Water Chestnut (Trapa natans L.) Infestation in the ... - SUNY Oneonta

Water Chestnut (Trapa natans L.) Infestation in the 

Susquehanna River Watershed: 

Population Assessment, Control, and Effects 

Willow Eyres 

BIOLOGICAL FIELD STATION 

ONEONTA, N.Y. 

STATE UNIVERSITY OF NEW YORK 

COLLEGE AT ONEONTA 

Occasional Paper No. 44 

June 2009

OCCASIONAL PAPERS PUBLISHED BY THE BIOLOGICAL FIELD STATION 

No. 1. The diet and feeding habits of the terrestrial stage of the common newt, Notophthalmus viridescens (Raf.). M.C. MacNamara, April 

1976 

No. 2. The relationship of age, growth and food habits to the relative success of the whitefish (Coregonus clupeaformis) and the cisco (C. 

artedi) in Otsego Lake, New York. A.J. Newell, April 1976. 

No. 3. A basic limnology of Otsego Lake (Summary of research 1968-75). W. N. Harman and L. P. Sohacki, June 1976. 

No. 4. An ecology of the Unionidae of Otsego Lake with special references to the immature stages. G. P. Weir, November 1977. 

No. 5. A history and description of the Biological Field Station (1966-1977). W. N. Harman, November 1977. 

No. 6. The distribution and ecology of the aquatic molluscan fauna of the Black River drainage basin in northern New York. D. E Buckley, 

April 1977. 

No. 7. The fishes of Otsego Lake. R. C. MacWatters, May 1980. 

No. 8. The ecology of the aquatic macrophytes of Rat Cove, Otsego Lake, N.Y. F. A Vertucci, W. N. Harman and J. H. Peverly, December 

1981. 

No. 9. Pictorial keys to the aquatic mollusks of the upper Susquehanna. W. N. Harman, April 1982. 

No. 10. The dragonflies and damselflies (Odonata: Anisoptera and Zygoptera) of Otsego County, New York with illustrated keys to the genera 

and species. L.S. House III, September 1982. 

No. 11. Some aspects of predator recognition and anti-predator behavior in the Black-capped chickadee (Parus atricapillus). A. Kevin Gleason, 

November 1982. 

No. 12. Mating, aggression, and cement gland development in the crayfish, Cambarus bartoni. Richard E. Thomas, Jr., February 1983. 

No. 13. The systematics and ecology of Najadicola ingens (Koenike 1896) (Acarina: Hydrachnida) in Otsego Lake, New York. Thomas 

Simmons, April 1983. 

No. 14. Hibernating bat populations in eastern New York State. Donald B. Clark, June 1983. 

No. 15. The fishes of Otsego Lake (2nd edition). R. C MacWatters, July 1983. 

No. 16. The effect of the internal seiche on zooplankton distribution in Lake Otsego. J. K. Hill, October 1983. 

No. 17. The potential use of wood as a supplemental energy source for Otsego County, New York: A preliminary examination. Edward M. 

Mathieu, February 1984. 

No. 18. Ecological determinants of distribution for several small mammals: A central New York perspective. Daniel Osenni, November 1984. 

No. 19. A self-guided tour of Goodyear Swamp Sanctuary. W. N. Harman and B. Higgins, February 1986. 

No. 20. The Chironomidae of Otsego Lake with keys to the immature stages of the subfamilies Tanypodinae and Diamesinae (Diptera). J. P. 

Fagnani and W. N. Harman, August 1987. 

No. 21. The aquatic invertebrates of Goodyear Swamp Sanctuary, Otsego Lake, Otsego County, New York. Robert J. Montione, April 1989. 

No. 22. The lake book: a guide to reducing water pollution at home. Otsego Lake Watershed Planning Report #1. W. N. Harman, March 1990. 

No. 23. A model land use plan for the Otsego Lake Watershed. Phase II: The chemical limnology and water quality of Otsego Lake, New 

York. Otsego Lake Watershed Planning Report Nos. 2a, 2b. T. J. Iannuzzi, January 1991. 

No. 24. The biology, invasion and control of the Zebra Mussel (Dreissena polymorpha) in North America. Otsego Lake Watershed Planning 

Report No. 3. Leann Maxwell, February 1992. 

No. 25. Biological Field Station safety and health manuel. W. N. Harman, May 1997. 

No. 26. Quantitative analysis of periphyton biomass and identification of periphyton in the tributaries of Otsego Lake, NY in relation to 

selected environmental parameters. S. H. Komorosky, July 1994. 

No. 27. A limnological and biological survey of Weaver Lake, Herkimer County, New York. C.A. McArthur, August 1995. 

No. 28. Nested subsets of songbirds in Upstate New York woodlots. D. Dempsey, March 1996. 

No. 29. Hydrological and nutrient budgets for Otsego lake, N. Y. and relationships between land form/use and export rates of its sub -basins. M. 

F. Albright, L. P. Sohacki, W. N. Harman, June 1996. 

No. 30. The State of Otsego Lake 1936-1996. W. N. Harman, L. P. Sohacki, M. F. Albright, January 1997. 

No. 31. A Self-guided tour of Goodyear Swamp Sanctuary. W. N. Harman and B. Higgins (Revised by J. Lopez),1998. 

No. 32. Alewives in Otsego Lake N. Y.: A Comparison of their direct and indirect mechanisms of impact on transparency and Chlorophyll a. 

D. M. Warner, December 1999. 

No.33. Moe Pond limnology and fish population biology: An ecosystem approach. C. Mead McCoy, C. P. Madenjian, V. J. Adams, W. N. 

Harman, D. M. Warner, M. F. Albright and L. P. Sohacki, January 2000. 

No. 34. Trout movements on Delaware River System tail-waters in New York State. Scott D. Stanton, September 2000. 

No. 35. Geochemistry of surface and subsurface water flow in the Otsego lake basin, Otsego County New York. Andrew R. Fetterman, June 

2001. 

No. 36 A fisheries survey of Peck Lake, Fulton County, New York. Laurie A. Trotta. June 2002. 

No. 37 Plans for the programmatic use and management of the State University of New York College at Oneonta Biological Field Station 

upland natural resources, Willard N. Harman. May 2003. 

No. 38. Biocontrol of Eurasian water-milfoil in central New York State: Myriophyllum spicatum L., its insect herbivores and associated fish. 

Paul H. Lord. August 2004. 

No. 39. The benthic macroinvertebrates of Butternut Creek, Otsego County, New York. Michael F. Stensland. June 2005. 

No. 40. Re-introduction of walleye to Otsego Lake: re-establishing a fishery and subsequent influences of a top Predator. Mark D. Cornwell. 

September 2005. 

No. 41. 1. The role of small lake-outlet streams in the dispersal of zebra mussel (Dreissena polymorpha) veligers in the upper Susquehanna 

River basin in New York. 2. Eaton Brook Reservoir boaters: Habits, zebra mussel awareness, and adult zebra mussel dispersal via 

boater. Michael S. Gray. 

No. 42. The behavior of lake trout, Salvelinus namaycush (Walbaum, 1972) in Otsego Lake: A documentation of the strains, movements and the 

natural reproduction of lake trout under present conditions. Wesley T. Tibbitts. 

No. 43. The Upper Susquehanna watershed project: A fusion of science and pedagogy. Todd Paternoster. 

Annual Reports and Technical Reports published by the Biological Field Station are available from Willard N. Harman, BFS, 5838 St. Hwy. 80, 

Cooperstown, NY 13326. 

2

Abstract 

TABLE OF CONTENTS 

Introduction…………………………………………………………………… 4 

Background 

Biology of Trapa natans L. …………………………………………… 5 

Distribution …………………………………………………………… 7 

Vectors of Invasion and Dispersal …………………………………… 8 

Effects on Ecosystem Processes ……………………………………… 9 

Impacts on Humans…………………………………. ………………. 11 

Invasion of Chesapeake Bay Watershed……………………………… 11 

Invasion of Hudson River Basin……………………………………… 12 

Methods and Attempts of Control……………………………………. 14 

Herbicidal Action of 2,4-D: Biochemistry and Physiology…………… 15 

Methods 

Site description………………………………………………………… 17 

Plant and Water Sampling……………………………………………. 17 

Plant Biomass and Distribution………………………………………. 18 

Herbicide Application………………………………………………… 18 

Results 

Plant Survey……………………………………………………………. 20 

Water Quality Analysis Data…………………………………………… 20 

Plant Biomass and Distribution Data…………………………………… 22 

Effects of 2,4-D Herbicide……………………………………………… 25 

Population Demographics- PIRTRAM Data…………………………… 28 

Orthoimagery/ GIS Data ……………………………………….. 30 

Discussion 

Nutrient Concentrations………………………………………………… 33 

Plant Biomass…………………………………………………………. 35 

Conclusion/ Recommendations………………………………………………… 36 

References……………………………………………………………………… 38

ABSTRACT 

A four acre population of the invasive European water chestnut (Trapa natans L.) 

was treated with 150, 200 and 200 pounds per acre of 2,4-D herbicide in the consecutive 

summers of 2006, 2007 and 2008 respectively, in a 40 acre wetland in Oneonta, NY, in 

close proximity to the Susquehanna River. The water chestnut is an aquatic weed found 

throughout the northeastern United States that can dominate ponds, shallow lakes, and 

river margins. It displaces native vegetation and limits navigation and recreation. Water 

nutrient analysis and aquatic plant biomass and distribution procedures were undertaken 

to monitor the impacts of the herbicide on both Trapa and non-target plants and changes 

of nutrient concentration within the system. Due to flooding of the Susquehanna in June 

2006, results of water quality analysis proved inconclusive in regard to any effect of any 

impacts on the water chestnut population. Nutrients in the wetland before flooding 

showed very high concentrations of nitrogen and phosphorus compared to post flood data 

collection. 

Despite a decline in overall biomass of non-target plants, the decrease was not 

statistically significant. Water chestnut biomass increased by about 450g (dry weight) 

between pre-treatment and post-treatment periods in 2006, though it can be speculated 

that if the herbicide not been applied, the increase may have been much greater. After the 

second herbicide application in 2007 there was a substantial decline in both water 

chestnut biomass and growth rate. Population estimates indicate that with continued 

annual herbicide applications, at this rate of 0.62 per year, the number of water chestnut 

rosettes may be reduced enough in five years to continue eradication efforts by hand 

pulling alone. As seen with other infestations, simply preventing the spread of this 

population does not ensure that the ramets of the current population could not be carried 

or travel to the Susquehanna and perpetuate a harmful infestation. However, failing to do 

so will virtually guarantee its further spread. Therefore, the success of this project will 

continue by maintaining funding for yearly applications through 2013, followed by 

dedicated volunteers’ hand pulling each spring.

