersey, the burn period is restricted to the winter months through March 15 to minimize impact to nesting birds. In the past 10 years at EBFNWR, controlled burns have followed glyphosate treatments. The pools are then intermittently flooded with fresh water in the summer months to further stress the plants. A drawdown follows during later summer to early fall to provide foraging habitat for migrating shorebirds.

c.  Flooding in Freshwater Systems:  The Northwest Pool has been managed in the last two years under a moist soil regime which involves drawdown in April and subsequent slow flooding across several months in the fall. The Southwest Pool has been maintained with deeper water to provide for submerged aquatic vegetation and to serve as a reservoir for other pools.  This has little if any effect on suppressing Phragmites growth except in some areas where combined with other treatments according to personnel (personal communication, Paul Steblein). Northwest and Southwest Pools are still managed as fresh water impoundments.

d.  Flooding with Salt Water (Tidal Enhancement):  The 535 acre East Pool has been maintained as a brackish impoundment since circa 1953.  The proposed plan to restore this site to tidal flow in 1999 was a management decision for the control of Phragmites.

e.  Combined Treatments: Prescribed burns and glyphosate applications alone may result in a reduction of aboveground biomass initially, but probably do not reduce the growth of Phragmites the following year due to the extensive underground rhizome system.  Burning may actually favor recolonization by Phragmites due to the reduction of aboveground biomass during the spring months and enrichment of the sediments.  Management personnel at EBFNWR combine glyphosate applications and/or burning which may result in greater suppression of Phragmites populations but the long-term ramifications are questionable due to cost [~$73 per acre for glyphosate and $ 43 per acre for prescribed burning- Paul Steblein, personal communication] and environmental impact.

Where appropriate, tidal enhancement combined with the above treatments may hold the greatest promise for long-term control of Phragmites.  However, there are only a handful of unpublished studies (most of which are not quantitative) to substantiate this claim.  To our knowledge, there are no other studies that involve impoundments this large.

EXPERIMENTAL DESIGN

a.  Macroplot Descriptions and Methods

Six macroplots are currently located in the EBFNWR wetland complex to identify both pre- and post-treatment Phragmites growth response and hydrogeochemical change (Figure 1). The macroplots were assigned the following treatments: 1) controlled burn + tidal enhancement (no glyphosate); 2) glyphosate application + controlled burn + tidal enhancement; 3) Tidal enhancement alone (no burn or glyphosate); 4) glyphosate application + tidal enhancement (no burn); 5) Freshwater Impoundment--no treatment; and 6) Saltmarsh--no treatment. An additional macro plot (7) was assigned recently to the Southwest Pool as a new control site.  The macroplots were subdivided into one to five (usually 3) subunits.  Subunits were assigned to areas that were most accessible and that represented target treatment zones. The subunits have been monumented with t-stakes. One-way ANOVA power tests were performed to identify sample size and minimum detectable difference in plant density and height needed to conclude whether a given treatment was effective or not (Elzinga et al., 1999; Zar, 1984).  Each subunit consisted of 8--0.5 m2 quadrats ( 8 to 40 per macroplot) randomly positioned on the interior of the Phragmites patches. Measured plant growth characteristics included culm density and height. For hydrogeochemical characterization, in situ measurements were made of porewater salinity (using a hand-held refractometer), and pH (using a portable pH meter). Porewater samples were collected simultaneously for the quantitative analysis of K, Ca, Fe, Mg, and Zn in solution. Samples were filtered, diluted with double distilled H20 and analyzed with a Baird ICP 2070 Sequential Plasma Spectrometer.

c.  Methods for the Treatments Imposed

Macroplots 1 - 4 in the East Pool were exposed to the appointed treatments as described in C-6a by the following methods (see Figure 1):

In September 1998, glyphosate was administered by plane along the eastern margins of the levees of the East Pool.  In the February of 1999, personnel at EBFNWR burned certain patches of Phragmites in the East Pool, leaving others untouched. Some regions were neither burned nor treated with glyphosate. On July 14 1999 three—36” Carolina water control structures and a large concrete tidal gate with 2—4 x 4’ bays were opened to restore tidal flow into the East Pool to assess the combined effects of the different treatments on Phragmites growth.  Freshwater influx from the two adjacent pools has been minimal.

The adjacent Northwest and Southwest Pools are still managed as fresh water impoundments.  Macroplot 5 in the Northwest pool served as a control site for our study.
 

