Northeast Natural History Conference 2011: Selected Papers 2012 Northeastern Naturalist 19(Special Issue 6):181–193 Long-term Management of an Invasive Plant: Lessons from Seven Years of Phragmites australis Control Karen B. Lombard1,*, Dena Tomassi1,2, and John Ebersole3 Abstract - We treated the invasive wetland grass, Phragmites australis (Common Reed), with herbicide in 99 (total of 14 ha) interdunal wetland swales of Sandy Neck barrier beach on Cape Cod, MA from 2002–2008. The herbicide applications have significantly reduced the number of Phragmites stems within invaded swales, with parallel reductions in personnel and chemical costs of the control program. After seven years of treatment, we have achieved substantial containment and suppression of Phragmites, but the plant persists in all but a few of the treated swales. Whether eradication or near-eradication can be ultimately achieved remains uncertain. Introduction Phragmites australis (Cav.) Trin. ex Steud. (Common Reed, hereafter referred to as Phragmites), a perennial grass, is historically native to the eastern United States, including New England. However, a non-native lineage of Phragmites from Europe has aggressively invaded coastal marshes and inland wetlands of the United States over the past century (Chambers et al. 1999; Meyerson et al. 2009, 2000; Saltonstall 2002). The invasive lineage is most common where wetlands have been altered, but it also invades both tidal and freshwater wetlands where anthropogenic disturbance seems minimal (Meyerson et al. 2009). Phragmites stands expand by sprouting from rhizomes to form extensive monocultures, and new stands can be established by both dispersal of rhizome fragments and seed (Bart and Hartman 2002, 2003; Belzile et al. 2010; Farnsworth and Meyerson 2003; McCormick et al. 2010; Meyerson et al. 2000; Saltonstall 2003). Phragmites grows rapidly, with high biomass production above and below ground, and its litter decomposes slowly, which hinders the growth of other species by reducing nutrients, light, and space availability (Coleman 2003, Meyerson et al. 2000, Minchinton et al. 2006). As a result, Phragmites invasions often decrease the diversity of plant communities (Back and Holomuzki 2008, Hocking et al. 1983, Meyerson et al. 2009, Rice et al. 2000, Whyte et al. 2008). On Sandy Neck barrier beach in Barnstable, Cape Cod, MA, several habitats have established stands of Phragmites. Sandy Neck (620 ha) is six miles long and ranges from 60 m to 1 km wide (Fig. 1). Although comprised of migrating sand dunes, the spit is relatively stable, which has facilitated the development of 1The Nature Conservancy, 99 Bedford Street, 5th floor, Boston, MA 02111. 2Current address - 25 Diamond Street, Plymouth, MA 02360. 3University of Massachusetts-Boston, Biology Department, 100 Morrissey Blvd., Boston, MA 02125-3393. *Corresponding author - klombard@tnc.org. 182 Northeastern Naturalist Vol. 19, Special Issue 6 several natural communities including maritime forest, cranberry bogs, and over 170 interdunal wetland swales (totaling >28 ha) (Fig. 2). Despite being first offi cially documented in 1971, the majority of the non-native Phragmites invasions on Sandy Neck is in the wetland swales and is believed to have occurred after two large storms in 1990 and 1991 that caused extensive flooding in the dune system (Coleman 2003). Further, large, healthy stands of Phragmites are present across the Great Marsh (which borders Sandy Neck to the south; labeled as “salt marsh” in Fig. 1) and probably became established due to disturbances created by shoreline development (Silliman and Bertness 2004) and hydrological alteration (Chambers et al. 1999). Swales are isolated wetland habitats that develop in the low-lying areas between surrounding dunes. They access water through accumulated precipitation or groundwater and are characteristically nutrient-poor (Shumway and Banks 2001, Swain and Kearsley 2001). Sandy Neck swales range in size from 0.004 to 1.3 ha and contain common species such as: Vaccinium macrocarpon Aiton (Cranberry), Morella pensylvanica (Mirb.) Kartesz (Northern Bayberry), Juncus canadensis J. Gay ex Laharpe (Canadian Rush), and Toxicodendron radicans (L.) Kuntze (Poison Ivy) (Shumway 1996). Native Phragmites, a subspecies that historically has been part of freshwater and brackish marshes in