INTRODUCTION 

There are currently only three known populations of European water chestnut 

(Trapa natans L.) inhabiting the Susquehanna River Basin. Other invasive infestations 

have dominated wetlands, shallow ponds, and slow moving streams throughout the 

northeastern United States for the past few decades; altering water chemistry, 

temperature, and light penetration and therefore out competing native plants, disturbing 

local fish and invertebrate populations (Naylor 2003). 

One small population inhabits Goodyear Lake near Portlandsville NY, where 

hand pulling efforts could still be very effective. A second population is in Cincinnatus 

Lake (Chenango County), while the third inhabits a 40 acre pond in west Oneonta, NY 

and is the focus of this study (FIG 3). The pond is privately owned by Mr. Louis Blasetti. 

The water chestnut population was first observed in 2000 and has since grown to 

approximately four acres in size, crowding out other previously dominant aquatic plant 

species: Ceratophyllum demersum, Potamogeton crispus, and Elodea canadensis. 

Propagules of the plant are presumed to have been deposited in the pond by attaching to 

the plumage of local birds and being transported from another region. 

Many hectares of wetlands are chemically treated annually to control exotic plants 

in New York State. Does control of large populations of ecologically dominant plants 

release significantly large amounts of nutrients into aquatic systems already stressed by 

eutrophication? Given the potential for federal regulation of nutrient loading via total 

maximum daily loads (TMDLs) in the Susquehanna Drainage Basin in the near future, 

how important are such considerations to agencies impact on large plant control programs 

in the region? The work described in this thesis and analysis of water quality information 

will begin to provide insight into some of the mechanisms of water chestnut invasion, and 

its potential control. 

Much attention has been brought to this project by the biologists of the 

Department of Environmental Conservation, members of the Otsego County 

Conservation Association, and local environmentalists due to concern over the adverse 

effects that the chestnut could have in the Susquehanna River drainage basin. Water 

chestnut in the Chesapeake Bay watershed demonstrates its invasive and dominating 

abilities in the waters of the Mid- Atlantic and the Northeast. During one year in the 

Sassafras River, a Maryland tributary of the Chesapeake Bay, the water chestnut 

population grew from about 50 to thousands of plants to covering over three acres of 

surface water. There, the water chestnut caused major problems with hydrology (currents, 

temperatures, and light penetration), native plant populations, as well as navigation and 

recreation (Naylor 2003). Management plans for the water chestnut in that case included 

hand-pulling efforts as well as several applications of the herbicide 2,4-D. After several 

years of continued control, the region has begun to recover (Naylor 2003). 

4

In an effort to prevent further dispersion of the Oneonta water chestnut 

population, it was thought best by project advisors to use a chemical weedkiller. In past 

years, hand-pulling efforts were ineffective in the wetland. Monies from the New York 

Power Authority, the Millennium Pipeline Company and a legislative grant from the 

NYS Senate were donated and used to purchase a quantity of the herbicide 2,4-D and to 

develop a work plan. The herbicide was applied by the Allied Biological Company in 

August, 2006, in June, 2007 and in June, 2008. All necessary permits were issued by the 

Department of Environmental Conservation. 

Biology of the Water Chestnut 

BACKGROUND 

Water chestnut (Trapa natans L.) is an aquatic plant that grows perennially in its 

native Europe, Asia and the northern countries of Africa (NRCS 2007). Non-native 

populations in New York and New England act as annuals because of the regions’ 

freezing temperatures (Groth et. al. 1996). The genus was previously placed in the family 

Trapaceae; however modern molecular research puts Trapa under Lythraceae in the order 

Myrtales (Stevens 2001). It grows best in shallow, nutrient rich lakes, rivers and ponds 

and is generally found in waters with a pH range of 6.7 to 8.2 and alkalinity of 12 to 128 

mg/L of calcium carbonate (Naylor, 2003). 

Water chestnut is a dicot with a floating rosette of leaves around a central stem. 

Rosette forming species respond to water movements and buoyant tissues in the stem, 

and leaves maintain stability on the surface of the water. The spongy inflated leaf 

petioles of T. natans also help the rosette to float (FIG. 1). Many aquatic species with 

rosettes, including Trapa, have a leaf mosaic with a wide range of leaves that develop on 

petioles containing layers of spongy tissue with large air spaces, termed aerenchyma 

(Groth et al. 1996). The leaves of floating plants are forced to physiologically handle 

exposure to air and water simultaneously. Carbon dioxide and oxygen are absorbed 

through stoma in the upper epidermis. The leaves also have a prominent waxy cuticle that 

prevents external conditions from raising the rate of transpiration. Also, floating leaves 

will usually take a circular peltate form (Sculthorpe 1967). The lamina of T. natans is 

rhombic in shape and is toothed toward the tip of the leaf (Naylor 2003). The leaves have 

little or no lignin and the vascular tissues are generally poorly developed in the leaves 

(Sculthorpe 1967). The upper stem swelling has a lacunate pith and four or five rings of 

air spaces in the cortex whereas the pith is compact having only two rings of cortical 

lacunae in the lower stem (Naylor 2003). 

The inconspicuous flowers are found on the leaf axils of younger leaves above the 

water. As the meristem elongates and produces new leaves, the older leaves and 

developing fruit become submerged (Sculthorpe 1967). The single seeded mature fruit 

are woody and bear four sharply pointed horns. Water chestnuts begin to flower in early 

June and the nuts will mature approximately a month later. Flower and seed production 

continue into the fall until the first frost kills the rosettes. When mature, the fruits fall 

from the plant and sink to the bottom of the body of water. Seed dormancy can be from 

four months to twelve years. The horns may act as anchors to limit movement of the seed, 

5

thus keeping them at suitable water depths (Naylor 2003). Winter survival of the nuts 

generates the bed of Trapa at that site the following year. A small fraction of nuts are also 

carried on buoyant, detached floating ramets and in this way the nuts are dispersed to 

downstream sites (Groth et al. 1996). 

Aquatic annuals are unique in that a large number of them propagate clonally, 

whereas most terrestrial annuals do not. In an annual species, clonal growth multiplies 

the opportunities of an individual for sexual reproduction without producing overlapping 

generations of ramets. A ramet is defined as a clonal offshoot. The explosive growth of 

this exotic plant may be due to several phenomena. There is some evidence that this 

species behaves as a perennial in parts of North America, and the rapid expansion of 

populations of the plant may be due to the proliferation of clonal fragments that 

subsequently proliferate the following year. The increase may also be via an increase in 

the rate of seed production. Typically, in this species, only one seed within the nut 

develops, but it may be that under low density conditions, both seeds develop. It is also 

possible that phenotypic plasticity allows it to develop more flowers per rosette, or more 

flowers may successfully develop into nuts, at low densities (Groth et al. 1996). 

Water chestnut has adventitious roots that develop in pairs on either side of the 

leaf scars at lower nodes of the floating stem. The roots are feathery and can often reach 

to the sediment, but usually remain suspended in the water column (Groth et al. 1996). 

The roots also contain chlorophyll which has often misled people to think they were 

submerged leaves with segments comparable to the terrestrial roots of such species as 

Bergia capensis or Heteranthera zosteraefolia (Sculthorpe 1967). Some oxygen reaches 

the internal tissues of the roots by diffusing in solution along the epidermal gradient but 

oxygen also diffuses from photosynthetic sites. There is a system of cortical lacunae for 

diffusion (Sculthorpe 1967). Like many other aquatic plants, Trapa has no primary root 

system, just the adventitious roots that extend from the stem region of the seedling below 

the cotyledons (hypocotyls). The lateral roots contain only one strand of xylem and 

phloem. Although the most important function of the roots is to absorb nutrients, they 

also provide an anchor for the plant, but the developmental origin of the roots is unclear 

(Groth et al. 1996). 

Typically a Trapa plant is capable of producing three primary ramets and they 

develop in a specific order. The first ramet arises from the center of the nut; the second 

develops on the side opposite the hypocotyls, and the third between the first shoot and the 

hypocotyls (Groth et al. 1996). 

6

FIGURE 1. The distinguishing rosette (1), nut (2), leaflet showing buoyancy bladder (3) 

and root stalk with filiform rootlets (4) of T. natans. Modified from 

http://aquat1.ifas.ufl.edu/tranatdr.jpg 

Distribution 

Water chestnut is native to the warm temperate regions of Eurasia and North 

Africa. There is some discrepancy in the literature as to which decade Trapa natans first 

entered the U.S., and when it became established. Naylor (2003) states water chestnut 

was first recorded in North America near Concord, Massachusetts in 1859. Hummel and 

Findlay (2006) state that it was first introduced to the U.S. in 1875, while Pemberton 

(1999) states it was first observed in 1884, growing in Sanders Lake, Schenectady, New 

York. 

Populations have become established in many locations in the northeastern United 

States, including the Hudson River, Lake Champlain and six of its tributaries, the Nashu 

River in New Hampshire and the Connecticut River in Connecticut. The plant has also 

been documented in Delaware, Maryland, Massachussets, Pennsylvania and Virginia 

(Pemberton 1999) (FIG 2). 

Trapa also thrives in the Great Lakes Basin, the Oswego River, Oneida Lake, and 

the Erie Barge Canal System (Pemberton 1999). In 1998, water chestnut was found in the 

South River in Quebec, which is connected to the Lake Champlain outlet via the 

Richelieu River. Its spread has continued because of the suitability of habitat. In 2001, 

7

Trapa was found in the Pike River in Quebec, which flows into Mississquoi Bay in 

Vermont (Pemberton 1999). 

FIGURE 2. Distribution of Trapa natans in the US (2004). 

Source: http://www.anr.state.vt.us/dec/waterq/lakes/images/ans/lp_wc-usamap.gif 

Vectors of Invasion and Dispersal 

Water chestnut was introduced into the wild sometime before 1879 by a gardener 

at the Cambridge Botanical Garden in Cambridge, Massachusetts. The gardener reported 

planting it in several ponds. It was also introduced in Concord, Massachusetts, where it 

was planted in a pond adjacent to the Sudbury River. By the turn of the century, it 

proliferated in the pond and the river. Since then, water chestnut has spread to other states 

and other river and estuary systems (Naylor 2003). Ballast waters also offered an easy 

means for Trapa nuts to gain entrance to America (Mills et al. 1996). Water chestnut has 

also become naturalized in Australia as well (Heywood 1993). 

