Evaluating the Success of Seed Sowing in a New England Grassland Restoration

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, researchlibraries, and research funders in the common goal of maximizing access to critical research.

Evaluating the Success of Seed Sowing in a New England Grassland RestorationAuthor(s): Chad C. Jones, Glenn D. Dreyer, Nels BarrettSource: Natural Areas Journal, 33(2):214-221. 2013.Published By: Natural Areas AssociationDOI: http://dx.doi.org/10.3375/043.033.0211URL: http://www.bioone.org/doi/full/10.3375/043.033.0211

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, andenvironmental sciences. BioOne provides a sustainable online platform for over 170 journals and books publishedby nonprofit societies, associations, museums, institutions, and presses.

Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance ofBioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use.

Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiriesor rights and permissions requests should be directed to the individual publisher as copyright holder.

http://dx.doi.org/10.3375/043.033.0211

http://www.bioone.org/doi/full/10.3375/043.033.0211

http://www.bioone.org

http://www.bioone.org/page/terms_of_use

214 Natural Areas Journal Volume 33 (2), 2013

Natural Areas Journal 33:214–221

•

Evaluating the Success of Seed Sowing in a New

England Grassland Restoration

Chad C. Jones1,4

1Botany DepartmentConnecticut College270 Mohegan Ave

New London, CT 06320

Glenn D. Dreyer2

Nels Barrett3

2Connecticut College Arboretum270 Mohegan Ave

New London, CT 06320

3USDA Natural Resource ConservationService Connecticut Office344 Merrow Road, Suite ATolland, CT 06084-3917

•

R E S E A R C H N O T E

4 Corresponding author:[email protected]; 860-439-5304

ABSTRACT: Grassland habitat is declining in the northeastern United States, leading to a decline in as-sociated native species. Consequently, there is considerable interest by land managers in conserving andrestoring grassland habitats in the Northeast. However, unlike the Great Plains and Europe, quantitativemonitoring of restoration sites is uncommon, making it difficult to improve new restoration projects.Here we evaluate a grassland restoration in Waterford, Connecticut, to determine if mechanical clearingof woody vegetation combined with sowing 23 native grasses and forbs led to successful establishmentof these species. We also compared cover, diversity, and colonization by exotic and woody species inplanted and unplanted areas over time. In the third and fifth growing seasons after planting in 2006,we sampled the vegetation in the planted site, an unplanted zone within the planted grassland, and anadjacent unplanted grassland. Twenty of the 23 sown species established by 2010, and sown speciesdominated the planted area (70% of total cover). Despite the successful establishment of most sownspecies, species richness and diversity were no higher in the sown grassland than in adjacent unseededareas. However, the sown grassland contained lower cover of non-native and invasive species. Bigbluestem (Andropogon gerardii Vitman) established aggressively, potentially reducing both exoticcolonization and native diversity. This study shows that sowing native grassland species can lead to thesuccessful development of native-dominated grasslands. Results can inform future grassland restora-tion efforts in the Northeast and show that seeding with aggressive grass species may greatly impactrestored plant communities.

Index terms: early successional habitat, exotic species, monitoring, New England grassland restoration,species diversity

INTRODUCTION

Grassland habitats have declined over thelast century throughout the northeasternUnited States due to agricultural aban-donment, suburban development, and firesuppression (Vickery et al. 1994). Thisdecrease has led to considerable interestin conserving and restoring these openhabitats throughout the region.

It is uncertain how prevalent grasslandhabitats were in the northeastern UnitedStates prior to European colonization(Askins 1997; Foster and Motzkin 2003).However, some areas dominated by na-tive grassland species were present in theregion, and the recent loss of grasslandshas clearly led to the decline of nativegrassland plants, insects, and birds (Marks1983; Askins 1997). The importance ofgrasslands in preserving biodiversity inhuman-dominated landscapes has beenrecognized for decades in Europe (Töröket al. 2010; Piqueray et al. 2011). As inEurope, conservation and restoration ofnative plant dominated grasslands in thenortheastern United States has been ac-knowledged as an important componentof the conservation of biodiversity in theregion, even at specific sites that were notgrasslands prior to European settlement(Foster and Motzkin 2003).