PARAMETERS OF POST-TREATMENT MEASUREMENT

The post-treatment evaluations include the following:

(1) Plant Growth:  Phragmites density and height
(2) Hydrogeochemistry:  Interstitial sampling in each subunit for measuring salinity, pH, Eh, and cation concentration (Na, Mg, Ca, K).
(3) Rhizome Bioassay:  Preliminary tests of regeneration potential in rhizomes collected from the G + B + TE site suggest a useful test for assessing treatment affects.

POST-TREATMENT RESULTS FOR YEAR 1

a. The First Year Post-Treatment Results of Phragmites Treatments

Field plant density, height data were collected in late January and early February 2000.
Analysis of the first-year post-treatment data has been conducted. Overall, tidal enhancement (TE) resulted in an increase in mean (± SE) salinity from 7 ± 2.3 to 27 ± 0.95 ppt in the East Pool and an increase in the concentration of potassium and magnesium (Figure 3 and Figure 4) while calcium levels remained the same. For the tidally enhanced East Pool, the highest cation concentrations approximated levels found in seawater (Langmuir, 1997). For the freshwater Southwest Pool levels of Mg++, K+ and Ca++ were at the lower end of the range reported for agricultural soils (Tisdale et. al, 1993). Surprisingly, pore water pH remained low (from 4.6 to 4.8) even one year after tidal enhancement.  For Phragmites, in the TE treatment site, there was a significant increase in mean plant density (31 to 67 plants 0.5 m-2, p = .0001) without a concomitant change in mean plant height (166 to 161 cm); 2) in the TE + Glyphosate (G) site, plant density increased significantly (57 to 88, p= .0001) coupled with a significant decrease in plant height (217 to 165, p = .0001); 3) for the TE + G + Burn treatment, there was no significant change in plant density (41 to 43), but a significant decrease in plant height was observed (173 to 146 cm, p = .0001) (Figure 4).  Data for the TE + B treatment is not available due to the fact that this macroplot was accidentally left unburned. Of the treatments imposed, these preliminary results suggest that the greatest reduction in height (24%) was achieved by the G + TE treatment.  Part of this difference might be attributed to the porewater salinity in this macroplot which was higher (30 –33 ppt) than in the other macroplots (22 – 28 ppt).  For all the macroplots combined, plant height was negatively correlated to increased salinity and associated cation levels (P = .0001), but plant density appeared to be unrelated (p = .33). The density increase in some treated sites may reflect the accumulation of standing biomass from previous years in macroplots that weren’t burned. Moreover, since tidal enrichment of the East Pool was not implemented until halfway through the growing season, initial densities may be higher in these sites since the culms counted had already sprouted by the time of inundation. Macroplots of the east pool were compared to a control site in a neighboring impoundment (Northwest Pool) that was not treated.  No significant changes in salinity (9 to 5 ppt) or pH (4.7 to 4.3) were observed in comparing sample data from the two years. But both plant density (35 to 79) and plant height (203 to 221, p < .05) increased in the control site. We assume that there are no substantial changes in the soil mineral condition from the pre-treatment (Figure 5).

b. Discussion of Our Results and Directives for Further Study

1). Did the first year post-treatment data reflect treatment effects? Clearly, three years minimum of post treatment vegetation surveys during the winter and summer months are needed at all of the macroplots.  In Fairfield, Connecticut, a 25 acre diked marsh was restored to saltwater tidal action with a self-regulating tide gate (Bongiorno et. al.  1984).  Over three years time, plant height was reduced by 1 to 3 feet.  Plant density also showed a marked decline in one year from 11.3 plants m-2  to 3.3 plants m-2.  Will tidal enhancement have the same effect on the much larger (535 acre) East Pool?  Will the additional treatments using glyphosate and/or controlled burns facilitate faster and longer term suppression of Phragmites?

2).   To what extent does the surface flood water impact subsurface porewater conditions?  How tolerant is the Phragmites population at EBFNWR to increased salinity levels, and do the levels change with increased soil depth?  We would like to assess the effect of tidal flow on interstitial pore water salinity levels and ion concentrations at depths that reflect the rooting depth of Phragmites. Lissner and Schierup (1997) found that die-back of Phragmites occurred in lower fringe stands where the salinities exceeded 15 ppt within the rooting depth.  They measured salinities at depths up to 125 cm.  Surface flood water salinities of  up to 30 ppt sometimes dropped to less than 5 ppt at greater soil depths (Lissner and Shierup, 1997).  Phragmites may tolerate higher salinities in surface soils due to the fact that the roots may extend deeper into soils with lower salinity levels (Chambers, 1997).  In the East Pool, slight elevational differences between the different subunits may mean they are not equally exposed to the effects of tidal enrichment. Interstitial samplers are needed in each of the subunits at depths approximating the rooting depth of the plant for measuring porewater salinity and other hydrogeochemical variables.