Hellquist (1997) believes that once introduced, Trapa is dispersed primarily by 

ducks and geese, but it is unlikely that the nuts could be carried over long distances. 

Although observations have been made of Canada geese with Trapa fruits attached to 

their feathers, the size and weight of the propagules make it unlikely they would remain 

attached during prolonged flight. Because Trapa fruits fall to the bottom of lakes and 

rivers, there is a low probability of getting tangled in plumage. It has also been 

determined that muskrat eat Trapa fruits and many also facilitate their dispersal (Les and 

Merhoff 1999). 

Trapa is also believed to be a determined “hitchhiker”, which accounts for its 

dispersal from the Hudson River to Lake Champlain on boats, clinging to ropes and nets 

(Countryman 1970). Wind and wave action disperse plant pieces and fruits locally. Trapa 

8

fruits have long been consumed by humans and were sold by street vendors in western 

New York State from about 1925 to 1935. Canned Trapa fruits are sold in gourmet food 

shops and plants are still being cultivated for the edible nuts, however, most tinned 

“water chestnuts” are in fact Cyprus esculentus (Les and Merhoff 1999). The Trapa fruit 

contains much starch and fat, and are a staple food in eastern Asia, Malaysia, and India 

(Heywood 1993). 

Effects on Ecosystem Processes 

Trapa natans and many other invasive aquatics can invade an area and severely 

alter an ecosystem. Wetlands, in particular, seem especially vulnerable to these invasions 

because they are landscape sinks, which accumulate debris, sediments, water and 

nutrients. Even though less than 6% of the earth’s land mass is wetland, 24% of the 

world’s most invasive plants are wetland species (Zedler and Kercher 2004). Wetland 

invaders contrast with many terrestrial invaders in that: seeds are often dispersed by 

water, plants and plant parts can be dispersed by flotation, and aerenchyma protects 

below ground plant tissues from flooding in anoxic soils and has the ability for rapid 

nutrient uptake, thus allowing for rapid growth (Zedler and Kercher 2004). 

In wetlands, non-indigenous species abundance is associated with road density, 

suggesting that landscape position interacts with dispersal pathways and disturbances to 

help plant establishment. Wetlands fed by surface water from agricultural and urbanized 

watersheds usually have many invasive species. Wetlands that are not fed primarily by 

surface water have small watersheds, depending on other sources for their water supply 

like rainfall or groundwater. These wetlands are usually species rich and relatively free of 

invasive plants (Zedler and Kercher 2004). 

There are many characteristics of wetlands that provide an area for opportunistic 

plant invaders such as: runoff, nutrient cycles, sediment composition, open standing 

water, human made structures, and salinity cycles. The characteristics that benefit an 

invasion by Trapa natans will be discussed in more detail. Floodwaters accumulate in 

wetlands, and anoxia becomes a cumbersome challenge for most species, except those 

that are flood tolerant. These species usually possess aerenchyma tissues. Plants with 

aerenchyma can also achieve high plant biomass, potentially growing very rapidly. Trapa 

stems contain aerenchyamtous tissues and therefore have that advantage. Trapa also has a 

great advantage in that its adventitious roots positively respond to changes in water depth 

and nutrient availability. Dense, floating mats of rhizomes provide another advantage for 

reasons discussed earlier (Orth and Moore 2004). 

Wetlands have shown to be significantly altered by plant invaders. Many invasive 

plants are unwanted because of the effects they have on habitat structure. Species that 

alter the physical structure of a site have high potential for shifting hydrological 

conditions and animal uses. Invasive plants are commonly understood to reduce both 

plant and animal diversity. (Zedler and Kercher 2004). Invasive plants that differ from 

native species in biomass and productivity, tissue chemistry, plant morphology, or 

phenology, can alter soil nutrient dynamics. Invasive species can affect food webs in 

multiple ways, by altering the quantity and quality of food, by changing food supply, or 

by changing susceptibility to predators (Zedler and Kercher 2004). 

9

Sedimentation is both a cause and effect of wetland invasions. In wetlands in 

which sediments are entering, invasive plants find canopy gaps and bare soils to colonize. 

Where sturdy invasive plants colonize stream banks, sediments accumulate and alter 

geomorphology. The outcomes are similar in both habitats in that the topography is 

simplified and this is detrimental to the recipient community’s ability to support diversity 

in vegetation. At the same time, sediments carry nutrients (especially phosphorous) that 

cause eutrophication and more aggressive growth of many invasive plants (Zedler and 

Kercher 2004). Overall, invasive species are reported to significantly alter 

geomorphologic processes by increasing erosion rates, increasing sedimentation rates, 

increase soil elevation, or impact the effective geometry or configuration of water 

channels (Gordon 1998). 

Species that alter geomorphology are also likely to influence hydrological systems 

by altering hydrological cycling, changing water tables, or altering surface flow patterns. 

Non-indigenous species with evapotranspiration rates higher than those of the native flora 

may significantly alter water cycles. Gordon (1998) found that nitrogen-fixing invaders 

will alter biogeochemical cycles, effect soil nutrient availability and significantly alter 

water chemistry. This in turn will effect submerged vegetation and phytoplankton. 

Aquatic macrophytes that form canopies also have their own set of effects on 

bodies of water. Extensive covers of floating plants, such as those produced by Trapa, 

shelter the surface from wind, reduce turbulence and aeration, restrict mixing and 

promote thermal stratification. Frodge et al. (1990) hypothesized that the structure of the 

plant canopies are functionally important to variations in water quality and that in dense 

beds the canopies can vertically divide the water column. Their study found that water 

quality differences and daily changes were strongly connected to the development of 

dense surface canopies. Significant differences in water temperature and dissolved 

oxygen were observed between the surface and the sub-canopy water. The low subcanopy 

dissolved oxygen concentrations, and lack of daily change in dissolved oxygen, 

indicated a reduction of sub-canopy photosynthesis, even during daylight hours. The 

self-shading by macrophytes can, therefore, change the lower stem area to a site of 

oxygen demand rather than an area of oxygen surplus. However, the plant canopy effect 

appeared to be dependent on the size and geometry of the body of water. A deeper lake 

with a larger ratio of open water would be naturally buffered to the impacts of the plant 

beds. Their study even suggested that the areas above and below the canopies could be 

considered fundamentally different habitats (Frodge et al. 1990). 

In eutrophic waters, aquatic macrophytes such as Trapa can grow vigorously and 

play a significant role in removing nutrients from polluted water. Floating-leaved plants 

are characterized by a short life span, which results in high rates of biomass turnover. 

Nutrient availability has been described to affect the leaf life-span of terrestrial plants; 

and even though there are small amounts of data for aquatic macrophytes, the same is 

hypothesized to be true. In a study by Tsuchiya and Iwakuma (1993), the data showed 

that with increased nitrogen availability, net production of T. natans increased as well, 

concluding that growth may be restricted by nitrogen flux. The study also discussed 

Trapa’s ability to take up nitrogen from both the water and from the sediment (Tsuchiya 

and Iwakuma 1993). 

10

Yet another effect of Trapa on ecosystem processes is in the area of invertebrate 

communities. As mentioned before, water chestnut leaves release oxygen into the 

atmosphere while the stems and roots consume oxygen from the water, so beneath the 

large, dense beds the water may become hypoxic or even anoxic. Because Trapa has a 

different architecture than submerged plants, and depletes water of dissolved oxygen, it 

has been thought to support distinctive communities of macroinvertebrates and fish 

(Strayer et al. 2003). 

Impacts on Humans 

The impacts of a water chestnut invasion are not only devastating ecologically, 

but also negatively affect humans. In most areas the biggest problem has become the 

interference of water chestnut in recreational and economic uses of navigable waters. 

Dense mat and root systems can completely cover the surface of the water, preventing 

swimming and canoeing and tangling in propellers of motor boats. In addition, the spiny 

seeds of the chestnut have been known to cause harmful injury to bathers and beach 

users. Similarly to infestations of Eurasian water milfoil (Myriophyllum spicatum), the 

mats are favorable sites for mosquito breeding. Water chestnut also affects the aesthetic 

value of an area. The plant is likely to be regarded as unattractive in large quantities and 

can be unsightly when washed ashore. Recreational fishing is also affected as many fish 

populations tend to avoid the infested areas because normal biological processes are 

terminated or severely reduced (NEMESIS 2005) 

Economically, efforts to reduce plant population sizes and stop its spreading have 

been costly. In the Chesapeake Bay region alone, $2.8 million have been spent in the past 

20 years for mechanical harvesting, herbicide applications and hand pulling and 

monitoring programs (NEMESIS 2005). 

Because of the nuisance of water chestnut and other aquatic invasives, more 

precautions are being taken and more legislation is being created to address these 

concerns. For example, many states have created strict legislation to require permits for 

all water withdrawals and water transports to prevent the spread of any invasive plants, 

including New York. Bulk water transporters that offer such services as filling swimming 

pools, hydroseeding, irrigation, spraying for dust control and roadbed compaction at 

construction sites, and similar activities often withdraw water from rivers or lakes at 

convenient access points. Many states now require pipes, hoses and tanks of trucks to be 

inspected and thoroughly cleaned (Mills et al. 1996). 

Invasion of Chesapeake Bay Watershed 

A distinct feature, and one of the Chesapeake Bay’s vital natural resources, is the 

beds of submerged aquatic plants that inhabit the shallow water areas. In addition to its 

high primary productivity, this vegetation is significant because it is a food source for 

waterfowl, a habitat and nursery for many species, a shoreline erosion control system and 

a nutrient buffer. However, over the past 50 years, there have been several distinct 

periods in which significant changes occurred within the submergent aquatic vegetation. 

These ecological changes began with the Zostera marina wasting disease in the 1930s 

11

and Myriophyllum spicatum and Trapa natans proliferation in the 1950s. These two 

periods caused widespread changes in plant communities during the 1960’s and 70’s 

(Orth and Moore 1984). 

