Effects of plant invasions on the species richness of ......Effects of plant invasions on the species richness of abandoned agricultural land. - Ecography 24: 633-644. While exotic

ECOCRAPHY 24: 633-644. Copeiihagt'ii 2(K)I

Effects of plant invasions on the species richness of abandonedagricultural land

Scott J. Meiners, Steward T. A. Pickett and Mary L. Cadenasso

Meiners, S. J., Pickett. S. T. A. and Cadenasso, M. L. 2001. Effects of plant invasionson the species richness of abandoned agricultural land. - Ecography 24: 633-644.

While exotic plant invasions are thought to lead to declines in native species, thelong-term impacts of such invasions on community structure are poorly known.Furthermore, it is unknown how exotic plant invasions compare to invasions bynative species. We present data from 40 yr of continuous vegetation sampling ol~ 10fields released from agriculture to examine the effects of invasions on species richness.The effects of both exotic and native species invasions on species richness werelargely driven by variations among fields with most species not significantly affectingspecies richness. However, invasion and dominance by the exotics Agropyron repens,Lonicera juponlca. Rosa imillifiora. Trifuliiim praiense and the native Solidagocanadensis were associated with declines in richness. Invasions by exotic and nativespecies during old field succession have similar effects on species richness withdominance by species of either group being associated with loss of species richness.Exotic species invasions tended to have stronger effects on richness than nativeinvasions. No evidence was found of residual effects ol~ invasions because the impactsof the invasion disappeared with the decline of the invading population. Whenpooled across species, heavy invasion by exotic species resulted in greater loss ofspecies richness than invasion by native species. Studies of invasion that utilizemultiple sites must account for variability among sites. In our study, had we notincluded field as a factor we would have incorrectly concluded that invasionconsistently resulted in changes in species richness.

S. J. Mtincr.s (cfsjm2(<i,c'iu.edii), Depi tij Biological Sciences. Eastern Illinois Univ., 600Lincoln Ave.. Charleslon. IL 61920-3099. USA. S. T. A. Pickett and M. L.Cadenasso. Inst. of Ecosystem Studie.^i. Box AB. Miltbrook. NY 12545-0129, USA.

The widespread invasion of exotic plant species is a (Baker 1965, Bazzaz 1996. Rcjmanek 1999), conditions

major concern from both the perspectives of conserva- that make communities subject to invasion (Peart andtion and ecological research. Exotic invasions are often Foin 1985, Fox and Fox 1986, Kotanen et al. 1998,associated with 1) declines in local plant diversity Stapanian et al. 1998, Knops et al. 1999) and the

{Richardson et al. 1989, Woods 1993, Wyckoff and subsequent impact of invasions on community structureWebb 1996, McCarthy 1997, Hutchinson and Vankat (Richardson et al. 1989, Hutchinson and Vankat 1997,1997, Christian and Wilson 1999). 2) declines in forest McCarthy 1997, Woods 1997, Parker et ai. 1999).regeneration (Thomas 1980, Woods 1993, WyckotTand Three factors limit our understanding of the impacts

Webb 1996, Hutchinson and V:inkat 1997) and 3) of exotic species invasion into plant communities: 1) areductions in the productivity of agricultural land {Pi- lack of information comparing native and exotic speciesmentel et al. 2000). In addition, exotic invasions often invasions. 2) the confounding effects of disturbance

accrue the financial costs of biological or chemical history, and 3) the limitations of correlational ap-control (Pimcntcl et al. 2000). Ecologists struggle to proaches. Most ecological studies of invasive plantunderstand the factors that make species invasive species inadequately address these problems, potentially

Accepted 12 February 2001

Copyright © ECOGRAPHY 200]ISSN 0906-7590Printed in Ireland - all rights reserved

ECOGRAPHY 24:6 |2(KII) 633

leading to the development of inappropriate conclti-sions and management priicUces.

An unresolved question facing ecologists is whetherexotic and native plant species differ fundamentally intheir function within plant communities. In experimen-tal comparisons of native and exotic plants, exoticspecies are generally found to be competitively superior(D"Antonio and Mahall 1991, Dillenburgh et al. 1993,Fogarty and Facelli 1999, Hamilton et al. 1999. Marlerel al. 1999), have faster and more plastic growth(Schiercnbeck ct al. 1994, Bazzaz 1996. Schweitzer andLarson 1999, Milberg et al. 1999) and higher reproduc-tive capacities (Rcjmanck 1989, Bazzaz 1996). Becausethese comparative studies foctis on problematie exoticspecies, they may provide biased information on therelative performance of native and exotic plant speciesin general. Systematic comparisons of native and exoticspecies at a community-wide scope arc needed to ade-quately eompare these two groups.