3). What factors contribute to the low pore water pH in the impoundments and are these related to the success of Phragmites? Phragmites typically flourishes in alkaline and brackish environments (Haslam 1972, 1971), although it has been reported  in highly acidic wetlands and mine tailings (See Marks et al., 1994).  Normally, an increase in porewater pH is expected within several weeks after tidal enhancement (Ponnamperuma, 1979 ) primarily due to reduction of iron compounds.  Are low pH’s related to soil properties,  current management practices (Ponnamperuma, 1979),  or do they reflect the accumulation of leaf litter and other factors associated with invasive stands of Phragmites (Armstrong and Armstrong, 1999;  Templer et al., 1998)?  PH was measured within the top 5 to 10 cm of soil where leaf litter and organic debris had accumulated and may not reflect conditions deeper in the root profile.  Again, this necessitates installing interstitial samplers for this purpose.

4).  What are the baseline hydrological conditions defining the impoundments? Currently land managers believe the tidal range is 1 –2 feet inside the East Pool but this has not been substantiated. By monitoring the water levels with automated piezometers in each macroplot we will be able to predict the long-term water level change based on the tidal harmonic analysis.  At least 29 days of data are needed for the least square harmonic analysis. We also need to be able to calculate the water budget in the impoundment from the water level. The water level data will be the basis for analyzing geochemical variations, including pH, Eh, and the levels of cations in the impoundment. In general, a long-term deep water inundation will likely result in a more reduced condition, but it might also intensify the salinization of the soil and change the stress level on Phragmites growth. Due to the control of geochemical conditions on Phragmites growth and the effect of tidal fluctuation on the geochemical conditions in the soil, eventually, we will expect a qualitative to semi-quantitative integrative model of plant growth, geochemical and hydrological condition based on our understanding of their interrelationship. We can use this knowledge to predict, for impoundment systems, the number, the size, the distribution, and the water height and range required to optimize Phragmites control subject to the constraints of other management objectives.

5).  What is the effect of other treatment combinations on Phragmites growth? Unfortunately macroplot 1  (assigned for tidal enhancement +  burn alone) was accidentally left unburned and data is unavailable for assessing this particular treatment combination. However, there are nearby sites that were burned that could be assigned as replacement macroplots for the treatment.  Baseline estimates of plant height and density for all macroplots combined will be used for post-treatment comparisons in the newly monumented  macroplot.  We would also like to expand our experimental design to include macroplots in the freshwater impoundments that would test the effects of burn alone, glyphosate alone, and burn + glyphosate.  This would give us a total of 10 macroplots including the controls.

6). To what extent do the treatments affect plant diversity in the different macroplots?  Does species richness and diversity increase or decrease in different areas?  What species are favored under a given management regime?  Under the brackish regime, the East Pool supported species like Iva frutescens, Toxicodendron radicans, Distichlis spicata, Solidago sempervirens, Spartina patens, Euthamia graminifolia, Juniperus virginiana, Baccharis halimifolia, and Glaux maritima.  To what extent are these species (along with Phragmites) replaced by more salt tolerant species?  How will burned plots ± glyphosate compare to unburned plots ± glyphosate with respect to species richness and diversity? Bongiorno et al. (1984) reported the return of salt marsh flora including Salicornia, Distichlis, Spartina alterniflora, and S. patens within three years of tidal enrichment of a diked marsh.  Over time we would expect that the existing flora in the  East Pool would be replaced by flora more typical of the surrounding salt marsh (Eg. Spartina alterniflora and S. patens) although  fire and herbicide applications (Thompson and Shay, 1989; Marks et al., 1994) may alter species diversity.