Within the Chesapeake Bay watershed, water chestnut first appeared in 1923, on 

the Potomac River near Washington D.C. as a two acre patch. The plant spread rapidly, 

covering 40 miles of river in just a few years. By 1933, 10,000 acres of dense beds 

extended from D.C. to Quantico, Virginia. Water chestnut was first recorded in the Bird 

River in Baltimore County for the first time in 1955 (Orth and Moore 1984). The 

Maryland Department of Game and Inland Fish and Tidewater Fisheries used mechanical 

removal and an herbicide (2,4-D, the only fully licensed herbicide successfully used 

against water chestnut) to control it. However, in 1964 it reappeared in the Bird River and 

an additional 100 acres were discovered in the Sassafras River in Kent County, of which 

30 acres were mechanically removed. This effort was highly successful as no plants were 

reported in surveys until 1997 when a water chestnut population was again discovered in 

the Bird River (Naylor 2003). The infestation spread from approximately 50 plants in the 

summer of 1997 to over three acres in 1998, demonstrating the explosive propagation 

ability of Trapa natans in some habitats. The Bird River population increased again into 

the Sassafras River and a substantial mechanical and volunteer harvesting effort began on 

both rivers in 1999, resulting in the removal of almost 400,000 pounds of plants from the 

two rivers. This undertaking was successful but researchers realized that viable nuts still 

remain in the sediments and that continuous follow-up measures will be necessary 

(Naylor 2003). 

Invasion of Hudson River Basin 

The Hudson River Basin drains parts of five states (New York, New Jersey, 

Massachusetts, Connecticut and Vermont) as well as six physiographic regions (the 

Canadian Shield, the Appalachians, the Catskills, the Hudson Highlands, the New 

England Upland, and the New Jersey Lowland). A study by Mills et al. (1996) found 

there to be 113 exotic species in the fresh waters of the Hudson River basin, including 

Trapa natans. In fact, the study placed water chestnut third on the list (after Potomogeton 

crispus and Rorripa nasturtium) of plant species to have had the most significant 

ecological impacts on the basin. In the Hudson River Basin, water chestnut is typically 

found in lakes and rivers, especially in the freshwater tidal sections. The authors suggest 

that in many regions, alterations of the environment through cultural eutrophication, 

siltation, and hydrological modifications only contributed to the success of Trapa, as well 

as many other invasive species in the basin such as Myriophyllum spicatum and Lythrum 

salicaria (Mills et al. 1996). 

Most of the exotic plants were first reported in the Hudson River Basin in the 19 th 

century. Several vectors brought in large numbers of exotics. Plants, in particular, 

originated chiefly as escapees from cultivation or in the solid ballast of ships. The high 

number of exotics in the Hudson River is probably due to the long history of human 

commerce throughout the region. Therefore, this human activity has influenced the 

number of species in the Hudson River Basin and has strongly influenced the kinds of 

species that are present (Mills et al. 1996). 

12

In the Hudson River, from the Tappan Zee Bridge to Troy, water chestnut covers 

approximately 2% of the water’s surface. Bed sizes range from 12m 2 to almost 

100,000m 2 , with an average size of 1500m 2 . These numbers again demonstrate the 

explosive propagating capability of Trapa (Hummel and Findlay 2006). 

A study by Hummel and Findlay (2006) analyzed the effects of water chestnut 

beds on water chemistry and therefore its detrimental effects on the Hudson River. 

Because under favorable conditions Trapa is capable of covering almost 100% of the 

water’s surface, it often shades out submerged aquatic species such as Vallisneria 

americana, Potamogeton perfoliatus and even the extremely invasive Myriophyllum 

spicatum. These dense beds also affect gas exchange, light penetration and invertebrate 

and fish populations. Water chestnut was also observed to be a source of dissolved 

organic carbon in the Hudson and Mohawk Rivers, which indicates that there is a direct 

correlation between rates of photosynthesis and increases in dissolved organic carbon 

(Hummel and Findlay 2006). 

The effect of aquatic plants on water velocity has direct implications for transport 

of water column constituents such as particulate matter, plankton, and detritus. Because 

sedimentation, deposition increasing as flow decreases, water chestnut beds may enhance 

settling of suspended solids thus reducing turbidity and contributing to local 

accumulation of fine sediment (Pierterse and Murphy 1990). The presence of water 

chestnut and other vegetation can also affect flow in a channel of water in one or more of 

the following ways: (1) reducing water velocities, thus raising water levels. (2) raising the 

water table on adjacent lands causing waterlogged soils and leaching of nutrients, and (3) 

changing the magnitude and direction of currents, therefore increasing the risk of local 

erosion, and interfering with other water uses (navigation, recreation) (Pierterse and 

Murphy 1990). These detrimental effects can be seen in various sites in the Hudson 

River Basin. 

The effects of large water chestnut beds on fish populations in the tidal freshwater 

Hudson River have also been studied. Fish species diversity is low under the beds, and 

the species with the largest populations are those that tolerate low dissolved oxygen 

content. Constant movement of fish into and out of the beds suggests water chestnut is 

not used continuously as protection from predators. The high plant surface area of the 

beds, however, provides habitat for various invertebrate species and significantly 

increases potential prey for fish. Very large beds, however, reduce dissolved oxygen 

which negatively affects some fish and invertebrates. Very large beds exert the greatest 

control on water quality and the two largest beds constitute 50% of the total Trapa 

coverage on the Hudson. Invertebrate and fish communities might gain from the 

separation of large beds into small disconnected beds so that they provide foraging 

habitat for fishes without creating the harmful effects of the large beds (Hummel and 

Findlay 2006). 

In many respects, the aquatic plant invasion history of the Great Lakes is similar 

to the nearby Hudson basin. Both regions have a large number of exotic vascular plants, 

fish and large invertebrates. Most are Eurasian in origin. Both areas can contribute the 

presence of exotic species to unintentional and deliberate releases. A large number of 

these species have had a significant ecological impact; however, the Hudson River 

13

eceived much higher numbers of exotic introductions in the 19 th century, while the 20 th 

century was the high point for introductions in the Great Lakes region (Mills et al. 1993). 

Methods and Attempts of Control 

Biological control possibilities were investigated for Trapa in the early 1990s. 

Surveys were conducted by the U.S. Department of Agriculture in 1992 and 1993 that 

sought natural herbivores of water chestnut in Northeast Asia (Pemberton 1999). 

Galerucella birmanica, a beetle that consumes up to 40% of water chestnut leaf tissue, 

was found to have various other plant hosts, thereby making it unsuitable for bio-control 

purposes in the U.S. Other insects that fed exclusively on water chestnut were identified 

but were found to be non-damaging to the overall health of the plant. Predators found in 

the warmer climate of India have potential but could not withstand the cooler 

temperatures of water chestnut-infested Northeast regions of the United States 

(Pemberton 1999). Other promising candidates include the beetles: Galerucella 

nymphaeae L. and Nanophyes japonica Roelofs. 

Hand removal is an effective means for eradication of smaller populations; when 

water chestnut roots are easily uplifted. Their removal and storage out of the water is 

important because floating plants can spread seeds downstream. The potential for water 

chestnut seeds to lay dormant for up to 12 years makes total eradication difficult. 

However, hand-harvesting from canoes and raking have been useful. Research has also 

attempted to reduce populations by manipulating water levels (Naylor 2003). 

For large-scale control of water chestnut populations herbicides and mechanical 

harvesting can be effective. Aquatic plant harvesting boats are often employed in 

instances where waterways are blocked. For example, mechanical harvesting in 1999 on 

the Sassafras River removed an estimated 260,000 pounds of water chestnut (Naylor 

2003). Unfortunately, mechanical harvesting boats cannot operate in some of the shallow 

areas that water chestnut can inhabit. For this reason, mechanical harvesting has been 

supplemented by hand harvesting in Maryland on the Bird and Sassafras rivers. 

The herbicide 2,4-D has been tested, and deemed safe for use by federal and state 

agencies. Used widely in the U.S., it has shown to be non-adverse on non-target species. 

Maryland and Virginia used 2,4-D in the 1960s to eradicate Eurasian water milfoil 

populations in the Chesapeake Bay. Due to public perception, the use of herbicides is 

seen as a last resort option. Integrating all possible methods for water chestnut removal 

will be the most effective course for eradication (Naylor 2003). 

The best method for control, however, remains to be prevention. Programs in 

many areas have developed systems for boat cleaning and inspection to prevent the water 

chestnut, and other invasive species, from entering a water source altogether. The “Clean 

Boats Clean Waters” Program (originally started in Michigan), has been established for 

the lakes of many states, including Otsego Lake (Cooperstown, NY) to increase public 

awareness and help prevent species from hitchhiking from lake to lake on boats. This has 

proven to be effective with cooperation and much more economical. For example, the 

14

state of Maine began a courtesy boat inspection program in 2001 to reduce the risk of 

transporting invasive species via boats, trailers and equipment (Maine Dept. of Env. 

Protection 2005). 

Herbicidal Action of 2,4-D: Biochemistry and Physiology 

Basic plant metabolism includes the production of organic compounds by 

photosynthesis, the generation of high energy chemical bonds through respiration and 

oxidative phosphorylation (in the form of ATP), and the synthesis of the basic polymers 

such as proteins, nucleic acids, starch and cellulose. Through intermediate metabolism all 

degrading and synthetic pathways are connected and this comprises a pool of small 

organic molecules. In the secondary metabolism innumerable different and specific plant 

compounds like alkaloids, pectins, lignin, growth hormones, and tannins are synthesized. 

Herbicides may interfere with any one of these pathways that contribute to the 

metabolism and growth of plants. Metabolic pathways specific for plant tissues contain 

most of the known sites of action, (e.g. photosynthesis, carotenoid synthesis, and specific 

plant regulatory systems). Therefore, in general herbicides are relatively nontoxic to 

animals (Fedtke, 1992). 

The herbicide 2,4-D is a highly selective herbicide that is toxic to broad leafed 

plants but less harmful to grasses. The formulation used on the water chestnut population 

is a butoxyethyl ester of 2,4-D, also termed Aqua-Kleen® by Cerezagri-Nisso. Aqua- 

Kleen® has been used successfully for selective control of noxious aquatic plants 

including water milfoil (Myriophyllum spicatum), coontail (Ceratophyllum demersum), 

and spatterdock (Nuphar polysepala) for more than two decades (Aqua-Kleen 2005). 