New England, has not been recorded in these natural communities (Coleman 2003, Saltonstall 2002). Colonization of swales by Phragmites propagules at Sandy Neck likely occurs by wind, by movement of wrack during storm surges, and by humans via recreational activities (Minchinton 2006). A previous study at Sandy Neck documented that Phragmites litter reduced the density of native plants by 94.5% in several of the swales due to the reduction of light transmission to the soil surface (Coleman 2003). Such negative effects of non-native Phragmites in these Figure 1. Swales treated for Phragmites invasion and uninvaded swales in Sandy Neck, Cape Cod, MA. Inset shows a close up of some of the interdunal swales. 2012 K.B. Lombard, D. Tomassi, and J. Ebersole 183 swales threaten the diversity of native plants and the habitat quality for several state-threatened or special-concern species that inhabit these swales, including Scaphiopus holbrookii (Eastern Spadefoot Toad) and Sabatia kennedyana Fernald (Plymouth Rose Gentian) (Anderson et al. 2006). In 2002, The Nature Conservancy initiated a long-term herbicide-based control program of Phragmites at Sandy Neck, with the goal of reducing Phragmites to levels where it would presumably have limited effects on native swale communities and future control would require limited (1–2 weeks) staff time per year. Although herbicide application reduces Phragmites cover greatly in the short term (after 1–2 years of treatment), research indicates that reapplication of herbicide is required for long-term effectiveness (Ailstock et al. 2001, Back and Holomuzki 2008, Carlson and Kowalski 2009, Jones 1987, Modzer et al. 2008, Moreira et al. 1999), and control is generally most successful when herbicide application is combined with other techniques (burning, flooding, cutting, etc.; Marks et al.1994). Consequently, it may take many years to either reduce the invasion to a maintenance level or eradicate the species from a particular location. Methods In a previous study at Sandy Neck, Coleman (2003) mapped most of the swale plant communities at Sandy Neck using GPS, characterized them by Figure 2. A typical wetland swale without Phragmites invasion at Sandy Neck. Photograph © Nina Coleman. 184 Northeastern Naturalist Vol. 19, Special Issue 6 recording prevalent native plants, and surveyed them for the presence of Phragmites. The patchy distribution of Phragmites makes detailed assessment of density or abundance logistically unfeasible, so we continued the method developed by Coleman (2003) of using broad categories to score Phragmites density and abundance (Tables 1, 2; see supplemental appendix 1, available online at https://www.eaglehill.us/NENAonline/suppl-files/n19-sp6-1032b- Lomabard-s1, and, for BioOne subscribers, at http://dx.doi.org/10.1656/ N1032b.s1 for descriptions of other monitoring techniques evaluated for their usefulness but not ultimately chosen for data collection). In subsequent years of the project (2002–2008), we trained seasonal staff to score density and abundance of Phragmites according to Coleman’s methods. We also trained seasonal staff in herbicide application, using the method of hand cutting Phragmites stems with clippers and then dripping a 50% solution of RodeoTM or AquamasterTM (active ingredient: isopropylamine salt of glyphosate) into the center of each stem (cut and drip). We used 2- to 4-person crews hired from late August to early October each year, except when we applied treatments for only two weeks in 2002 and only two days in 2006, due to personnel and funding constraints. We expanded treatment methods after 2003 to include plants too small for the cut and drip method, which were swiped with a cotton glove dipped in 33% solution of glyphosate herbicide. We also used backpack sprayers to apply 2% glyphosate in high-density Phragmites stands with limited native vegetation. We found Phragmites in 55 (9 ha) of the 133 swales that were assessed in 2001–2002, and treated 14 of these swales (3 ha) with herbicide in 2002. Up to 2008, we attempted to retreat previously treated swales and initiate treatment in Table 1. Phragmites density scores used to estimate the total number of Phragmites stems per swale area for yearly classification of swales before treatment. Phragmites density scores Description 0 No live Phragmites stems present 1 Light: ≤200 stems per swale, native vegetation appears normal under Phragmites 2 Moderate: >200 stems per swale, patchy Phragmites with some dense areas, but also some areas with native vegetation 3 Heavy: Phragmites and poison ivy dominant in swale, significant Phragmites thatch present Table 2. Phragmites abundance scores used to estimate the percent cover of Phragmites stems for yearly classification of swales before treatment. Phragmites abundance scores Description 0 No live Phragmites stems present 1 <25% of swale covered 2 25-49% of swale covered 3 50%-75% of swale covered 4 >75% of swale covered 2012 K.B. Lombard, D. Tomassi, and J. Ebersole 185 as many new swales as could be reached during the treatment period. By 2009, the last year of assessment, we located and treated Phragmites in an additional 44 swales for a total of 99 treated swales (14 ha). The invaded swales varied in initial level of Phragmites infestation and have different treatment histories, since we were not able to treat all swales in all possible years. To examine the impact of herbicide applications on Phragmites infestations in swales, we analyzed how the estimated densities (total number of Phragmites stems per swale area) and abundance (percent cover) changed over time in the 133 swales originally mapped in 2001/2002. To determine whether herbicide application had a significant impact on Phragmites infestations, we performed linear mixed model regressions on estimated density and abundance scores from 2002–2009, with time (year) and number of treatments as independent variables. Mixed model regression takes into consideration the repeated measures that may have a significant effect on the estimated density and abundance of a Phragmites infestation. All monitoring data were analyzed using PASW® Statistics 18 (IBM.com). In addition to tracking swales treated, we examined the personnel hours required for a treatment period and annual costs of the project, including supplies, transportation, communication, labor, data analysis, and report writing; indirect costs and housing (donated) were not included. Results Over the seven years of the control project, we found that Phragmites was steadily reduced in terms of both density and abundance in the 133 swales first mapped in 2001–2002 (Fig. 3), with the highest reductions in areas that initially had high abundance and/or high density of Phragmites. More specifically, just before herbicide applications had begun in 2002, 15% of swales received the highest score for estimated Phragmites density (score of 3), indicating a heavy invasion (Fig. 3A); however, only 1.5% of these swales were given the highest density ranking in 2009. The number of swales in the moderately dense category (score of 2) decreased during the same time period: about 16% were moderately dense swales in 2002, compared to 8% in 2009. During this treatment period, the proportion of swales described as having a light invasion or no invasion (density score of 0 or 1) rose from 68% to 90%. Although Phragmites density decreased in the swales, the swales show only a slightly decreasing trend in the proportion with Phragmites over time (Fig. 3). Comparing 2002 (pre-treatment) to 2009 (last year of data showing 2008 treatment results) shows little change in the proportion of swales infested with Phragmites (41% versus 35%). This pattern of Phragmites reduction over the course of the control project is also evident from the estimated abundance data (Fig. 3B). In 2002, 16% of swales were given the highest abundance score of 4, indicating heavy invasion, while none of the swales were given this maximum score in 2008 and 2009. Similarly, the number of swales in the second highest abundance category (score of 3) fell 186 Northeastern Naturalist Vol. 19, Special Issue 6 from 7% in 2002 to 0% in 2009. Finally, swales given an abundance score of 1 increased over time, from ≈13% in 2002, to almost 30% in 2009. During the same Figure 3. Phragmites density scores (A) and Phragmites abundance scores (B) from 2002–2009 for swales initially mapped in 2001 or 2002. 