The frequency and intensity of disturbance is one ofthe principal factors considered to regulate comrntmityinvasibility (Fox atid Fox 1986, Hobbs 1989, Kolanenet al. 1998). However, an association between distur-bance and invasion makes it difficult to separate theeffects of an exotic species from the effects of thedisturbance that originally led to the invasion (Woods1997). Because exotic plants commonly dominate aban-doned agricultural land (Bard 1952, Pickett 1982, Rej-manek 1989. Bazzaz 1996. Bastl et al. 1997).successional studies, with continuous species invasionsand losses, provide a tmiqtie opportunity to test the roleof exotic species in determining community structureand dynamics under a known disturbance histot"y.

The litTiited temporal duration of tnost studies con-strains our understanding of the function of exoticspecies in plant communities {Lodge 199.1. Fike andNiering 1999. Smith and Knapp 1999. Parker et al.1999). Inferences about long-tertn impacts and popula-tion persistence based on short-term investigations maynot accurately predict the influence of exotic invaderson plant communities. Eeologists have attempted toaddress the limitations of short-term studies through achronosequencc approach, where areas of differing lev-els of invasion are assumed to represent areas differingin time since invasion. Greater abundance of an invaderindicates longer times since initial invasion. This ap-proaeh often uses correlations to infer ecological im-pacts of the exotic species invasion (Thomas 1980,Richardson ct al. 1989. 'Woods 1993. Hutchinson and'Vankat 1997). Beeause of the long time spans involvedin most sueeessions, space-lbr-time substitutions arecommonly used (Glcnn-Lewin and van der Maarel1992). Critical assumptions of the chronoscquence ap-proach are that I) all sites will be. or have beensubjected to similar processes, and that 2) the patternsamong sites reflect temporal change (Pickett 1989,Glenn-Lcwiu and van der Maarel 1992). These assump-

tions may not be met and ehronosequences may suggestquite different patterns from those revealed from long-term studies (.lackson et al. 1988, Pickett 1989). Pat-terns revealed in ehronosequencc studies may reflectassociations caused by disturbance history or other sitecharacteristics unrelated to invasion (Woods 1997). Forthis reason, long-term studies of permanent plots arefar superior to chronosequence studies iu verifying thepotential impacts of exotic species on plant communitystructure.

To address the three limitations outlined above, weutilized 40 yr of permanent plot data, determining theimpacts of both native and exotic invasions on thestructure of successional plant communities. Plot datawere derived ft"om 10 abandoned agricultural fields ofknown disturbance histories in the piedmont region ofNew Jersey. USA. All species abundant enough foranalysis were included to provide an unbiased compari-son of both the native and exotic assemblages. Thisstudy addresses the following questions. 1) How doesinvasion by native or exotic species affect richness insuccessional communities? 2) Are species effeets ondiversity consistent among fields? 3) Do native andexotic speeies invasions have dilTerent effects on speciesrichness? 4) Do residual effects of species invasionspersist after the invading poptilation declines?

Methods

Data collection

Since 1958. the vegetation of abandtined agriculturalfields at the Hiitcheson Memorial Forest Center(HMFC) has been monitored with the use of perma-nent plots (Small et al. 1971, Pickett 1982). Tliis studyis known as the Buell-Small Succession Study (BSS).named for the project initiators (www.ecostudies.orgXbss). Within each of 10 fields, 48 plots are arrangedin a regular pattern thai varies somewhat with theshape of the field. Most of these fields abut the old-growth forest of HMFC. Fields were abandoned aspairs in alternate years from 1958 until 1966. Sincerelease, data collection occurred every year until 1979,when sampling was switched to alternate years. At eachsampling, the percent cover of all speeies present ineach permanently marked 0.5 x 2.0 m plot is recorded.These data represent the longest continuous data set onold field successional ehange known.

Statistical analyses

Data were summarized at two spatial scales in thisstudy, the plot and field scale. To determine the tempo-ral pattern of each species invasion, data were pooledacross all fields. All subsequent analyses were con-

634 F.COCiRAPIIY 24:(< (20011

ducted at the plot scale within invaded fields. Theseanalyses included field as a categorical variation toaccount for variation among fields. To determine theeffects of plant invasion on species riehness, the changein total species richness of plots was related to changein cover of the invading species.