Known as a hormone weedkiller, the herbicide is an aryloxyalkanoic acid or a “phenoxy 

herbicide”. These chemicals have complex plant interactions resembling those of auxins 

(growth hormones). Once absorbed, 2,4-D is translocated within the plant and 

accumulates at the growing points of roots and shoots where it stimulates growth. This 

herbicide has low soil absorption, a relatively short half-life and a high potential for 

leachability. Aqua-Kleen® can be used in specific areas without impacting untreated 

areas of the lake or water body. While some formulations of 2,4-D, particularly esters, are 

highly toxic to fish, the compounds used for this project are not (Aqua-Kleen 2005). 

Aquatic invertebrates do not in general seem to be sensitive to the herbicide and toxicity 

to birds is low (Herbicide 2,4-D 2006). In 2007, Navigate® was applied, manufactured 

by Advantis Technologies. Navigate® is chemically identical to AquaKleen. 

While auxin herbicides are known to interfere with many plant processes 

(depending on concentration) such as the synthesis of hormones, oxidases, peroxidases, 

and nucleic acids, the effectiveness of 2,4-D is particularly known for its interaction with 

the gaseous plant hormone ethylene. Ethylene is continuously produced by most plant 

tissues. When 2,4-D is applied, this production is greatly increased. Since some effects of 

auxins and of ethylene are similar, it was originally thought that auxins might act via the 

induction of excessive ethylene production. Then it was discovered that firstly, the effects 

on nucleic acid metabolism are not mediated by ethylene. (Fedtke 1992). 

15

One important example is the abscission (shedding) of leaves and fruits. In the 

presence of externally applied excess amounts of auxins and the artificial production of 

high concentrations of ethylene, fruit ripening and abscission are promoted. The surface 

of the treated fruit, however, remains fresh and undergoes color changes in response to 

the rejuvenating effect of the auxin. Naturally, abscission is initiated by indoleacetic acid 

(or IAA), a plant hormone promoting elongation of the stems and roots, is no longer 

produced in the aged leaves and ethylene is then produced in the abscission zone. 

Therefore, the treatment of a leaf with 2,4-D produces a green treated zone and a yellow 

surrounding ring. The explanation being that, 2,4-D at the site of application, keeps the 

tissue juvenile whereas the mobile ethylene produced in response to the treatment 

diffuses out and softens and ages the surrounding tissue. In summary, auxin and ethylene 

are in many respects antagonistic (in growth, abscission and ripening). Ethylene might 

also be responsible for many of the sublethal effects observed after 2,4-D treatments 

(Fedtke 1992). 

16

Site Description 

METHODS 

Located in Oneonta, Otsego County, NY, and owned by Louis Blasetti, the 40 

acre wetland is a formerly agricultural bottomland of an old Susquehanna River oxbow, 

now inundated because of beaver activity. As seen in Figure 3, it drains directly into the 

Susquehanna. It is essentially an old Brownfield undergoing 20 years or so of ecological 

succession. The mean depth is about 0.3 meters; and the maximum depth is about 1.3 

meters. At least 90% of the shoreline has been disturbed in the past by filling from home 

construction along Lower Oneida Street, the right-of-way of Lower River Street and 

filling by the Delaware and Hudson rail yards. Practically all is overgrown with 

emergent vegetation and larger wetland plants including dogwoods (Cornus amomum, C. 

sericea), alders (Alnus incana), honeysuckle (Lonicera sp.), buckthorn (Rhamnus 

cathartica) and a diversity of pioneer hardwoods. Purple loosestrife (Lythrum salicaria) 

is the dominant emergent along about 25% of the shoreline. 

Public access is available all along Lower River Street, but is difficult because of 

tangled vegetation and the nature of the filled bank. Private access is facilitated from 

paths behind (east of) homes along Oneida Street. There is effectively no shore 

development other than a few places to stash canoes or john boats behind homes in the 

woods. The property owner involved with our control activities professes great 

largemouth bass fishing; a diversity of waterfowl has been observed. 

Plant and Water Sampling 

Several activities were undertaken to characterize the relatively unexplored 

Blasetti wetland. A plant survey was performed to collect and identify the local 

submerged aquatic plant species (TABLE 2). Water samples were extracted from the 

wetland about every two weeks throughout the growing seasons of 2006 through 2008 

(May through September). Samples were taken in two consistent places, at the outlet at 

the southwest corner of the wetland and at its deepest area (FIG 3). A water sample of 

125mL was collected at each site in an acid-washed bottle, preserved with 0.8mL of 

sulfuric acid (H2SO4) and analyzed using a Lachat QuikChem FIA+ Water Analyzer. 

This machine tested for total phosphorus using ascorbic acid following a persulfate 

digestion (Liao and Martin 2001), total nitrogen using the cadmium reduction method 

(Pritzlaff 2003) following the peroxodisulfate digestion as described by Ebina et al. 

(1983), ammonia using the phenolate method (Liao 2001), and for nitrate+nitrite nitrogen 

using the cadmium reduction method (Pritzlaff 2003). The nutrients quantified during 

testing were ammonia, nitrate, total nitrogen and total phosphorous. The results of 

analysis of the water samples were then compared to determine the effects of 

eutrophication and nutrient loading on the chemistry of the water. 

17

Plant Biomass and Distribution 

The aquatic plant distributions and abundances are described using a sampling 

technique developed at the Ponds Research Center at Cornell University called 

PIRTRAM, or the Point Intercept Rake Toss Abundance Method (Lord and Johnson 

2006). The system is an adaptation of a technique used by the Army Corp of Engineers 

(ACOE) to quantify populations of terrestrial plants. The rake toss sampling method is 

relatively simple. By placing a UTM NAD27 100x100m grid on a map of the wetland, a 

set of predetermined locations is determined. These points serve as the plant sampling 

collection sites, totaling 23 sites that fell upon the surface of the water. Then using a 

handheld GPS out on the water, each location is found and at each spot a double sided 

iron garden rake is thrown the length of an attached 10m rope from the boat. After the 

rake is allowed to sink, the rope is slowly pulled back to the boat making sure to allow 

the rake to drag along the bottom. The rake and all attached plants are then lifted into the 

canoe and an overall plant abundance is determined using the Cornell Abundance Scale 

(TABLE 1). After the plants are removed from the rake and identified, they are separated 

by species and each species was given a plant abundance category based on the same 

scale. The scale is determined by using associated dry weights to estimate biomass. 

All data was recorded on spreadsheets on the water. The rake was thrown and 

retrieved three times at each GPS location (Lord and Johnson 2006) followed by plant 

identification and quantification. The recorded data were then analyzed using Microsoft 

Access and Excel and ArcGIS software. Because the Cornell Abundance Scale displays 

a range of dry weights, the largest biomass weight for each category was used for ease of 

calculation. 

TABLE 1. Cornell Plant Abundance Scale with associated dry weights 

Raw Data Code Rake Toss Retrieval Associated Dry Weight 

Z= zero plants no plants on rake 0g 

T=trace plants fingerful on rake 0.0001- 2.000g 

S= sparse plants handful on rake 2.0001- 140.00g 

M= medium plants rakeful of plants 140.001- 230.000g 

D= dense plants difficult to bring in boat 230.001- 450.000g 

Herbicide Application 

In an effort to prevent further dispersion of this Oneonta water chestnut 

population, it was thought best by project advisors to use a chemical weedkiller. In past 

years, hand-pulling efforts were ineffective. Donated monies were used to purchase a 

quantity of 2,4-D herbicide and to develop a work plan. The herbicide was applied by the 

airboats and trained applicators of Allied Biological Company, NJ, on August 23, 2006 

and July 25, 2007 and July 26, 2008. The first application involved the use of 150 

pounds of herbicide per acre while 200 pounds of the same herbicide were applied per 

acre in the second and third applications. All necessary permits were issued by the New 

York State Department of Environmental Conservation. 

18

FIGURE 3. Aerial map of Blasetti wetland indicating location of both water sampling 

sites (blue points) and the 23 plant sampling sites (yellow points, 100x100m grid). 

Numbers in red indicate those sites that received herbicide treatment. Also note close 

proximity of wetland to Susquehanna River. 

19

Plant Survey Results 

RESULTS 

A plant survey was performed in the late spring of 2006. Voucher specimens of 

each were collected, identified and reported in TABLE 2. All species reported in the table 

are obligate wetland species, meaning these plants are found in wetland habitats 99% of 

the time. It should also be noted that curly pondweed (P. crispus) is also a non-native 

plant, though it did not display the same invasive capabilities as the water chestnut. It was 

found to be in lesser quantities than coontail (C. demersum) and waterweed (E. 

canadensis), both native. It could be speculated that plant diversity in this wetland is 

relatively low compared to other aquatic communities of its size when considering its age 

and that decreased species richness is common in wetlands that are found in disturbed 

areas and also those with invasive populations (Houlahan and Findley 2004). 

TABLE 2. List of aquatic plant species inhabiting Blasetti Wetland, 2006-2007. 

Source: USDA NRCS Plants Database http://plants.usda.gov. 

Species ID Common Name Scientific Name 

1. Cd Coontail Ceratophyllum demersum 

2. Ec Waterweed Elodea canadensis 

3. Pa Water smartweed Polygonum amphibium 

4. Pc Curly pondweed Potamogeton crispus 

5. Pp Sago pondweed Potamogeton pectinatus 

6. Tn Water chestnut Trapa natans 

7. Wc Water meal Wolffia columbiana 

8. Lm Duckweed Lemna minor 

9. Bs Watershield Brassenia schreberi 

Water Quality Analysis Data 

Tests for Ammonia and Nitrates, Total Nitrogen and Total Phosphorous were run 

by Matthew Albright at the SUNY Oneonta Biological Field Station. It should be noted 

that in June of 2006, there was massive flooding of the Susquehanna River. The water 

level of the wetland rose approximately one meter and within two weeks returned to 

normal level. 

20

TABLE 3. Nutrient concentrations in Blasetti Wetland, May through September 2006. 

DEEP designates samples collected from the deepest part of the wetland (about 1.5 

meters) and OUTLET indicates samples collected near the outlet in the southwest corner 

of the wetland, these areas are indicated on FIG. 3. 

Ammonia NO3 + NO2 Total Nitrogen Total Phosphorus 

2006 Samples (mg/l) (mg/l) (mg/l) 

(mg/l) 

22-May 0.79 2.331 8.62 2.512 

30-May 0.61 3.064 11.50 2.952 

15-Jun 0.64 2.393 8.19 2.411 

10-Jul 0.00 0.162 0.63 0.038 

21-Jul 0.01 0.136 0.63 0.033 

22-Aug 0.08 0.010 0.62 0.079 

DEEP 3-Sept 0.02 0.010 0.40 0.025 

OUTLET 3-Sept 0.08 0.008 0.46 0.046 

DEEP 15-Sept 0.05 0.013 0.40 0.037 

OUTLET 15-Sept 0.08 0.008 0.42 0.034 


Below detection (BD) records are those nutrient levels too small to be recorded. 