In areas where grasslands have convertedto forest, mechanical removal of the woody

species is necessary to restore grasslandhabitat (Lezberg et al. 2006; Dzwonkoand Loster 2007). Even after removal ofwoody vegetation, however, recolonizationby grassland species is necessary becausethese species usually do not persist in theseed bank in forested areas (Bisteau andMahy 2005). If existing grasslands areadjacent to the restoration site, naturalcolonization may be sufficient (Pärtel et al.1998). However, the majority of evidenceshows that, for isolated sites and for manygrassland species, sowing may be necessaryto restore species-rich grassland habitats(Dzwonko and Loster 2007; Piqueray etal. 2011). Because seed mixes of nativeplants, especially those using regionalecotypes, can be both difficult and costlyto obtain (Jongepierová et al. 2007), it isimportant to understand if seed sowingimproves restoration outcomes. If not,then limited resources may be better usedin other aspects of restoration.

Numerous studies have examined the ef-fectiveness of seed sowing in grasslandrestoration and have uncovered severalcritical themes. First, establishment successof sown species can vary greatly amongsites (Lepš et al. 2007) and among species(Pakeman et al. 2002; Hellstrom et al.2009). The degree of initial disturbancecan influence establishment success, withgreater disturbance (e.g., tilling or clearing)usually increasing establishment (Pywellet al. 2007). Second, sowing with more

Volume 33 (2), 2013 Natural Areas Journal 215

diverse seed mixes has been shown to in-crease species richness compared to naturalregeneration or less diverse seed mixes(Pywell et al. 2002; Foster et al. 2007),especially when natural seed sources aredistant. Thus, seed sowing is often used asa tool to increase the grassland diversity.Third, sowing native seed can help to pre-vent the establishment and spread of exoticor invasive species (Lawson et al. 2004;Sheley and Half 2006; Foster et al. 2007;Török et al. 2010). Invasive species are amajor cause of restoration failure (Berger1993; Kettenring and Adams 2011), sopreventing invasions is a major goal ofgrassland restoration efforts.

Although grassland restoration has beenwell documented in Europe and the Mid-west United States, there is little publishedinformation about clearing or seed sowingin restoration efforts in the northeasternUnited States. Because there can be largeregional variation in species establishment(Lepš et al. 2007), it is unclear whetherresults from other regions can be appliedto grassland restoration in the Northeast.

Restoring grassland and shrubland habitatis a conservation priority in the Northeast,and funding is available for landownersthrough federal programs and the Con-necticut Department of Energy and Envi-ronmental Protection (DEEP). The DEEPhas worked to restore or enhance 524 haof grassland habitat on state land between1998 and 2008, with 422 ha involvingmechanical clearing of woody vegetationand 133 ha involving sowing seeds (PaulRothbart, DEEP, pers. comm.). In addition,land trusts and private landowners havebeen restoring grassland habitat.

Despite these efforts, it is unclear howmuch monitoring is occurring. The onlypublished example with quantitativemonitoring that we could find focused onthe effects of clearing rather than on theeffects of sowing seeds (Lezberg et al.2006). Results from quantitative monitor-ing are necessary to inform future grasslandrestoration efforts in the Northeast.

In this study, we assessed the results ofa project in Waterford, Connecticut, torecreate grassland habitat by mechanically

clearing woody vegetation and plantingseeds of species native to local grasslands.We compared vegetation change over fivegrowing seasons in an area seeded with23 species of native grassland herbs withtwo adjacent unseeded areas. We asked thefollowing questions: (1) Does seed sowinglead to establishment of native grasslandspecies? (2)Are species richness, diversity,grass cover, and total plant cover changingover time in these grasslands? and (3) Arethe abundances of exotic, invasive, andwoody species increasing over time?

This project is similar to those commonlycarried out by land managers in NewEngland. We recognize that this studyonly represents a single site, and is not areplicated test of the effect of seed sow-ing in New England. Nevertheless, thisstudy provides quantitative monitoring ofgrassland restoration that is lacking in thisregion and can thereby help inform futuregrassland restoration efforts.