7).   Can we predict the long-term effectiveness of a given treatment with a rhizome bioassay test? Rhizomes and roots are sustained during the winter months by oxygen transported through dead culms and in the summer months by anaerobic respiration (Gries et al, 1990). The osmotic pressure of the rhizomes has also been shown to increase with increased salinity levels, potentially enduring salinities up to 30 ppt (Lissner and Shierup, 1997).  To what extent are the under ground reserves of Phragmites protected from the stresses imposed by the treatment combinations explored in our project? A preliminary experiment looking at the effects of TE + G + B on regeneration potential of rhizomes provided some interesting and potentially useful insights.  We removed horizontal and vertical rhizomes from the top 10 cm of soil from macroplot 2, cut them into ~2.5 cm segments of equal weight, and planted the pieces individually in 3” pots filled with vermiculite.  The plants were watered identically to maintain saturated conditions for the duration of the experiment (12 weeks).  During this time the plants were fed every other week with equal amounts of Hoaglands solution. Twenty five of the 36 rhizome pieces sprouted, producing 4 – 5 cm long shoot tips.  Of the sprouted shoot tips, however, only one developed into a young plantlet, while the rest died.  Rhizomes extracted from treated and untreated macroplots may provide a useful bioassay for predicting which treatment will be most effective. Our predictions can then be correlated to the seasonal plant density data for each macroplot.  We would like to repeat this experiment in 2001 and 2002 following some of the methods described by Hellings and Gallagher (1992) and Lissner and Shierup (1997).

RECOMMENDATIONS FOR FOLLOW-UP  DATA COLLECTIONS

As planned in the initial proposal, we recommend at least two additional years of monitoring to establish a credible database for assessing the treatment effects, and developing cost effective management strategies.

1. Biannual measurements of Phragmites density and height, and hydrogeochemical variables.  These should be conducted in August, to reflect changes in live biomass in association with hydrogeochemical conditions and again, in the winter months (January, February) for comparison to baseline conditions.
2. A plant species inventory in the treatment areas is needed.   Funding was not available for a thorough pre-treatment documentation of plant species diversity in the East Pool.  Since the site is still undergoing transition from a brackish to a saline regime, it is possible to document the changes as transitional species are replaced by more salt tolerant species.  This is critical information for management considerations.
3. Water level data in the East Pool needs to be collected for the estimation of the water budget.  There has not been any hydrological monitoring in the site yet.
4. Bioassays of rhizomes collected from treatment areas may help us to predict which treatments were most effective, and the likelihood of the plant returning once the current management regime is changed.

ACKNOWLEDGMENTS

The authors want to thank Paul Steblein and Steve Atzert from EBFNWR for their help with the field work and coordination of the project. We also want to thank Jonathan Husch who is a professor in the Department of Geological and Marine Sciences at Rider University for many hours spent helping us with the ICP instrumentation. Jim Davidson deserves many thanks for assisting with field work and providing vehicular access to the site.  We also thank students Rebecca Crescitelli for her library research and assistance with a greenhouse experiment, and  Craig Tobe for data entry and lab work.   It is with the help from these people that we were able to accomplish so much on this project. Finally, we want to express our gratitude to the Department of Geological and Marine Sciences, and the Department of Biology at Rider University for access to their instruments and van.

REFERENCES

Armstrong, J., and Armstrong, W. 1999.  Phragmites die-back:  toxic effects of propionic, butyric, and caproic acids in relation to pH.  New Phytologist.  v142, P.201.

Armstrong, J., Armstrong, W. and Sinclair, A., 1996: Pathways of aeration and the mechanisms and beneficial effects of humidity- and Venturi-induced convections in Phragmites australis (Cav.) Trin. ex Steud.  Aquatic Botany. v 54, n2 / 3 P.177

Bongiorno, S.F., J.R. Trautman, T.J. Steinke, S. Kawa-Raymond and D. Warner.  1984.  A study of restoration in Pine Creek Salt Marsh, Fairfield, Connecticut.  In F. J. Webb (ed.) Proceedings of the 11th Annual Conference in Wetlands Restoration and Creation.  Hillsborough Community College, Tampa, FL.Elzinga, C.L., Salzar, D.W.   1998

Brady, N.C. and Weil R. R,1998: The Nature and Properties of Soils. Prentice Hall.

Chambers, R. M, 1997: Porewater chemistry associated with Phragmites and Spartina in a Connecticut tidal marsh.   Wetlands :  the journal of the society of the wetland scientists. v 17, n3    P.360

Elzinga, C.L., Salzar, D.W., and J.W. Willoughby, 1998.  Measuring and Monitoring
 Plant Populations.  BLM Technical Reference  1730-1. 477 pp.