2012 K.B. Lombard, D. Tomassi, and J. Ebersole 187 seven-year period, the proportion of swales described as having a light invasion or no invasion (abundance score 0 or 1) increased from 71.5% to 93.2%. The linear mixed model regressions show that year (time) and the number of herbicide applications are both highly significant predictors of Phragmites abundance and Phragmites density scores. Phragmites was significantly reduced with repeated herbicide treatments over the years of the program (Tables 3, 4). The parameter estimates of both independent variables are negative, indicating a decreasing relationship with the dependent variable. In other words, as time and the number of treatments increase, the estimated density and abundance scores decrease. Five years of treatment were needed before every invaded swale was treated at least once, and large numbers of treated swales begin to have no Phragmites stems only after seven years of treatment (Fig. 4). In several cases, we observed that Phragmites appeared again in a swale after it was recorded as absent in a previous year. For example, in 2007, four swales that had previously been treated had no Phragmites; however, in 2008 two of these had Phragmites stems and had to be retreated. In 2008, twelve swales were reported with no Phragmites, but, in two of these, Phragmites was present again in 2009. In both cases, the re-invasion was a small number of stems that were presumed to be from rhizomes not killed by herbicide. As the invasion was controlled and Phragmites levels declined at Sandy Neck, personnel hours, chemical amounts, and cost generally decreased (Fig. 5). The cost of the project ranged from almost $3000/ha in the early years to less than $1000/ha, with the overall cost to date being approximately $110,000 USD (not including indirect costs). When we started the project, we used a crew of 3–4 people for 8 weeks or 131 hrs/ha (not including the first year of treatment). By 2009, we reduced the crew to 2 people for 2–6 weeks or 77 hrs/ha. In addition, we reduced the liters of glyphosate used from 10 l/ha to 1 l/ha over the same time period. Table 4. Linear mixed model regressions for estimated abundance scoring from 2002-2009 (** indicates a result below an alpha level of 5%). 95% confidence interval Parameter Estimate Std. Error df t P Lower Upper Intercept 94.80 23.20 998.0 4.1 less than 0.001 49.40 140.20 **Number of treatments -0.20 0.02 1070.9 -9.7 less than 0.001 -0.28 -0.20 **Year -0.05 0.012 998.0 -4.05 less than 0.001 -0.07 -0.02 Table 3. Linear mixed model regressions for estimated density scoring from 2002-2009 (** indicates a result below an alpha level of 5%). 95% confidence interval Parameter Estimate Std. Error df t P Lower Upper Intercept 50.90 19.00 989 2.7 0.008 13.60 88.20 **Number of treatments -0.20 0.02 1059 -8.8 less than 0.001 -0.20 -0.10 **Year -0.03 0.01 989 -2.6 0.009 -0.04 -0.01 188 Northeastern Naturalist Vol. 19, Special Issue 6 Discussion Our analysis of the 2002–2008 Phragmites treatment data demonstrates the control project’s success in reducing the density and abundance of Phragmites plants in interdunal swales of Sandy Neck. By 2009, 79% of the swales treated had either no Phragmites or were at a low density in 2009—levels less likely to affect native vegetation. Although the Sandy Neck project did not include untreated swales as experimental controls, it is likely that without these chemical treatments, many of the wetland swales at Sandy Neck would now be heavily invaded by Phragmites. In Virginia, along the seaside of the eastern shore, comparisons between untreated and treated landholdings showed that in areas of no treatment, Phragmites acreage expanded 68% (130 acres) over four years from 2004–2008, whereas Phragmites acreage in aerially treated areas decreased 34% (238 acres) (Myers et al. 2009). On Lake Erie, seven years of herbicide applications reduced Phragmites from 18 to 6 percent of emergent vegetation (Back and Holomuzki 2008). Although the control efforts have resulted in a significant reduction in Phragmites stems, the plant persists in all but a few swales. The influence of application methods and timing on herbicide effectiveness may help explain this continued persistence. Glyphosate may be most effective when applied in the fall (Cross and Fleming 1989, Derr 2008, Moreira et al. 1999; Figure 4. Number of previously treated swales without live Phragmites stems in succeeding years. No data were collected in 2006. 