METHODS

The study site was part of a small complexof grassland and savannah in the Con-necticut College Arboretum, Waterford,Connecticut. The overall project goalwas to reclaim grassland habitat from en-croaching forest vegetation and to expandthe total open area to approximately 4.2ha. The site slopes gently to the northeastand adjoins a field to the east that hasbeen kept in predominantly herbaceousvegetation by mowing since cultivationended approximately 60 years ago. Thesoil is mapped as Sutton very stony, finesandy loam (NRCS 2011).

Approximately 2 ha of woody vegetationwas mechanically cleared during 2004(Figure 1), stumps were ground down tobelow the soil surface, and larger woodydebris was removed with a York rake.Prior to clearing, the site was covered byyoung forest of Prunus serotina Ehrh., Acer rubrum L., Pinus strobus, and Quercus velutina Lam. The understory was domi-nated by invasive exotic shrubs and vines(e.g., Celastrus orbiculatus Thunb. andBerberis thunbergii DC.). Native shrubswere uncommon and there was very littleherbaceous vegetation.

Three separate treatments were appliedto parts of the cleared area. A 1.5 ha area(hereafter “planted”) was seeded on 22June 2006 with a mixture of native grassand forb species supplied by New EnglandWetland Plants, Amherst, Massachusetts,(Table 1) using a tractor-pulled Truax FlexII seed drill. Seeds were applied at ~30kg/ha based on recommendations fromthe seed supplier. Approximately 6 cm ofrain fell during the next three days, allow-ing rapid germination. No supplementalirrigation or mulch was applied. A 0.07ha area (hereafter “cleared”) within thisfield was cleared of all vegetation, but notplanted due to some large boulders thatprevented access by the seed drill (Figure1). Approximately 0.5 ha of the adjacentfield to the east was cleared of encroach-ing woody vegetation, but existing groundlayer vegetation, mainly grasses and forbspersisting from the old agricultural fields,remained intact (hereafter “existing”).The herbicides Crossbow® and Garlon®3A were selectively applied to regrowingwoody and invasive plants in all grasslandsevery summer since 2005. All grasslandareas were mowed once each year in latewinter.

Sampling

In July 2008, a permanently marked, 100m long, east-west transect was establishedthrough the middle of the “planted”grassland and into the “cleared” grassland(Figure 1). Thirty 1-m2 sample plots wereplaced every 3 m on the north side of thisline with 23 plots in the planted, and 7 inthe cleared grasslands.

A second permanent transect was estab-lished in the adjacent “existing” grassland(Figure 1). Fifteen plots were placed every3 m on the east side of this 50-m north-south transect.

The percent cover of each species in eachplot was estimated between 20 Augustand 1 September in both 2008 and 2010.Percent cover was measured by visualestimates of areal projection allowing foroverlapping canopies; hence plot totalsexceeded 100%. Visual estimates weremade by at least two of the three authorsfor each plot. Species nomenclature fol-lows the Integrated Taxonomic Information


System (http://www.itis.gov). Voucherspecimens are contained in the CharlesGraves Herbarium at Connecticut College(CCNL). We classified species as native orexotic using Tucker (1995) and as invasivebased on the official list of Connecticutinvasive species (http://www.hort.uconn.edu/cipwg/list.html). Artemisia vulgaris L.,an aggressively spreading exotic species,was included as an invasive species.

Data Analysis

The abundance of species in the seedmixture (percent of total seeds by weight)was compared with the percent cover ofthose species in the planted grassland in2008 and 2010 using two-tailed Pearsoncorrelation. Species richness and diversity(Shannon-Weiner H`) were calculated foreach plot.

We used paired t-tests to compare rela-tive cover of sown species, total cover,species richness, diversity (question 2),exotic cover, invasive cover, and woodycover (question 3) between sample datesfor each of the three grasslands.

RESULTS

Nineteen of the 23 sown species occurred

within plots in at least one year (Table 1)and one more (Penstemon digitalis Nutt.ex Sims) was seen in the grassland but wasnot sampled. Establishment success washighly variable, with some sown speciesbecoming abundant while others remainedrare in the grassland. Four species occurredon the transect for the first time in 2010(Table 1), suggesting that germination andestablishment of these species may takeseveral years.