For each species, the temporal pattern of plant coversummed across all fields was examined and a sarnplingperiod was chosen that approximated the initiation ofthe species invasion (tj,.,;,! ,) and extended to the peakcover of the invading species (tp ..,,,). The value of t,,,,,!,,,was set at the time when cover of the invading specieswas ^ 5% or less, or at year 1 for species invadingimmediately after abandonment. The value of tp . ^ wasset at the year of maximum cover for the invadingspecies. When tj,,j|j,,i and t^^.,y. were > 10 yr apart, t ..,,was set as tinnj i + n, to limit changes in communitystructure that were external to the invasion. Only fieldsin which cover of the invading species at tp ,, , averaged> 1% per plot were included in the analyses. Withininvaded fields, only those plots that were invaded attpeak were included. This removes from comparisonlocal areas that may not be suitable for invasion by aspecies or areas that have not received sufficient seed toresult in an invasion (Treberg and Husband 1999),thereby focusing the analyses on the effects of invasion.

The increase in cover (C) of the invading species(* peak ~ C|,,,ii.j|) and the change in species richness(Speak ~ Sjn ji,,]) was calculated for each plot. By calcu-lating change in richness and cover over time, weaccount for successional changes in vegetation overtime. Plots were separated into three invasion classesbased on change in cover of the invading species: light(0.5-33% cover), moderate {33.5-66'Xi cover) andheavy (66.5-100'X> cover). Plots that had a decrease incover of the invading species from tj,,i,jy| to tp ak weredropped from analysis. All species that had at least 5plots representing each invasion class were analyzed fortheir impacts on species richness. The influence ofinvasion class on change in species richness was evalu-ated with ANOVA (Anon. 1989),

To compare the effects of native and exotic speciesinvasions on community diversity, the average changein species richness from ti,,njj| to tp ,,,, was calculated ineach invasion class for all plots within a field. This wasdone separately for each species within each field toaccount for differences among fields. These values werethen analyzed with ANOVA with species nested withinnativity (Anon, 1989),

Species that exhibited a pronounced decline in abun-dance within 10 yr of tpe ^ were further analyzed for thepresence of residual effects of the invasion on speciesrichness, A sampling time was selected that had asubstantial decrease in cover of the invading species(t,,j-, ,) but that was temporally as close to tp .,,, aspossible to minimize external influences on richness.The value of t^^^^ had at least a 50'/<i decline in cover of

the invading species. The same invasion classes wereused as in the previous analysis, substituting changes inspecies richness from l,,,,i,i,t to t ner as the responsevariable. Residual efTects of the invasion on speciesrichness were evaluated with ANOVA (Anon, 1989).

Results

The fields contained I 17 exotic and 216 native speciesover the entire 40-yr span of data collection. A total of12 exotic and 14 native species were sufficiently abun-dant to meet all criteria for analysis of their influenceson species richness (Table I). These species included awide range of life forms from annuals to woody vines,and ranged temporally from immediately post-aban-donment to 40 yr after release from agriculture. Invad-ing species cover increased on average at least oneorder of magnitude from ti,,!,;.,, to tp .., ;. Fifteen species(9 exotic and 6 native) had sufficient population de-creases to allow for analysis of residual effects of theinvasion on species richness-

Invasions by most species did not reduce the speciesrichness of plots. In this study, the vast majority (90%exotic, 94'/<i native) of species did not attain sufficientabundance within the community to have testable influ-ences on diversity. Of those that became abundant,only 5 species (1.5'Mi of the total species pool) wereassociated with significant declines in species richnessonce site factors were accounted for. In this survey,97'%* of exotic plant invasions did not result in a mea-surable effect on species richness in these communities.

Exotic species generally elicited a decrease in Sp ,| -SjiiiULii w'ith increasing severity of invasion (Fig. 1). Mostof these invasions (8 of 12) resulted in a net loss ofspecies over the course of the invasion, with the mostspecies lost from the high invasion class. ANOVAresults indicate that only four of these species hadsignificant effects of invasion class when differencesamong fields were aeeounted for (Table 2). Field wasgenerally the most important factor in the analysis.When variation among fields in invasion and speciesrichness was not accounted for, the invasion class oftenbecame significant- The relationship between change inspecies richness and plant invasion was generally con-sistent among fields. Two species, Barharea vulgar'is andChrysanthemum leucanthemum. were exceptions, with asignificant invasion x field interaction. Both of thesespecies had invasions associated with decreases in spe-cies richness in some fields and increases in others.