Ammonia NO3+NO2 Total Nitrogen Total Phosphorous 

2007 Samples (mg/l) (mg/l) 

(mg/L) 

(mg/L) 

OUTLET 5/23 0.16 0.080 0.61 0.059 

5-Jun 0.09 BD 0.49 0.081 

Jun-31 BD 0.010 0.31 0.018 

13-Jul BD BD 0.39 0.046 

31-Jul 0.28 0.049 0.81 0.012 

14-Aug 2.34 BD 1.57 0.509 

6-Sep 0.01 BD 0.44 0.045 

30-Sep 0.08 0.289 0.94 0.041 

BW DEEP 5/23 0.14 BD 0.43 0.065 

5-Jun 0.05 BD 0.56 0.081 

Jun-31 BD BD 0.36 0.035 

13-Jul 0.41 BD 0.81 0.027 

31-Jul 0.08 BD 0.54 0.010 

14-Aug 0.02 BD 0.63 0.084 

6-Sep 0.11 BD 0.53 0.062 

30-Sep 0.07 0.042 0.68 0.041 

21


Below depth (BD) records are those nutrient levels too small to be recorded. 

Ammonia NO3+NO2 Total Nitrogen Total Phosphorous 

2008 Samples (mg/l) (mg/l) 

(mg/L) 

(mg/L) 

OUTLET May 26 0.15 0.070 0.54 0.065 

June 16 0.08 BD 0.62 0.045 

July 2 BD BD 0.41 0.022 

July 22 BD BD 0.42 0.037 

Aug 1 BD 0.058 0.63 0.012 

Aug 24 0.05 BD 0.75 0.074 

Sept 4 BD 0.049 0.44 0.068 

Sept 19 0.05 0.162 0.94 0.040 

DEEP May 26 0.12 BD 0.71 0.075 

June 16 0.06 BD 0.35 0.076 

July 2 BD BD 0.63 0.028 

July 22 BD 0.010 0.81 0.033 

Aug 1 0.07 BD 0.36 0.041 

Aug 24 0.03 BD 0.48 0.048 

Sept 4 0.09 0.030 0.57 0.067 

Sept 19 0.08 BD 0.84 0.059 

Plant Biomass and Distribution Data Results 

The Point Intercept Rake Toss Relative Abundance Method (PIRTRAM) data 

were collected in June of each year. The grid used was 100x100m 18T049 UTM datum. 

Three records per site were included using the predetermined dry weight scale in Table 1. 

A paired t-test comparing non-target biomass in 2006 (3671.34g/m 2 ) and 2007 

(2390.35g/m 2 ), found that the decrease was not significant (p= 0.25). Similar results 

were calculated when comparing 2007 (2390.35g/m 2 ) and 2008 (2279.57g/m 2 ) data; the 

decrease was not significant (p=0.46). Tables 6 and 7 summarize the PIRTRAM data. 

Tables 8 and 9 include PIRTRAM data for plant biomass comparisons in areas sprayed 

with herbicide and those not sprayed with herbicide. 

22

TABLE 6. PIRTRAM data summary by species, 2006- 2008. 

2006 2007 2008 

Species Biomass 

(g/m 2 #Sites Biomass 

) 

(g/m 2 # Sites Biomass 

) 

(g/m 2 ) # Sites 

1. Cd 1286.21 21 469.21 13 418.63 11 

2. Ec 624.14 8 636.70 11 690.38 11 

3. Pa 873.54 7 252.21 2 215.52 2 

4. Pc 541.62 13 787.52 16 685.04 14 

5. Pp 304.03 11 230.5 4 211.74 4 

6. Tn 1118.04 11 1590.39 9 987.19 6 

7. Lm 2.10 7 3.01 12 4.30 13 

8. Wc 1.92 10 1.75 8 2.50 10 

9. Bs 37.78 4 9.45 1 51.46 6 

TOTAL 4789.38 88 3980.24 76 3266.76 77 

23

TABLE 7. Total plant biomass by collection site. A retrieval of benthic algal species is 

indicated by the + symbol. *Sites that received herbicide application. 

Site 2006 Total Biomass 2007 Total Biomass 2008 Total Biomass 

(g) 

(g) 

(g) 

1 226.35 + 240.23 215.36 

2 75.64 + 110.52 + 165.57 + 

3 196.36 + 94.75 + 104.15 

4 222.78 + 23.05 + 25.0 + 

5 230.31 + 137.02 117.20 

6 229.19 143.25 133.19 

7 47.51 2.15 120.61 

8 50.10 + 137.36 + 108.13 

9 96.43 + 48.99 + 149.28 + 

10 49.47 + 1.03 + 29.24 + 

11 262.35 + 137.05 + 111.83 

12 443.30 226.10 201.18 

13 170.26 426.12 245.97 

14 200.07 129.61 56.72 

15 82.64 + 124.22 + 126.04 

16 186.59 + 150.62 + 200.77 + 

17 205.24 + 3.40 + 98.43 

18* 57.44 + 131.06 89.23 

19* 55.76 141.64 65.04 

20* 136.04 45.97 186.62 

21* 441.68 437.85 272.59 

22* 468.36 474.62 197.71 

23* 508.72 416.49 246.9 

TOTALS 4789.38 3927.79 3266.76 

24

Effects of 2,4-D Herbicide: Photos 

The 2,4-D herbicide was applied to beds of healthy water chestnut that are seen in 

Figure 4 to be lush, dense and bright green. Figures 5 and 6 document the degradation of 

the water chestnut over three weeks time after the application. Roots turned brown and 

lost all rigidity and form, petioles began to disassociate from rosettes, leaves lost color 

and form, and plants began to sink in the water column. Four weeks after the application, 

the plants were observed growing new shoots from degraded, old roots (Figure 7). All 

photos were taken by W. Eyres in 2006 on site. 

FIGURE 4. Chestnut bed prior to FIGURE 5. Bed two weeks after 

herbicide application; plants dense and green. Application; plants begin to sink and degrade 

FIGURE 6. Three weeks post application; FIGURE 7. Four weeks; plants showing regrowth 

plants brown almost completely disassociated of new shoots from old roots 

25

TABLE 8. Biomass of non-target species in areas sprayed with herbicide (sites 1-17, FIG 

3) and not sprayed with herbicide (sites 18-23) in all years. Data displayed in Figure 8. 

NON TARGET 

Species Biomass Sprayed Non Sprayed Total 

(g) 

Area Area 

2006 656.32 3014.66 3670.96 

2007 144.83 2047.91 2192.71 

2008 121.53 2158.07 2279.57 

FIGURE 8. Herbicide effect on non-target species biomass in sprayed and non-sprayed 

areas. Paired t-Test results showed that between the years of 2006 and 2007 there was a 

significant difference (p=0.03) between the sprayed and non-sprayed areas. Treatment 

effects were similar comparing 2007 and 2008 data (p=0.01). 

26

TABLE 9. Biomass of target (T. natans) species in areas sprayed (sites 1-17, FIG 3) with 

herbicide and not sprayed with herbicide (sites 18-23) in all years. Data displayed in 

Figure 9. 

TARGET 

Species Biomass Sprayed Non Sprayed Total 

(g) 

Area Area 

2006 1011.70 106.34 1118.04 

2007 1502.83 87.56 1590.39 

2008 936.59 50.60 987.19 

Dry Weight Biomass (g) 

1600 

1400 

1200 

1000 

Target Species Biomass in Sprayed and 

Non-Sprayed Areas 

800 

600 

400 

200 

0 

1 2 

Sprayed vs. Non Sprayed Areas 

FIGURE 9. Herbicide effect on target species biomass in sprayed and non-sprayed areas. 

Paired t-Test results showed that between the years of 2006 and 2007 there was a 

significant difference (p=0.04) between the sprayed and non-sprayed areas. Treatment 

effects were similar comparing 2007 and 2008 data (p=0.03). 

27 

2006 

2007 

2008

Water Chestnut Population Demographics 

PIRTRAM Data 

In the past decade, orthoimagery, aerial photography and raster and vector 

imagery used in Geographic Information Systems have become a powerful monitoring 

tool in biological research. Using a 2001 GIS map of the wetland from the Otsego County 

Online Mapping website, a 2004 map from NRCS, and 2006 orthoimagery from the New 

York State Clearinghouse GIS website, it was possible to calculate and extrapolate 

information about the water chestnut population growth over the past five years (FIGS 

10-14). The bright color of the dense chestnut beds is easily seen on the surface of the 

water on these maps with clear 1 meter resolution. It should be noted that these 

calculations do not include information about the random individual chestnuts found 

throughout the wetland, but only the main bed confined to the northwest corner. Also, it 

was not possible to locate information about which month of the year these photos were 

taken, therefore it should not be assumed that the area of chestnut coverage is at its full 

extent for any given year. The numbers reported can only be used as minimum values, 

but nevertheless increases are demonstrated. 

As previously noted, one individual water chestnut seed may produce one to five 

rosettes in a single growing season, and each rosette is capable of producing 3-20 ramets, 

and each ramet producing its own set of seeds. Because these numbers are not consistent 

and show such a range of productivity even on one individual, averages were used and 

therefore assumptions made. At full growth and biomass, densities were recorded at 

approximately 15 rosettes/m 2 . Because the number of individual chestnut plants could not 

be known, numbers of rosettes were used as reproductive individuals (because each 

produces its own set of seeds). Also, rosettes are easily distinguishable and quantifiable 

from the surface. 

TABLE 10. PIRTRAM biomass of T. natans, 2006-2008 (23 sampling sites). 

T. natans PIRTRAM Biomass (g) # Site Occurrences 

2006 1118.4 11 

2007 1590.3 9 

2008 987.1 6 

TABLE 11. Population size of T. natans over five years, Blasetti Wetland. 

Rosette density averaged approximately 15/m 2 . (Aerial photo/ GIS for 2001 and 2004). 

See Figures 10-14 for GPS maps. 