The proportion of sown species increasedin the planted grassland between years,reaching ~70% of cover by 2010 (t =5.62, p < 0.001, Figure 2a). Sown specieshad much lower cover in both unsowngrasslands and did not change over time (t< 1.2, p > 0.25). These species may havenaturally dispersed into these grasslandsfrom surrounding areas. Graminoids weremore common in the sown grassland thanin the other two sites (Figure 2b).

Species that were more abundant in theseed mixture tended to be more commonin the planted grassland in 2008 (r = 0.53,p = 0.009). This relationship weakened tomarginal significance by 2010 (r = 0.39, p =0.063), largely due to increased dominanceby a few of the sown species.

Total cover increased overall between 2008

and 2010 in the planted (t = 5.03, p < 0.001)and cleared grasslands (t = 4.24, p = 0.005),but not in the existing grassland (t = 0.05,p = 0.96). By 2010, total cover was similaramong all three sites (Figure 3a).

Species richness per plot did not changesignificantly over time in the planted orcleared meadow for either total richnessor native species richness (t < 1.08, p >0.32; Figure 3b.). In the existing meadow,native richness declined slightly (t = 2.06,p = 0.058). Despite planting 23 native spe-cies, there was no indication that richnesswas higher in the planted grassland thaneither of the other grasslands. The Shan-non-Weiner diversity index varied littleamong sites or years.

Cover of exotic species and invasive speciesdid not increase between 2008 and 2010 inany grassland (t < 1.27, p > 0.18; Figure4a). Exotic species cover in the plantedgrassland was very low (< 3%) compared tothe other two grasslands. Cover of invasiveexotic species showed a similar pattern,with lower cover in the planted than inthe existing grassland. However, invasivecover was < 6.5% of the total cover in allsites. The cover of woody species increasedover time in both the planted (t = 2.26,p = 0.034) and the cleared meadow (t =3.11, p = 0.021) despite targeted herbicide

Figure 1. Aerial photo of site in 2001 (A) prior to restoration and in 2006 (B) following restoration. White lines indicate the location of transects (1 – seeded,2 – unseeded). The dashed white line indicates the cleared grassland (unseeded portion of seeded grassland).


application (Figure 4b). The majority ofwoody cover consisted of native species(Table 2), including several species ofblackberry (Rubus spp.) and winged sumac(Rhus copallina L.).

DISCUSSION

After five growing seasons, native herba-ceous vegetation dominated both plantedand non-planted restored areas, regardlessof whether the site was initially cleared ofground layer vegetation. The abundanceof woody species did increase over time,suggesting that continued management isneeded to prevent the site from returningto dominance by woody species.

As seen in grassland restoration in otherregions (Lawson et al. 2004; Foster etal. 2009), the majority of sown speciesestablished on the site. The sown site hadgreater grass dominance than unsownsites, and nine species established in theplanted grassland that were not present inunplanted grasslands in 2008. Sowing withgrass-dominated seed mixtures has alsosuccessfully increased grass dominance inother regions (Bakker and Wilson 2004).

The abundance of the species in the seedmix corresponded to its abundance in thevegetation although this relationship de-clined over time. Thus, in the short term,the proportion of seed of each species canbe important; while, over time, factors suchas environmental tolerances and competi-

tion become more important.

While most species established success-fully, there was considerable variabilityamong species that is not explained byabundance in the seed mix alone. Resultsfrom grassland restoration in Europehave also shown considerable variability(Pakeman et al. 2002), but grasses andgeneralist species establish better thanforbs and specialists (Pywell et al. 2002;Lawson et al. 2004). Grasses establishedbetter than forbs overall at this site, butthere was considerable variation amonggrass species.