Native species had more varied responses of speciesrichness to their invasion than did exotie species (Fig,2). Most species (8 of 14) show some effects of invasionon change in species richness, with the high invasionclass having the most species lost or the fewest species

HCOGRAPHY 24:6 (20(11) 635

gained. These effects became non-significant with theinclusion of field in the ANOVA {Table 2). Only onespecies, Solidago canadensis. had a significant effect ofinvasion class on richness change, with plots containingmoderate invasions losing 4 species on average whilethe other two classes gained a species. Differencesamong fields accounted for the majority of the varia-tion, and again accounts for most of the invasioneffects seen in Fig. 2. Two species. Aster pilosus andEupalorium rugosum. had a significant invasion x fieldinteraction. For both species, half of the fields hadinvasions associated with increases in species richnessand the other half had invasions associated with de-creases in species richness.

Native and exotic species have similar effects onchange in species richness over successional time (Table3, Fig. 3), with species richness decreasing with increas-ing invasion class. However, invasion by exotic specieshad an overall larger effect on change in species rich-ness than invasion by native species (a significant nativ-ity effect). The invasion class of both native and exoticspecies had similar effects on species lichncss as seen bythe non-significant interaction term.

There is no evidence that the invasion of plots byeither native or exotic species has any long-term effectson species richness (Table 4, Fig. 4). As in the previousanalyses, field identity accounted for the majority of thevariation within the data. Once the effects of field wereaccounted for in the ANOVA. invasion class becamenon-significant for all species. Chrysanlhemtim leucan-theimmi had a significant invasion x field interaction,with one field showing a positive association betweeninvasion and richness and another field a negativeassociation. The other four fields did not have signifi-cant relationships.

Discussion

Invasion effects on richness

Invasions by most species in this post-agricultural suc-cession did not have a significant infiuence on speciesrichness (Williamson and Fitter 1996). However, a fewspecies had very strong effects on community richness.This result is in contrast to most field studies, whichfind significant negative associations between exotic

Table 1. Summary information on the populations of~ exotic and native species used in the analysis of invasion effects on speciesrichness. Cover values based on invaded fields and plots only (see text for invasion criteria), t, is years since release fromagriculture. Life form abbreviations: A = annual, AG = annual grass, B = biennial or short-lived perennial, P = perennial,PG = perennial grass, S = shrub. V = woody vine. Nomenclature, life form and nativity status follows Glcason and Cronquist(1991).

Exotic speciesAgropvron repensBarbarea vulgar isBromus racemo.susChrysanthemum leucanlhcimimDaucus carolaHieracium cacipito.sumLonicera JaponicaPoa pratcnsisRosa multi/loraRum ex ucetoseltaTrifolium hyhridumTrifolium pratense

Native speciesAmbrosia artemisiifotiaAster pilosusCalystegia sepiumErigeron annuusEupatorium rugosumEuthamia graminifotiaFragaria rirginianaFartlicnocissus c/uinc/uejoIiaI'lanlago rugcliiPotent it tu simplexSolidago canadensisSolidago JunceaSolidago rugosaToxicodendron rinJicans

Lifeform

PGBAGPBPVPGSPPB

ABPAPPPVPPPPPV

Number offields

38168

10108

10724

5888299863996

10

Initial

l,,>uu,l

11421454

10111

1211

3044

1013

105

1010

conditions

Cover

3.54.4LO0.90.55.40.22.02.50.91,1L2

13.50.10.92.12.62.00.9LI

12.81.10.70.41.70.9

Peak

tpe.,k

52

1064

10151020344

21062

40101020

31020152020

of invasion

Cover

39.027.752.824.016.741.241.844.232.720.025.329.2

48.923.814.522.127.624.332.919.433.640.616.118.530.621.5

After

I .LIUT

104

-101015-15-101010

515-

3--20

515---—

invasion

CoverI'i'i.)