Rosettes 2001 2004 2006 2007 2008 

Perimeter (m) 464.36 588.95 779.94 854.21 781.04 

Acres (ac) 1.75 2.54 3.64 3.98 2.16 

Total Area (m 2 ) 7118.47 10315.5 14762.3 16106.5 8797.61 

28

Using the data from Tables 6 and 9 and the models and formula of exponential 

population growth (Nt= No t ) (Gotelli 1995) it was possible to calculate changes in 

growth rates and spread. For these calculations the water chestnut population was 

assumed to be a discrete population in which the plants die off each year and return as a 

new generation the following spring. Therefore, considering the water chestnut as an 

annual made it appropriate to use lambda ( ) for the population equations, instead of r 

(where r= dN/dt). In the three years between the 2001 (FIG. 10) and 2004 (FIG. 11), 

approximately 16,000 new rosettes were added to the bed (TABLE 11). With the addition 

of these new rosettes came an increasing rate of growth; more chestnuts were producing 

that many more seeds (more than 1:1 replacement). Therefore, in the years between the 

2004 and the 2006 maps, over 33,000 new rosettes were added to the population. Using 

these rates of growth, it was then possible to calculate the time that would be needed for 

the bed to cover the entire 40 acre surface of the wetland, which was determined to be 

just less than 22.8 years (or more than 2.5 million rosettes). 

The growth rate ( ) after one herbicide treatment was calculated to be 1.42 per 

year (see calculations below and TABLE 10). When compared to the negative growth 

rate after the second treatment (0.62 per year), it can be suggested that there may be a lag 

time in the value of 2,4-D in the first two consecutive application years. It could also 

indicate that the first treatment of herbicide was applied too late in the first season and 

was not as effective. The growth rate after two herbicide applications was substantially 

lower, less than 1, meaning that there was a decline in chestnuts in 2008. If the reduced 

growth rate was continued with the yearly herbicide application, it was determined using 

the formula Nt= No t that the 221,430 rosettes in the pond in 2006 would decrease every 

year at the rate of 0.62 per year. This calculation assumes that the 0.62 growth rate is 

linear. Realistically, this growth rate will actually decline further with each subsequent 

herbicide application. Even using the linear rate, in year five, 2013, there would be 

approximately 20,286 rosettes in the wetland. At this stage, continued herbicide 

applications would not be necessary as a dedicated crew of 10 people could hand pull this 

amount easily in one day; each person removing about 2,000 rosettes each spring. Project 

management, however, must keep in mind that continued and consistent hand removal 

each spring will be necessary. Seeds can remain viable for over ten years and could quite 

feasibly regenerate large populations. 

2006-2007: rate of growth after 1 treatment 2,4-D, data from TABLE 10. 

= N2/N1= 1590.39/1118.04= 1.42 per year 

2007-2008: rate of growth after 2 treatments of 2,4-D, data from TABLE 10. 

= N2/N1= 987.19/1590.39= 0.62 per year. 

29

Water Chestnut Populations Demographic 

Orthoimagery/ GIS Data 

Figures 10-14 are a visual demonstration of both the water chestnuts dispersal 

capabilities as well as the changing demographic from the years of 2001 and 2008. Please 

note that Figures 10 and 11 are estimated from aerial photography and are not as accurate 

as the following maps of 2006-08 in which chestnut beds were surveyed on the water 

using GPS technologies. See Tables 10 and 11 for actual numbers referencing the 

population size changes. 

FIGURE 10. Approximate extent of water chestnut bed in 2001. 

Source: Otsego County Online Mapping, 2001. 

30

FIGURE 11. Approximate extent of water chestnut bed in 2004. 

Source: Natural Resources Conservation Service, 2004. 

FIGURE 12. Surveyed extent of water chestnut bed by GPS in 2006. 

31

FIGURE 13. Surveyed extent of water chestnut beds by GPS in 2007. 

FIGURE 14. Surveyed extent of water chestnut beds by GPS in 2008. 

32

Nutrient Concentrations 

DISCUSSION 

Unfortunately, the nutrient concentrations in the wetland can only be discussed in 

the context of the massive flooding of the Susquhanna River in June 2006. The first three 

samples before the flooding reported very high concentrations of nutrients and then a 

sudden decline, which corrosponds exactly with the time of the flooding (TABLE 3). It 

appears that this flushed the system and the levels remained low throughout the 

remainder of the project (FIG 15 and 16). Unfortunately, this could suggest that the 

effects of the flood were so substantial that they obscured impacts of the water chestnut 

population or its management on nutrient concentrations. 

There was a non reliable datum reported from the water testing. The ammonia 

concentration reported for 14 Aug 07 is higher than total nitrogen, which is impossible 

beause all nitrogen components are to be converted to nitrate, which is what is then 

analyzed. This could be due to incomplete conversion of ammonia during the digestion. 

Therefore, total nitrogen must have been higher that what was reported (TABLE 4). 

The concentrations of phosphorus in the first three samples in 2006 are nearly 50 

times the concentrations normally found in other areas of the Susquehanna; nitrogen 

concentrations were about 10 times greater (TABLE 3). However, when disregarding the 

very high concentrations recorded for the nutrients on 22 May, 30 May and 15 June of 

2006, the results of the water nutrients of the wetland are quite low when compared to 

other portions of the river. This comparison supports the speculation that the flood in 

June 2006 flushed the wetland and kept conentrations low. Nutrient data collected in 

more southern areas of the Susquehanna watershed in streams of Towanda, Marietta, and 

Conestoga (all in Pennsylvania) taken over the past 20 years have shown to have 

relatively high and consistent nitrogen levels, while phosphorous levels have been 

continuously improving (decreasing) (Edwards 1996). In comparison, nitrogen levels in 

the Blasetti wetland are extremely low, less than half of the nitrogen at the other sites in 

the watershed. Phosphorus levels are also low, approximately 0.01-0.03 mg/L less 

(TABLE 12). 

Although there are no water nutrient data from the wetland prior to the start of the 

project or before the chestnut invasion, it can only be assumed that the nutrient 

concentrations were high (a typical wetland charateristic) and consistent with the data 

reported for May and June 2006. It can be speculated that the increased nutrient levels 

only enhanced the suitability of the wetland for the infestation of the invasive water 

chestnut and curly pond weed. 

In addition, establishing causality of nutrient and sediment trends is not a simple 

task, given the complex interactions of human acitvites. In the Susquheanna River Basin 

over the past 30 years trends indicate that phosphorous and sediment concentrations have 

been holding steady or improving, but nitrogen levels are of concern. The cummulative 

effect of phosphate detergent bans, agricultural best management practices (BMPs) and 

33

sewage treatment plant upgrades may have played a role in the decreasing trends in 

phosphorus. Population growth and greater intensities of land use may have contributed 

to increasing (deteriorating) trends in nitrogen. Several studies by Ott (1990) have found 

that the streams of the Susquehanna River Basin have significantly higher nutrient levels 

than other New York water systems. The reduction of nutrients from the Susquehanna 

River Basin has been part of a multi-state effort to reduce the controllable nutrient 

loading reachig the Chespaeake Bay Watershed. Reduction of nitrogen and phosphorus 

loads is expected to increase dissolved oxygen levels (Edwards 1996). 

TABLE 12. Nutrient concentrations: Towanda, Marietta and Conestoga data are a 20 year 

average for nitrogen and phosphorus (Edwards 1996). * Indicates northern most site. 

Note that Blasetti 2006 averages exclude data recorded before the flood. 

Location Nitrogen (mg/L) Phosphorus (mg/L) 

Blasetti Wetland, 2006 0.513 0.042 



Towand, PA * 1.29 0.06 

Marietta, PA 1.30 0.07 

Conestoga, PA 1.32 0.09 

Nitrogen Concentration (mg/L) 

14 

12 

10 

8 

6 

4 

2 

0 

May- 

September 

Total Nitrogen Concentrations 

Growing Season Months 

FIGURE 15. Total nitrogen concentrations in 2006, 2007 and 2008, Blasetti Wetland. 

Data from TABLES 3, 4 and 5. 

34 

2006 

2007 

2008

Phosphorous Concentration (mg/L) 

3.5 

3 

2.5 

2 

1.5 

1 

0.5 

0 

Total Phosphorous Concentrations 

May- 

September 

Growing Season Months 

FIGURE 16. Total phosphorous concentrations in 2006, 2007 and 2008. Blasetti 

Wetland. Data from TABLES 3, 4, and 5. 

Plant Biomass 

Several statistical tests were performed to elucidate any significant changes in the 

biomass and distribution of the plants in the wetland. The first comparison was made 

between the overall plant biomass for the three project years. The data shows that in 2007 

overall plant biomass in the wetland had decreased about 800g, while the biomass of the 

water chestnut increased about 480g. However, the results from a paired t-test show that 

the decrease in non-target plant biomass was not significant in 2007 or 2008 with p 

values of 0.25 and 0.46 respectively (TABLE 6). The reported increase in water chestnut 

biomass in 2007 is not surprising given its expansion potential, and it can be speculated 

that the increase could have been much more substantial had the herbicide not been 

applied. 

Other findings were illustrated when comparing the biomass changes of target 

(water chestnut) and non-target plants in the herbicide sprayed areas and non-sprayed 

areas. Herbicide sprayed areas were designated by PIRTRAM sites 18, 19, 20, 21, 22, 

and 23. Despite an overall non-significant decrease in biomass, there was a significant 

decrease of nontarget species in the areas that were sprayed with the herbicide (p=0.01). 

Similar results were reported for 2008; there was a significant (p=0.016) overall decrease 

in non-target plant biomass (TABLE 8 and FIG 8). A decrease in nontarget plant biomass 

was expected due to the broad ranging effectiveness of the chosen herbicide, although 

when comparing the overall nontarget species biomass from 2007 to 2008, there was only 

35 

2006 

2007 

2008

a 50g decrease, indicating that about 60% of the overall loss of biomass (389.94g) within 

the wetland was from the target area (TABLE 9, FIG 9). 

The increased biomass of the water chestnut from 2006 to 2007 (TABLE 6) with 

a growth rate of 1.42 per year indicate a possible lag time in 2,4-D effectivness, though 

increases may have been higher had the herbicide not been applied. Also, the water 

chestnuts were recorded in fewer sampling sites in 2007 and 2008. The third herbicide 

treatment showed more effective results, with a substantial decline in water chestnut 

biomass of 603.2g, and an overall negative reduction in the growth rate ( = 0.62 per 

year). These results are quite promising for reducing propagule pressure enough to 

prevent infestation in the river. 