Species establishment in grassland res-toration can vary among regions so it is

Table 1. Establishment success of species included in the seed mixture in the planted grassland.


important to document species establish-ment in the Northeast. For example, somespecies (e.g., Schizachyrium scoparium(Michx.) Nash and Lespedeza capitata Michx.) established successfully both inthis study and in a restoration site in Kansas(Foster et al. 2009). However, Andropogon gerardii Vitman was more successful andSorghastrum nutans (L.) Nash and Andro-pogon virginicus L. much less successfulin Connecticut than in Kansas. Asclepias tuberosa L. established successfully inmore than a third of the plots in Kansasand did not establish at all in Connecticut.Establishment can also be greatly impactedby climate and weather conditions or site-specific factors (Hellstrom et al. 2009), sothis study is not a definitive evaluation ofestablishment success in the Northeast.Documentation of species establishmentat multiple sites is necessary to understandwhich species will be generally effective forgrassland restoration in the Northeast.

Despite successful establishment of themajority of the 23 sown species, nativerichness and diversity were no higher in the

planted grassland than non-planted areas.The effects of seed sowing on grasslandrichness have been mixed, with some stud-ies finding clear increases (Foster et al.2007; Foster et al. 2009), and others findingno effect (Piper et al. 2007; Rydgren et al.2010) or even decreases (Lepš et al. 2007).While sowing clearly added species to thegrassland system in this study, these speciesmay have simply replaced other species thatmight have naturally colonized.

Although the planted grassland did notdisplay increased richness, it did have thelowest level of invasion by exotic and in-vasive species. Several studies have foundthat seed sowing can reduce invasion byexotic species, but that native colonizationby unsown species is also often reduced(Lawson et al. 2004; Lepš et al. 2007; Töröket al. 2010). Reduced natural colonizationmay explain the similarity in species rich-ness among grasslands in this study.

Large or fast growing grass species areoften particularly effective at reducinginvasions of exotic species in grasslandrestoration, but they can also lead to re-duced diversity and low colonization by

native forbs (Fagan et al. 2008; Török etal. 2010). Because forb species providemost of the diversity in many grasslands(McCain et al. 2010), strong dominanceby grasses may reduce diversity. Evenif diverse seed mixes are used, diversitymay be low if aggressive sown grassescome to dominate the site (Lawson et al.2004). Andropogon gerardii is becomingdominant and is more abundant at this sitethan in other local grasslands, and thus maybe reducing potential diversity. Competi-tive dominance by large grasses may bemore pronounced when using commercialvarieties of these species rather than localecotypes. For example, available ecotypesof Panicum virgatum L. have been aggres-sive in previous plantings (G. Dreyer, pers.observation) and were not included in theseed mix for the current study. The use ofwidely available and inexpensive varietiesof large-statured grasses in grassland res-toration in the Northeast must be carefullyconsidered.

Because seed sowing can reduce bothinvasion of exotics and native diversity,it is important to consider if it is a neces-sary or desirable management approach

Figure 2. (A) Proportion of the total cover thatconsists of species that were sown into the plantedgrassland. (B) Proportion of total vegetation thatis grasses and related species (graminoids). Errorbars indicate SE.

Figure 3 Vegetation characteristics of the threegrasslands. Total cover (A) can add to more than100because species are overlapping. (B)Number ofspecies per 1 m2 plots. Error bars indicate SE.

Figure 4. Cover of all exotic species (A) andwoody species (B) in the three grasslands. Errorbars indicate SE.


Table 2. Percent cover of plant species with at least 1% cover in at least one of the three sites. Species in bold were included in the planted seed mix-ture.


in restoring grasslands in the Northeast. Ifexisting grasslands with native species arein close proximity to the restoration site,natural colonization will likely occur andseed sowing may not be necessary (Pärtelet al. 1998; Fagan et al. 2008). At our site,common species such as Schizachyrium scoparium and Solidago spp. colonizedunplanted areas rapidly. However, rareand locally uncommon species will notestablish without being sown. Sowing seedmay also reduce the need for exotic inva-sive species control. Thus the landscapecontext of the restoration site needs to beconsidered (Ruprecht 2006).