0.81.4

-5.13.4

15.4-22.1-0.90.61.4

0.512.1

5.9

-8.2

-8.32.3

---—

636 LCOGItAI'HV 24:6 (2001)

Eig. 1. Change in speciesrichness in association withinvasion by exotic plantspecies. Data for invadedplots only. Sampling timesfor each species listed inTable 1. Change in cover ofinvading species classes are:L = light, M = moderate,H = heavy.

nCO

co

o

cCO

-5

5

0

-5

-5

Agropyron repens

C. leucanthemum

Lonicera japonica

Rumex acetosetta

Barbarea vulgaris

Daucus carota

Poa pratensis

Trifotium hybridum

Bromus racemosus

Hieracium caespitosum

Rosa multiflora

Trifotium pratense

L M H L M H L M H

Change in cover of the invading species

species and community richness (Richardson ct al.1989, Woods 1993, Wyckoff and Webb 1996. Mc-Carthy 1997, Hutchinson and Vankat 1997, Christianand Wilson 1999, Meiners and Pickett 1999, but secWiser et al. 1998, Treberg and Husband 1999). Thisdifference is probably caused by the focus of fieldstudies on species that are thought a priori to havestrong impacts on plant communities. Community-widestudies should show much lower incidence of impacts(e.g. Mystcr and Pickett 1992, this study).

The effects of the invasion generally increa.se withincreasing cover of the invading species. This pattern isseen in both the native and exotic species studied. Thissuggests that ecological impacts of invasions arc moredependent on the dominance in cover of a patch by aspecies rather than the presence of the species (Richard-

son et al. 1989. Woods 1993, Morgan 1998). Onedifficulty in comparing our result to those in the ecolog-ical literature is the variation in methodologies used.Studies that use cover or dominance of exotic species toassess invasion impacts tend to find negative associa-tions between invasion and diversity (Richardson ct al.1989, Woods 1993, Hutchinson and Vankat 1997,Christian and Wilson 1999, but see Treberg and Hus-band 1999). In contrast, studies that relate speciesrichness of exotics to species richness of native speciestend to find positive associations (Stohlgren et al. 1998,Lonsdale 1999, Smith and Knapp 1999). Analysesbased on cover of the invader seem more appropriatefor assessing the impacts of invasions because theyincorporate estimates of the abundance of the invaderrather than presence alone.

ECOGRAPHY 24:6 (201111 637

The negative association between exotic plants andcommunity diversity has been used to suggest thatinvading exotic species do not merely fill vacantniches in natural communities (Tilman 1997), or re-place native species one-for-onc, but that they dis-place speeies disproportionately from the community,lowering diversity. This tnechanisrn is often cited infield studies of invasive exotic plants, cither as corre-lational or as anecdotal information, but has notbeen experimentally tested. Problematic exotic speciesare generally found to be competitively superior tonative species in two-species competition experiments(D'Antonio and Mahal! 1991, Dillenburgh et al. 1993.Marler et al. 1999), However, it is not knownwhether exotie species are on average competitivelysuperior to native plant species, which would be nec-essary to result in lowered diversity within a commu-nity. The competitive displacement of native plantspecies is often used as justification for the eradica-tion of individual exotic species, although the per-ceived relationship with diversity has not beendemonstrated (Anderson 1995, Hager and McCoy1998, Treberg and Husband 1999). Most exotic spe-cies have only trivial impacts on community structure,bccotning tninor cotnponcnts of the plant community,increasing local species diversity (Williamson and Fit-ter 1996, Parker et al, 1999. this study, Meiners et al,unpubl.).

Species characteristics

The species with significant effects on richness rangedacross most of the life histories tested in this study.Four of the 5 speeies are long-lived perennial herba-ceous or woody species that have become problematicweeds of agricultural or natural areas. The exception tothis is Trifolium pralense. a biennial to short-livedperennial species that is widely planted because of itsnitrogen fixation capabilities, Agropyron repens is aweed of agricultural land throughout North Americaand Europe that spreads rapidly in pastures (Palmerand Sagar 1963). Lonicera japonica and Rosa multiftoraarc widespread woody species of Asian origin thatdominate disturbed land in the eastern United States(Thomas 1980, Amrine and Stasny 1993), SoHdagocanach'nsis. a native Norlh American speeies, has be-come a pt"oblematic weed in European agriculturalareas (Weber 1997), All of these species are widespread,have high reproductive capacities, are capable of form-ing monospecific patches, and are associated with agri-cultural practices. While these characteristics are similarto those of many successful exotic invasions (Baker1965, Rejmanek 1989. Bazzaz 1996), there is no charac-teristic that would differentiate these species a priorifrom the larger pool tested. The only generalizationabout life history characteristics that can be made isthat annuals do not appear problematic in this system.

Table 2, Summary of 2-way ANOV.As on the effeets of exotic and native species invasion on change in species richness(Sp,,;.i,-S,,,,,,.,|) in old fields, * = p < 0 . 0 5 , * * - p < 0 , 0 1 , *** -p<O.OOf .