2012 K.B. Lombard, D. Tomassi, and J. Ebersole 189 Figure 5. Resources per hectare needed, including: A) liters of herbicide used/ha (Rodeo ® or Aquamaster®), B) person hours/ha, and C) costs/ha. In year one, we only treated for two weeks, and in year five, we used a contractor to treat just a few highdensity swales. The majority (over 85%) of costs is for personnel; indirect costs are not included in these estimates. 190 Northeastern Naturalist Vol. 19, Special Issue 6 but see Mozdzer et al. 2008), when Phragmites is preparing for dormancy and nutrients are being directed to the horizontal rhizomes (Haslam 1969a, 1969b, 1969c; Norris et al. 2002); herbicide applied during this period can be transferred throughout the entire root system to produce a high mortality rate (Derr 2008). Prior to 2006, the small work crew at Sandy Neck had to begin herbicide applications in summer to have enough time to treat the many, widespread, heavily infested swales before the first frost. In more recent years, the Sandy Neck control project has shifted to fall treatments, which likely is why we saw substantially reduced Phragmites invasions since 2006. Initiating treatments earlier in some years may have limited the effectiveness of the herbicide, allowing Phragmites to re-emerge. Logistical considerations also prevented the use of the most effective herbicide application method, cut-and-drip, in heavily infested swales during the early years of the control project. Although this method is both time- and laborintensive, cut–and-drip minimizes non-target negative effects, and provides a high probability that the herbicide will be translocated from the plant stem to the belowground system of roots and rhizomes (Norris et al. 2002). Application of herbicide by swiping or backpack spraying instead of cut-and-drip in some heavily infested swales at Sandy Neck may have reduced the impact of herbicide and allowed sprouting of new Phragmites individuals from these extensive belowground systems. Since application technique and timing have such a strong effect on the success of herbicide treatments, repeat herbicide applications are usually required to gain complete control and eventually eradicate expanding Phragmites stands (Havens et al. 1997, Marks et al. 1993, Warren et al. 2001), and this may be the case on Sandy Neck. Maintenance of currently low Phragmites density and abundance as well as further reductions will continue to require time and resources into the future. Although we did not reach our original goal of getting the invasion to a point that could be treated in 1–2 weeks each year, we did substantially reduce the number of personnel hours and cost, so that two staff members working over four weeks could treat all swales. This project was funded for many years with a combination of federal, state, and private grants; however, identifying new funding sources for necessary maintenance of these gains will continue to be a major challenge. Other studies indicate that eradication of a plant species is possible (Gardener et al. 2009; Panetta and Timmins 2004; Rejmánek and Pitcairn 2002; Simberloff 2003, 2008), but will likely require over 10 years to achieve (Mack and Lonsdale 2002). In this case, we did not start seeing many swales with no Phragmites stems until 5–7 years into the project, and many swales still exhibited low numbers of stems. Follow-up monitoring and treatment will probably be needed in perpetuity because stems are still present and re-invasion is likely due to adjacent source populations and continued anthropogenic impacts. Given the challenging site logistics of locating, monitoring, and potentially treating the large number of swales each year, containment of Phragmites at low densities in the currently invaded swales may be a more pragmatic goal, with near-eradication possible with continued treatments. 2012 K.B. Lombard, D. Tomassi, and J. Ebersole 191 Acknowledgments This work was funded by the Massachusetts Environmental Trust, the NRCS Wildlife Habitat Incentive Program, the USFWS Partners Program, the William Wharton Trust, and Entrust Fund. We thank the Town of Barnstable and in particular Sandy Neck Beach Managers, Nina Coleman and Ken Alfieri, for critical support during this project. We appreciated the assistance of the Massachusetts Audubon Society; Cape Cod AmeriCorps; as well as the numerous invasives crewmembers that provided many hours of dedicated work to control this invasion. Appreciation also goes to The Nature Conservancy’s Eastern Invasives Network and Rich McHorney for assistance in early stages of the project and to Erik Kiviat who encouraged us to tell this story to a wider audience. This paper also benefitted from the excellent input of Jessica Dyson, Peter Kareiva, Dan Majka, Laura Marx, Karen Poiani, and Stacey Solie. Literature Cited Ailstock, M.S., C.M. Norman, and P.J. Bushmann. 