We recognize that this study is not a for-mal scientific comparison of replicatedsown and unsown plots. Thus, differencesamong grasslands may reflect underlyingsite differences in addition to managementdifferences. However, this study imitatedwhat a land manager might do in restor-ing grassland habitat and documented theresults. Results from this study indicatea successful pathway towards grasslanddevelopment dominated by native species.Additionally, the restored sites appear tobenefit open-site animal species, provinghabitat for winter-resident sparrows (fam-ily Emberizidae) and helping sustain localpopulations of blue-winged warbler (Vermi-

vora cyanoptera), song sparrow (Melospiza melodia), and eastern bluebird (Sialiasialis; R. Askins, Connecticut College,pers. comm.). Given the scarcity of moni-toring grassland restoration in northeasternNorth America, these results will providea baseline for comparison and a source ofinformation for land managers who con-tinue to restore early successional habitatin the Northeast.

ACKNOWLEDGMENTS

We thank the Stewart B. McKinneyNational Wildlife Refuge, U.S. Fish andWildlife Service, for the use of the trac-tor and seed drill used in this study, theConnecticut Department of Environmen-tal Protection for the staff who operatedthe planting equipment, and the USDANRCS Natural Resources ConservationService, which supported establishment ofthis grassland with their Wildlife HabitatIncentive Program.

Chad Jones is an Assistant Professor of Botany and Environmental Studies at Con-necticut College. His research interests include invasive species, plant coloniza-tion, and restoration ecology.

Glenn Dreyer is the Charles & Sarah P. Becker ‘27 Director of the Connecticut College Arboretum, Executive Director of the college’s Goodwin-Niering Center for the Environment, and an Adjunct As-sociate Professor of Botany. His research interests include vegetation management, invasive species biology, and big and historic trees.

Nels Barrett is the State Ecologist for the Natural Resources Conservation Service in Connecticut. Much of his work focuses on applied aspects of vegetation science as well as habitat management and res-toration.

LITERATURE CITED

Askins, R.A. 1997. History of grasslands in thenortheastern United States: implications forbird conservation. Pp. 119-136 in P.D. Vick-ery, and P.W. Dunwiddie, eds., Grasslands ofNortheastern NorthAmerica. MassachusettsAudubon Society, Lincoln.

Bakker, J.D., and S.D. Wilson. 2004. Usingecological restoration to constrain bio-logical invasion. Journal ofApplied Ecology41:1058-1064.

Berger, J.J. 1993. Ecological restoration andnonindigenous plant species: a review.Restoration Ecology 1:74-82.

Table 2. Continued


Bisteau, E., and G. Mahy. 2005. Vegetation andseed bank in a calcareous grassland restoredfrom a Pinus forest. Applied VegetationScience 8:167-174.

Dzwonko, Z., and S. Loster. 2007. A functionalanalysis of vegetation dynamics in aban-doned and restored limestone grasslands.Journal of Vegetation Science 18:203-212.

Fagan, K.C., R.F. Pywell, J.M. Bullock, andR.H. Marrs. 2008. Do restored calcareousgrasslands on former arable fields resembleancient targets? The effect of time, methodsand environment on outcomes. Journal ofApplied Ecology 45:1293-1303.

Foster, B.L., K. Kindscher, G.R. Houseman,and C.A. Murphy. 2009. Effects of haymanagement and native species sowing ongrassland community structure, biomass,and restoration. Ecological Applications19:1884-1896.

Foster, B.L., C.A. Murphy, K.R. Keller, T.A.Aschenbach, E.J. Questad, and K. Kindscher.2007. Restoration of prairie communitystructure and ecosystem function in anabandoned hayfield: a sowing experiment.Restoration Ecology 15:652-661.

Foster, D.R., and G. Motzkin. 2003. Interpret-ing and conserving the openland habitatsof coastal New England: insights fromlandscape history. Forest Ecology andManagement 185:127-150.

Hellstrom, K., A.P. Huhta, P. Rautio, and J.Tuomi. 2009. Seed introduction and gapcreation facilitate restoration of meadowspecies richness. Journal for Nature Con-servation 17:236-244.

Jongepierová, I., J. Mitchley, and J. Tzanopou-los. 2007. A field experiment to recreatespecies rich hay meadows using regionalseed mixtures. Biological Conservation139:297-305.

Kettenring, K.M., and C.R. Adams. 2011.Lessons learned from invasive plant con-trol experiments: a systematic review andmeta-analysis. Journal of Applied Ecology48:970-979.