Invas ion (1) Field (F) I x F Mode! R-

Exotic speciesAgropyron repensBarharea vulgari.\Bromus rucer}io.siisChrysaiilheniuni IcucaiilhemwnDaucus ciirotalUeracium cucspiKisuniI.onieera JaponicaPoa praten.sisRosa nitdiifloraRumex aceloicllaTrifolium pratenseTrifolium hybridum

Native speciesAmbrosia artcmisiifoliaAster pilosiisCalysicgiu sepiumErigt'ron annuusEiipaloriiini rugosumEulhamia graininifoliaFragaria virginianaPin'lhcnocixsiis i/uique/oiiaPiaiilago rugcliiPotentilla simplexSolidago canudensisSoli dago JunceaSI dida go rugosaToxicodi'ndron rudiearis

638

3,43*1,070,]l0,130,092,57

10,25***0,908.73***1-23

10.89***0.11

1.892,770.140,822,100,491.681,580.080,583.29*0.231.300.39

8,83***7,71***

-17,64***12.27***15.20***6,86***5.40***1,96*9,05***7,70***1,08

LJ - ^ 2 * * *

17^88***20,10***

9,08***8,14**6,35***5,08***3-91***

27.82***0.581.966.86***1.310.81

0-012.15*-3.50**0,910,721,230,941,420,771,30-

1,712,10*0,191,723,52*1-081-001,050,491,360,540,850,421,49

0.600,400,130,560,500,430,460,320,290,460,370,03

0,480-500,530,300,210,390,410,250,480,190,200,400,190,23

RCXXiR.APllY 24:6 (2Q01)

Fig. 2, Change in speciesrichness in associationwith invasion by nativeplant species. Data forinvaded plots only.Sampling times for eachspecies listed in Table I.Change in cover ofinvading species classesare: L = light,M = moderate.H = heavy.

-5

CO0)c

0)O

c

— -5

CO

6 ^-5

Ambrosia artemisiifolia

Erigeron annuus

Fragaria virginiana

Potentilla simplex

Solidago rugosa

Aster pilosus

Eupatorium rugosum

P. quinguefolia

Solidago canadensis

Calystegia sepium

Euthamia graminifolia

Plantago rugeiii

Solidago juncea

T. radicans

L M H L M H

Change in cover of the invading species

Differences among fields

While the fields of the Buell-Smail Succession Study areadjaeent to each other, the differenees in pre-abandon-ment history have resulted in the development of dra-matieally different vegetation (Myster and Piekett1990). This variation in vegetation and disturbancehistory resulted in the dominance of site effects in al!

analyses. Analyses across fields often resulted in signifi-cant negative associations between species richness andinvasion, even though there v^as no pattern evidentwithin each field. The common practice of evaluatinginvasion effects by comparing sites differing in level ofinvasion may also lead to such ineorrect conclusions.This problem is analogous to the drawbacks found in

ECOGRAPHY 24:6 (2(1111) 639

Table 3. Nested ANOVA summarizing the influence of nativ-ity and invasion class on the impacts of invasion on plotspecies richness.

FieldSpecies (nativity)Invasion classNativitvInvasion class x nativitvFrror

DF

924

12

378

MS

114.2331.90

131.7493.3419.49

F

9.302.60

10.727.601.59

P

0.00010.00010.00010.00610.2062

0)

co(/)aj"oQ)Q.to

(DCD

5O

L M H L M H

Native ExoticFig. 3. Summary of change in species richness in associationwith plant invasion by native and exotic species. Invasionclasses ;ire: L = light. M = moderate, H = heavy.

using space for time substitutions in studying succes-siona! processes through time (Pickett 1989).

A few species showed inconsistency of effects amongsites, as shown by a significant field x invasion interac-

tion. While the effects of invasion cUtss were non-sig-nificant, these results illustrate the importance ofamong-sitc variation in determining the infiuence of aninvasion on community structure. Not all sites andspecies invasions will result in similar community-wideeffects. The identity of the invading species and thecomposition of the existing community are both impor-tant in determining the success and impact ol" theinvasion (Lodge 1993).