FWS.  April, 1999.  Jersey Coast Refuges Draft Comprehensive Conservation Plan. Edwin B. Forsythe National Wildlife Refuge, POB 72, Great Creek Road, Oceanville, NJ  08231-0072

Gries, C.,  Kappen, L. and  Losch, R. 1990:  Mechanism of flood tolerance in reed, Phragmites australis (Cav.) Trin. ex Steudel. The New Phytologist. v 114, n4, P.589.

Haslam, S.M. 1971.  The development and establishment of young plants of Phragmites communis Trin. Annals of Botany N.S. 35: 1059-1072.

Haslam, S.M. 1972. Phragmites communis Trin. biological flora British Isles. Journal of Ecology. 60: 585-610.

Havens, Kirk J.  Priest III, and Walter I. Berquist, Harry, 1997.   Investigation and Long-Term Monitoring of Phragmites australis within Virginia's Constructed Wetland Sites. Environmental Management. v 21, n4,  P.599

Hellings, S.E. and Gallagher, J.L., 1992. The effects of salinity and flooding on Phragmites australis. The Journal of Applied Ecology. v 29, n1, P.41

Hood, L. 1998.  Science-Based Stewardship:  Recommendations for Implementing the National Wildlife Refuge System Improvement Act.  Defenders of Wildlife Handbook.  1101 Fourteenth St., N.W.  Suite 1400, Washington, D.C.  20005. 26pp.

Langmuir, D.  1997.  Aquatic Environmental Geochemistry.  Prentice Hall, Inc.  U.K.

Lissner, J.,  and Schierup, H.-H., 1997: Effects of salinity on the growth of Phragmites australis. Aquatic Botany.,  v 55, n4,  247

Marks, M.,  Lapin, B. and Randall, J., 1994:  Phragmites australis (P. communis): Threats, Management,  and Monitoring. Natural Areas Journal. v 14, n4, P.285

Mckee, Juno, and  Richards, A.J., Variation in seed production and germinability in common  reed (Phragmites australis) in Britain and France with respect to climate.  The New Phytologist.  1996 v 133, n2,  P.233

Neuhaus, D.,  Kuhl, H. and Kohl, J. G., 1993: Investigation on the genetic diversity of Phragmites stands using genomic fingerprinting. Aquatic Botany. v 45, n4. P.357

Ponnamperuma, F.N. (1979) The chemistry of submerged soils. Adv. Agron, 24, 29-96.

Roman, C.T., W.A.Niering, and R.S. Warren.  1984.  Salt marsh vegetation change in response to tidal restriction.  Environmental Management 8:141 – 150.

Sun, H.,  Browne, K. and Husch, J., 1998: Pore water geochemistry in a transition zone of Spartina to Phragmites  in a tidal marsh in Stone Harbor, New Jersey. Geological Society of America, Abstract with programs, Vol.31, no.2.

Templer, Pamela, Findlay, S. and Wigand, C.  1998.  Sediment chemistry associated with native and non-native emergent macrophytes of a Hudson River marsh ecosystem.  Wetlands. V. 18,  n1 P.70.

Thompson, D.J. and  Shay, J. M., 1989: First-year response of a Phragmites marsh community to seasonal burning.  Canadian Journal of Botany. v 67, , n5, P.1448

Tiner, R. May, 1997.  Managing Common Reed (Phragmites australis) in Massachusetts-
 an overview of the Species and Control Techniques.  Mass. Wetlands Restoration and Banking Program.  Executive Office of Environmental Affairs.  100 Cambridge St., Boston, MA  02202.

Tisdale, S.L., W.L. Nelsen, J.D. Beaton, and J.L. Havlin.  1993. Soil Fertility and Fertilizers. 5th Ed.  Macmillan Publ. Co., New York. P. 365

Tucker, G.C.  1990.  The general of Arundinoideae (Gramineae) in the southeastern United States.  Journal of Arnold Arboretum  71:  145-177.

Windham, L. and Lathrop, Richard G. Jr. 1999. Effects of Phragmites australis (Common Reed) invasion on aboveground biomass and soil properties in a brackish tidal marsh of the Mullica River, New Jersey.  Estuaries.  V. 22, n4  P. 927

Zar, J.H. (1984) Biostatistical Analysis. Prentice Hall, New Jersey.