Lawson, C.S., M.A. Ford, and J. Mitchley.2004. The influence of seed addition andcutting regime on the success of grasslandrestoration on former arable land. AppliedVegetation Science 7:259-266.

Lepš, J., J. Doležal, T.M. Bezemer, V.K. Brown,K. Hedlund, A.M. Igual, H.B. Jörgensen,C.S. Lawson, S.R. Mortimer, A. Peix Geld-art, C. Rodríguez Barrueco, I. Santa Regina,P. Šmilauer, and W.H. van der Putten. 2007.Long-term effectiveness of sowing high andlow diversity seed mixtures to enhance plantcommunity development on ex-arable fields.Applied Vegetation Science 10:97-110.

Lezberg, A.L., K. Buresch, C. Neill, and T.Chase. 2006. Mechanical land clearing topromote establishment of coastal sandplaingrassland and shrubland communities. Res-toration Ecology 14:220-232.

Marks, P.L. 1983. On the origin of the fieldplants of the northeastern United States.American Naturalist 122:210-228.

McCain, K.N.S., S.G. Baer, J.M. Blair, andG.W.T. Wilson. 2010. Dominant grassessuppress local diversity in restored tallgrassprairie. Restoration Ecology 18:40-49.

[NRCS] Natural Reources ConservationService. 2011. Web Soil Survey. USDANatural Resources Conservation Service.Available online <http://websoilsurvey.nrcs.usda.gov/app/>.

Pakeman, R.J., R.F. Pywell, and T.C.E. Wells.2002. Species spread and persistence:implications for experimental design andhabitat re-creation. Applied Vegetation Sci-ence 5:75-86.

Pärtel, M., R. Kalamees, M. Zobel, and E.Rosén. 1998. Restoration of species-richlimestone grassland communities fromovergrown land: the importance of propa-gule availability. Ecological Engineering10:275-286.

Piper, J.K., E.S. Schmidt, andA.J. Janzen. 2007.Effects of species richness on resident andtarget species components in a prairie resto-ration. Restoration Ecology 15:189-198.

Piqueray, J., G. Bottin, L.M. Delescaille, E.Bisteau, G. Colinet, and G. Mahy. 2011.Rapid restoration of a species-rich ecosystemassessed from soil and vegetation indicators:the case of calcareous grasslands restoredfrom forest stands. Ecological Indicators11:724-733.

Pywell, R.F., J.M. Bullock, A. Hopkins, K.J.Walker, T.H. Sparks, M.J.W. Burke, andS. Peel. 2002. Restoration of species-richgrassland on arable land: assessing the limit-ing processes using a multi-site experiment.Journal of Applied Ecology 39:294-309.

Pywell, R.F., J.M. Bullock, J.B. Tallowin, K.J.Walker, E.A. Warman, and G. Masters.2007. Enhancing diversity of species-poorgrasslands: an experimental assessment ofmultiple constraints. Journal of AppliedEcology 44:81-94.

Ruprecht, E. 2006. Successfully recoveredgrassland: a promising example fromRomanian old-fields. Restoration Ecology14:473-480.

Rydgren, K., N. Jørn-Frode,A. Ingvild,A. Inger,and H. Einar. 2010. Recreating semi-naturalgrasslands: a comparison of four methods.Ecological Engineering 36:1672-1679.

Sheley, R.L., and M.L. Half. 2006. Enhancingnative forb establishment and persistenceusing a rich seed mixture. Restoration Ecol-ogy 14:627-635.

Török, P., B. Deák, E. Vida, O. Valkó, S.Lengyel, and B. Tóthmérész. 2010. Restor-ing grassland biodiversity: sowing low-diversity seed mixtures can lead to rapidfavourable changes. Biological Conservation143:806-812.

Tucker, G.C. 1995. The vascular flora ofsoutheastern Connecticut. Memoirs of theConnecticut Botanical Society No. 3, Con-necticut Botanical Society, New Haven.

Vickery, P.D., M.L. Hunter, and S.M. Melvin.1994. Effects of habitat area on the distribu-tion of grassland birds in Maine. Conserva-tion Biology 8:1087-1097.

Documents

Evaluating the Success of Seed Sowing in a New England Grassland Restoration