Native vs exotic invasions

Native and exotic species appear to function in similarways within plant communities, exhibiting similar infiu-ences on species richness in this system. However, themagnitude of the depression of species richness wasgreater for exotic than for native invasions. The greaterimpact of exotics within this study may be indicative ofan overall competitive superiority of dominant exoticsin comparison to dominant native species as is sug-gested by competition experiments (D'Antonio andMahall 1991, Dillenburgh et al. 1993. Marler et al.1999). The evoluiionary history of the exotic species ineastern North America may explain these differences incompetitive ability. Many oi" these species arc adaptedto persistence in highly disturbed Eurasian agriculturalsystems and generally have much stronger impacts onagricultural productivity than native agricultural weeds(Pimentel et al. 2000). Native species, in contrast,largely evolved in persistent openings within forestcommunities (Marks 1983).

While there may be a negative relationship betweenexotic species invasion and diversity at small scales(Stohlgren et al. 1998, 1999. this study), at larger scalesnative and exotic species are often positively associated(Stohlgren et al. 1998, 1999. Wiser et al. 1998, Lonsdale1999, Levine and D'Antonio 1999, Smith and Knapp

Table 4 Summary of 2-vvay ANOVAs on the residual effects of exotic and native speeies invasion on ehange in species riehnessafter decline of th'e population (S,,,,,,-Si,,,,,,,). * = p<().05, ** = p<0.()!. *** = p<O.OOI.

Invasion (I) Field (K) I x F Model

Exotic speciesAgropyron repensBar bar CO vulgar isChrysanthcniiiiii leucanthemumDaucus carotaHieracium caespitosumPoa pratensisRumcx acetosdiaTrifolium hyhridumTrijbiium pratense

Native speeiesAmbrosia artemisiifotiaAster pitosusErigeron annuusFragaria virginianaPlanlago rugctiiPotciifilla simplex

0.441.310.440.251.140.310.140.380.69

0.500.292.750.140.091.00

7.96***15,85***22.04***

15.12***4 47***5.82***0.804,98**

19 02***16.93***12.24***

7^62***11.0!***

0.470.422.67*1.400.981.280,05-1.14

0.901.540.731.240.630.41

0.230.480.600.330.430.270.320.020.14

0.610.550.280.370.210.43

640 ECOGRAPIIV 24:6 (2001)

Fig, 4, Change inspecies richnessfollowing the decline ofexotic and native speciesinvasions. Data forinvaded plots only.Sampling times for eachspecies listed in Table 1,Change in eover ofinvading species classesare: L = light,M = moderate,H = heavy.

CDC

oCO

o<DQ.(0

0)cCD.n.O

-5

-5

-5

Agropyron repens

Daucus carota

Rumex acetosella

Ambrosia artemisiifolia

Fragaria virginiana

Barbarea vulgaris

H. caespitosum

Trifolium hybridum

Aster pilosus

Plantago rugeiii

C, leucanthemum

Poa pratensis

Trifolium repens

Erigeron annuus

Potentilla simplex

L M H L M H L M H

Change in cover of the invading species

1999). This may indicate that the diversity of exotic andnative speeies is controlled by similar mechanisms, suchas habitat heterogeneity (Lonsdale 1999), disturbance(Stohlgren et al. 1998), or soil characteristics (Harrison1999). Because ecological interactions occur at smallspatial scales within plant communities, the small-scale

negative association between diversity and exotic inva-sion is probably indicative of competitive interactions.In contrast, diversity at regional scales is less deter-mined by species interactions and more the result oflarge changes in abiotic factors (Rosenzweig andAbramsky 1993. Tilman and Pacala 1993).

ECOGRAPHY 24li (2001) 641

Overall, native and exotic species may not be asdissimilar as eomparisons to problematic exotics wouldindicate (Levine and D'Antonio 1999). Over the broadrange of speeies' life histories and successional agesinvestigated in this study, our general eonclusion is thatboth native and exotic species show the same lowincidence of effeets on speeies riehness.

Residual effects of iuvasions

We found no evidence for invasion effects that persistbeyond the decline of the species during sttcccssion.Speeies that change soil nutrient status or other soilchemical properties wouid be expected to have long-lasting effects on the community {Vitousek et al. 1987.Walker and Smith 1997). The only species in thisanalysis capable of nitrogen fixation were the legumesTrifolium hyhridum and T. pratense. While the strongdeclines in species richness associated with T. pratenseduring the invasion may result from the alteration oflocal nutrient levels, these effects do not persist beyondthe time span of the population in the eommunity.Allelopathy has been implicated for several herbaceousspecies in old lields, including Solidago canadensis(Jackson and Willemsen 1976, Fisher et al. 1978. Bosyand Reader 1995). If allelopathic chemicals are present,they appear to have short residence times in the soil orto have only weak effects on species richness. Clearly,exotic species with much stronger effects on soil chem-istry than those studied, may exhibit prolonged residualeffeets. Our analysis of post-invasion effects is conser-vative, as only two of the species that had signiflcanteffects on richness {Agropyron repens and Trifoliumrepens) could be analyzed for residual effects.