Another promising factor was that the population appeared from the surface to 

have dispersed into five smaller beds possibly indicating that rosette density and clonal 

propagation had declined or that fewer seeds germinated. In addition, as of 15 August 

2008, the water chestnut has not been found in the drainages or surrounding areas in the 

Susquehanna River. 

CONCLUSIONS/ RECOMMENDATIONS 

The three year scope of this study provides only baseline information about the 

effectiveness of the 2,4-D herbicide, its impacts on water, and the future of the water 

chestnut in the wetland or the Susquehanna River Basin. It is unfortunate that the timing 

of the flood interfered with the water nutrient analysis of the wetland. It will be 

interesting to see how the wetland recovers and if any impacts of the herbicide and 

eutrophication will be found. It is a well established fact that microbes will degrade and 

use the herbicide in their life cycles and thus become part of the food web (Schmidt 

1999). It is no doubt that the decomposer community will be impacted because of the 

herbicide and this may in turn reduce the release of nutrients. In addition, the herbicide 

damage to the plants may provide a “carbon supplement” for microbial organisms, which 

may also prevent nutrient release into the system (Pers.Comm with Zedler 2008). 

After the third treatment, the results of this project showed that the herbicide had 

an effect on the water chestnut population, as a decrease in plant biomass and population 

growth rate were reported. The invasive population now has a declining growth rate of 

0.62 per year. At this rate, after five years there would be fewer than 20,000 rosettes of 

water chestnut left in the wetland, assuming other factors remain constant. Hand-pulling 

efforts showed to be a continued success, as the number of sites the chestnuts were found 

decreased from the original 11 in 2006 to now only 6 sites in 2008. If the decreased rate 

continues, the herbicide will decrease the population size enough so that the hand-pulling 

efforts of the dedicated volunteers every spring will be the only control measure needed. 

Because of the size, extent and location of the population, eradication may be 

impossible due to the long-lived propagules and heartiness of the plant. But it must be 

kept in mind that although eradication would be an ideal goal, the ultimate purpose of this 

36

project is to keep the water chestnut from escaping the wetland and taking hold the 

Susquehanna River where its impacts would be widespread. As shown with other 

infestations (i.e. the Chesapeake Bay), control of such invasives capable of rapid 

establishment and spread involves substantial long term commitment from researchers 

and financial supporters. With this project and as with any case of invasive plant species, 

early detection and rapid response always result in a higher probability of control and/or 

eradication. 

However, while preventing the spread of this population does not ensure that the 

seeds of the current population could not be carried or travel to the Susquehanna and 

begin a harmful infestation, failing to do so will virtually guarantee its further spread. 

Therefore, further study and continuation of control efforts are necessary. Close 

monitoring of the culvert and drainage areas must be sustained. The present study 

supports a prediction that with the continual timely application of the 2,4-D herbicide, the 

Blasetti wetland water chestnut population will continue decline in the coming years. 

37

REFERENCES 

Aqua-Kleen. 2005. Cerexagri: Aquatic habitat management. [Online]. Accessed 14 Nov 

2006. http://www.cerexagri.com/aquatic/aquakleen.asp 

Countryman, W.M. 1970. The history, spread and present distribution of some immigrant 

aquatic weeds in New England. Hyacinth Control Journal 8(2): 50-52. 

Dick, T.J. 2004. Water Chestnuts: Aquatic Exotic Animals and Plants. [Online]. 

http://www.iisgcp.org/EXOTICSP/waterchestnut.htm. 

Ebina, J., T. Tsutsui, and T. Shirai. 1983. Simultaneous determination of total nitrogen 

and total phosphorus in water using peroxodisulfate oxidation. Water Res. 17(12): 

1721-1726. 

Edwards, R.E. 1996. Trends in nitrogen, phosphorus and suspended sediment in the 

Susquehanna River Basin, 1974-1993. Susquehanna River Basin Commission. 

Pennsylvania Department of Environmental Resources: 10-14. 

Elton, Charles. 1958. The Ecology of Invasions by Plants and Animals. John Wiley and 

Sons, New York. 437-461. 

Fedtke, C. 1992. Biochemistry and Physiology of Herbicide Action. Springer-Verlag, 

New York. 139-145. 

Frodge, J.D. and G.L. Thomas, and G.B. Pauley. 1990. Effects of canopy formation by 

floating and submergent aquatic macrophytes on the water of two shallow Pacific 

Northwest lakes. Aquatic Botany 38: 231-248. 

Gordon, D.R. 1998. Effects of invasive, non-indigenous plant species on ecosystem 

processes: Lessons from Florida. Ecological Applications 8(4): 975-989. 

Gotelli, N. 1995. Primer of Ecology. Sinauer Associates, Massachusetts: 1-30. 

Groth, A.T, L.L. Doust, and J.L. Doust. 1996. Population density and module 

demography in Trapa natans, and annual, clonal aquatic macrophyte. American 

Journal of Botany 83(11): 1406-1415. 

Hellquist, C.B.. 1997. A guide to invasive non-native aquatic plants in Massachusetts. 

Massachusetts Department of Environmental Management, Boston, MA. 

Herbicide 2,4-D. 2006. Dow AgroSciences LLC. [Online]. Accessed 28 Oct 2006. 

http://www.dowagro.com/ca/prod/frontline-2.htm 

38

Heywood, V.H. 1993. Flowering plants of the world. Oxford University Press, New 

York: 47. 

Houlahan, J.E. and C.S. Findlay. 2004. Effect of invasive plant species on temperate 

wetland plant diversity. Conservation Biology (18) 4: 1132-1138. 

Hummel, M. and S. Findlay. 2006. Effects of water chestnut (T. natans) beds on water 

chemistry and in the tidal freshwater Hudson River. Hydrobiologia 559: 169-181. 

Levine, J.M., P.B. Adler, and S.G. Yelenik. 2004. A meta-analysis of biotic resistance to 

exotic plant invasions. Ecology Letters 7: 975-989. 

Les, D.H. and L.J. Merhoff. 1999. Introductions of non-indigenous aquatic vascular 

plants in southern New England: a historical perspective. Biological Invasions 1: 

281-300. 

Liao, N. 2001. Determination of ammonia by flow injection analysis. 

QuikChem®Method 10-115-01-1-F. Lachat Instruments. Loveland, Colorado. 

Laio, N. and S. Marten. 2001. Determination of total phosphorus by flow injection 

analysis colorimetry (acid persulfate digestion method). QuikChem®Method 10- 

115-01-1-F. Lachat Instruments. Loveland, Colorado. 

Lockwood, J.L., P. Cassey, and T. Blackburn. 2005. The role of propagule pressure in 

explaining species invasions. Trends in Evolution Ecology 20 (5): 223-228. 

Lord, P.H. and R.L. Johnson. 2006. Point Intercept Rake Toss Relative Abundance 

Method Software and User Guide. Cornell University Research Ponds: 1-7. 

Maine Department of Environmental Protection 2005. Maine Courtesy Boat Inspections. 

2005. [Online]. http://www.maine.gov/dep/blwq/topic/invasives/inspect.htm 

Mills, E.L., J.H. Leach, J.T. Carlton, and C.L. Secor. 1993. Exotic species in the Great 

lakes: A history of biotic crises and anthropogenic introductions. Journal of Great 

Lakes Restoration 19(1): 1-54. 

Mills, E.L, D.L. Strayer, M.D. Scheuerell, J.T. Carlton. 1996. Exotic species in the 

Hudson River Basin: A history of invasions and introductions. Estuaries 19(4): 

814-823. 

Naylor, M. 2003. Water chestnut in the Chesapeake Bay Watershed: A Regional 

Management Plan. Maryland Department of Natural Resources. 

NEMESIS: National Exotic Marine and Estuarine Species Information Systems. 2005. 

Trapa natans. [Online]. http://invasions.si.edu/nemesis/CH- 

IMP.jsp?Species_name=Trapa+natans 

39

Orth, R.J. and K.A. Moore. 1984. Distribution and abundance of submerged aquatic 

vegetation in Chesapeake Bay: An Historical Perspective. Estuaries 7(4): 531- 

540. 

Ott, A.N. 1990. Nitrogen and phosphorus loads for Susquehanna River Basin 1985-87. 

Resource Quality Management and Protection Division. Susquehanna River Basin 

Commission. 15-19. 

Pemberton, R.W. 1999. Water Chestnut. Invasive Plant Research Laboratory, US 

Department of Agriculture, Florida. 

Personal correspondence with Joy Zedler. Botany Professor, University of Wisconsin. 

January, 2008. 

Pierterse, A.H. and K.J. Murphy. 1990. Aquatic weeds: The ecology and management of 

nuisance aquatic vegetation. Oxford University Press, New York: 193-197 

Pritzlaff, D. 2003. Determination of nitrate-nitrite in surface and wastewaters by flow 

injection analysis. QuikChem®Method 10-115-01-1-F. Lachat Instruments. 

Loveland, Colorado. 

Schmidt, R. 1999. Classification of herbicides according to mode of action. WSSA 

Herbicide Handbook, Weed Society of America: 57-63. 

Sculthorpe, C.D. 1967. The biology of aquatic vascular plants. Edward Arnold Publishers 

Ltd., London: 65-75. 

Stevens, P.F. 2001. Angiosperm Phylogeny Website. Version 6, May 2005 [and more or 

less continuously updated since]. 

Steward. K. 1990. Aquatic weed problems and management in the eastern United States. 

In: Piertarse and Murphy (eds) Aquatic Weeds: The ecology and management of 

nuisance aquatic vegetation, 391-405. Oxford University Press, Oxford. 

Strayer, D.L., C. Lutz, H.M. Malcom, K. Munger and W.H. Shaw. 2003. Invertebrate 

communities associated with a native and an alien macrophyte in a large river. 

Freshwater Biology 48: 1938-1949. 

Tsuchiyai, T. and T. Iwakuma. 1993. Growth and leaf life-span of a floating-leaved plant, 

Trapa natans L., as influenced by nitrogen flux. Aquatic Botany 46: 317-324. 

National Resources Conservation Service: US Department of Agriculture. 2007. Trapa 

natans L. [Online]. http://plants.usda.gov/java/profile?symbol=TRNA 

40

Zedler, J.B. and S. Kercher. 2004. Causes and consequences of invasive plants in 

wetlands: Opportunities, opportunists, and outcomes. Critical Reviews in Plant 

Sciences 23(5): 431-452. 

41

Water Chestnut (Trapa natans L.) Infestation in the ... - SUNY Oneonta

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?