2001. Common Reed Phragmites australis: Control and effects upon biodiversity in freshwater nontidal wetlands. Restoration Ecology 9:49–59. Anderson, M.G, K. Lombard, J. Lundgren, B. Allen, S. Antenen, D. Bechtel, A. Bowden, M. Carabetta, C. Ferree, M. Jordan, S. Khanna, D. Morse, A. Olivero, N. Sferra, and M. Upmeyer. 2006. The North Atlantic coast ecoregional assessment: Conservation status Report and resource CD. The Nature Conservancy, Boston, MA. Back, C.L., and J.R. Holomuzki. 2008. Long-term spread and control of invasive Common Reed (Phragmites australis) in Sheldon Marsh, Lake Erie. Ohio Journal of Science 108:108–112. Bart, D., and J.M. Hartman. 2002. Environmental constraints on early establishment of Phragmites australis in salt marshes. Wetlands 22:201–213. Bart, D., and J.M. Hartman. 2003. The role of large rhizome dispersal and low salinity windows in the establishment of Common Reed, Phragmites australis, in salt marshes: New links to human activities. Estuaries 26:436–443. Belzile, F., J. Labbé, M-C LeBlanc, and C. Lavoie. 2010. Seeds contribute strongly to the spread of the invasive genotype of the Common Reed (Phragmites australis). Biological Invasions 12:2243–2250. Carlson, M.L., and K.P. Kowalski. 2009. Promoting species establishment in a Phragmites- dominated Great Lakes coastal wetland. Natural Areas Journal 29:263–277. Chambers, R.M., L.A. Meyerson, and K. Saltonstall. 1999. Expansion of Phragmites australis into tidal wetlands of North America. Aquatic Botany 64:261–273. Coleman, N. 2003. Interdunal swales and the colonization, effect, and removal of the invasive grass Phragmites australis. M.Sc. Thesis. University of Massachusetts, Amherst, MA. 94 pp. Cross, D.H., and K.L. Fleming. 1989. Control of Phragmites or Common Reed. Waterfowl Management Handbook. Fish and Wildlife Leaflet 13:1–5. Derr, J. 2008. Common Reed (Phragmites australis) response to mowing and herbicide application. Invasive Plant Science and Management 1:12–16. Farnsworth, E.J., and L.A. Meyerson. 2003. Comparative ecophysiology of four wetland plant species along a continuum of invasiveness. Wetlands 23:750–762. Gardener, M., R., Atkinson, and J.L. Rentería, J. 2009. Eradications and people: Lessons from the plant eradication program in Galapagos. Restoration Ecology 18:20–29. Haslam, S. 1969a. Stem types of Phragmites communis Trin. Annuals of Botany 33:127–131. 192 Northeastern Naturalist Vol. 19, Special Issue 6 Haslam, S. 1969b. The development of shoots in Phragmites communis Trin. Annuals of Botany 33:695–709. Haslam, S. 1969c. The development and emergence of buds in Phragmites communis Trin. Annuals of Botany 33:289–301. Havens, K.J., W.I. Priest, and H. Berquist. 1997. Investigation and long-term monitoring of Phragmites australis within Virginia’s constructed wetland sites. Environmental Management 21:599–605. Hocking, P.J., C.M. Finlayson, and A.J. Chick. 1983. The biology of Australian weeds. 12. Phragmites australis (Cav.) Trin. Ex Steud. Journal of the Australian Institute of Agricultural Science 49:123–132. Jones, W.L. 1987. Phragmites control and revegetation following aerial applications of glyphosate in Delaware. Pp. 185–196, In W.R. Whitman and W.H. Meredith (Eds). 1987. Waterfowl and Wetlands Symposium: Proceedings of a Symposium on Waterfowl and Wetlands Management in the Coastal Zone of the Atlantic Flyway. Delaware Coastal Management Program, Delaware Department of Natural Resources and Environmental Control, Dover, DE. Mack, R.N., and W.M. Lonsdale. 2002. Eradicating invasive plants: Hard-won lessons from islands. Pp. 164–172, In C.R. Veitch and M.N. Clout (Eds.). Invasive Species Specialist Group of the World Conservation Union (IUCN), Auckland, NZ. Marks, M., B. Lapin, and J. Randall. 1994. Phragmites australis (P. communis): Threats, management, and monitoring. Natural Areas Journal 14:285–294. McCormick, M.K., K.M. Kettenring, H.M. Baron, and D.F. Whigham. 2010. Extent and reproductive mechanisms of Phragmites australis spread in brackish wetlands in Chesapeake Bay, Maryland (USA). Wetlands 30:67–74. Meyerson, L., K. Saltonstall, and R.M. Chambers. 2009. Phragmites australis in eastern North America: A historical and ecological perspective. Pp. 57–82, In B.R. Silliman, E. Grosholz, and M.D. Bertness (Eds.). Salt Marshes Under Global Siege. University of California Press, Berkeley, CA. Meyerson, L.A., K. Saltonstall, L. Windham, E. Kiviat, and S. Findlay. 2000. A comparison of Phragmites australis in freshwater and brackish marsh environments in North America. Wetlands Ecology and Management 8:89–103. Minchinton, T.E., J.C. Simpson, and M.D. Bertness. 2006. Mechanisms of exclusion of native coastal marsh plants by an invasive grass. Journal of Ecology 94:342–354. Minchinton, T.E. 2006. Rafting on wrack as a mode of dispersal for plants in coastal marshes. Aquatic Botany 84:372–376. Modzer, T.J., C.J. Hutto, P.A. Clarke, and D.P. Field. 2008. Efficacy of imazapyr and glyphosate in the control of non-native Phragmites australis. Restoration Ecology 16:221–224. Moreira, I., A. Monteiro, and E. Sousa. 1999. Chemical control of Common Reed (Phragmites australis) by foliar herbicides under different spray conditions. Hydrobiologia 415:299–304. Myers, K., R. Kevin, E. Heffernan, and P.A. Clarke. 2009. Census and mapping Phragmites australis in Virginia: Basic information to support control and monitoring strategies. Presentation to the 36th Natural Areas Conference, Vancouver, WA. Norris, L., J.E. Perry, and K.J. Havens. 2002. Wetlands program technical report: A summary of methods for controlling Phragmites australis. VIMS Publication Center, Gloucester Point, VA. Panetta, F.D., and S.M. Timmins. 2004. Evaluating the feasibility of eradication for terrestrial weed incursions. Plant Protection Quarterly 19:5–11. 2012 K.B. Lombard, D. Tomassi, and J. Ebersole 193 Rejmánek, M., and M.J. Pitcairn. 2002. When is eradication of pest plants a realistic goal? Pp. 249–253, In D. Veitch and M. Clout (Eds.). Turning the Tide: The Eradication of Invasive Species. Invasives Species Specialty Group of the World Conservation Union (IUCN), Auckland, NZ. Rice, D., J. Rooth, and J.C. Stevenson. 2000. Colonization and expansion of Phragmites australis in upper Chesapeake Bay tidal marshes. Wetlands 20:280–299. Saltonstall, K. 2003. Microsatellite variation within and among North American lineages of Phragmites australis. Molecular Ecology 12:1689–1702. Saltonstall, K. 2002. Cryptic invasion by a non-native genotype of the Common Reed, Phragmites australis, into North America. Proceedings of the National Academy of Sciences of the United States of America 99:2445–2449. Shumway, S.W. 1996. The swales of Sandy Neck. A Report to the Massachusetts Natural Heritage and Endangered Species Program, Westborough, MA. Shumway, S.W., and C.R. Banks. 2001. Species distribution in interdunal swale communities: The effects of soil waterlogging. The American Midland Naturalist 145:137–146. Silliman, B.R., and M.D. Bertness. 2004. Shoreline development drives invasion of Phragmites australis and the loss of plant diversity on New England salt marshes. Conservation Biology 18:1424–1434. Simberloff, D. 2003. Eradication: Preventing invasions at the onset. Weed Science 51:247–253. Simberloff, D. 2008. We can eliminate invasions or live with them. Successful management projects. Biological Invasions 11:149–157. Smith, S.M., M. Hanley, and K.T. Killingbeck. 2008. Development of vegetation in dune slack wetlands of Cape Cod National Seashore (Massachusetts, USA). Plant Ecology 194:243–256. Swain, P.C., and J.B. Kearsley. 2001. Classification of the Natural Communities of Massachusetts. Version 1.3. Natural Heritage and Endangered Species Program, Division of Fisheries and Wildlife, Westborough, MA. Warren, R.S., P.E. Fell, J.L. Grimsby, E.L. Buck, C. Rilling, and R.A. Fertik. 2001. Rates, patterns, and impacts of Phragmites australis expansion and effects of experimental Phragmites control on vegetation, macroinvertebrates, and fish within tidelands of the lower Connecticut River. Estuaries 24:90–107. Whyte, R.S., D. Trexel-Kroll, D.M. Klarer, R.S. Shields, and D.A. Francko. 2008. The invasion and spread of Phragmites australis during a period of low water in a Lake Erie coastal wetland. Journal of Coastal Research 55:111–120.