Exotic species removals generally result in an increasein the abundance and diversity of native species (Mc-Carthy 1997, Pickart et al. 1998). Such removal treat-ments may duplieate the consequences of a naturalpopulation decline or may prevent the development ofan alternative plant eommunity (Pike and Niering1999). Our results suggest that, in this system, invasivecontrol by species retnoval would be successful in allevi-ating any effects of the invasion. Alternatively, in caseswhere species removal is inappropriate or impractical,management practices that lead to the suecessionalreplacement of an invader should have similar benefi-cial effects on diversity. This type of management maybe more sustainable and cost-cfTicient. and should notineur any long-term effects on community structure.

Conclusions

While plant invasion certainly alters plant communitycomposition, lhe direet effects on patterns of diversityare largely unquantifled. From our survey, most inva-

sions v\ill have little to no effect on {he diversity ofplant communities. Although invasions by exotic spe-cies have stronger effects on community structure, na-tive and exotic invasions appear to have similar overalleffects. Long-term data are necessary to adequatelyevaluate the impacts of invasions without the problemsassociated with analyses across sites.

cniL'nls - We thank S, Picard for dalii managementand four decades of field workers for collection of vegetationdata. The Cooperative State Research, hdiication. and Exten-sion Service. U.S. Dept of Agriculture supported this research,under Agreement No. 99-35315-7695 to SJM and LTREBgrant DRB 97-26993 to STAP.

References

Amrine. J. W. and Stasny. T. A. 1993. Biocontrol of multiflorarose. - In; McNight. M. N, (ed.). Biological pollution: thecontrol and impact of invasive exotic species. IndianaAcad. of Sci.. Indianapolis, pp. 9-21.

Anderson. M. G. 1995. Interactions between Lylhrum salictniaand native organisms: a critical review. Environ. Man-age. 19: 225-231.

Anon. 1989. SAS.STAT User's guide, ver. 6. 4lh ed., Vol. Iand 2. - SASInst., Cary, NC.

Baker. H, G. 1965. Characteristics and modes of origin ofweeds. In: Baker. II. G. and Stebbins, G. L. {eds). Thegenetics oi colonizing species. Academic Press, pp. 147172.

Bard. G. H. 1952. Secondary succession on the piedmont ofNew Jersey. Ecol. Monogr. 22: 195 215.

Basil. M. et al. 1997. The effect of successional age anddisturbance on the establishment of alien plants in man-made sites: an experimental approach. - In: Brock. J. H. etal. {eds). Plant invasions: studies from North America andhurope. Backhuys Publishers. Leiden. The Netherlands,pp. 191 201.

Bazzaz. L. A. 1996. Plants in changing environments. Cam-bridge Univ. Press.

Bosy. J. L. and Reader. R. J. 1995. Mechanisms underlyingthe suppression of forb seedling emergence by grass {Poapraiensis) litter. Fund. Ecol. 9: 635-639.

Christian, J. M. and Wilson. S. D. 1999, Long-term ecosystemimpacts of an introduced grass in the northern great plains.

Ecology 80: 2397-2407.D'Aiitonio. C M, and Mahall. B. E. 1991. Root profiles and

competition between the invasive exotic perennial. Carpo-brotus eduli.s. and two native shrub speeies in Californiacoastal scrub. - Am. J. Bot. 78: 885-894.

Dillenburgh. L, R. et al. 1993. Effeets of vine competition onavailability of light, v\'ater. and nitrogen to a tree host{IJc/uidauihar slyracifluci). - Am. J. Bot. 80: 244-252,

Fike. J. and Niering. W. A. 1999. Lour decades of old fieldvegetation development and the role of Celastrus orhicula-tiis in the northeastern United States. - J. Veg. Sci. 10:483-492.

Fisher. R. !•.. Woods. R. A. and Glavicic. M. R. 1978.Allelopathic effects of goldenrod and aster on young sugarmaple. - Can. J. For. Res. 8: I 9.

Fogarty. G. and Facelli. J, M. 1999. Growth and competitionof Cviisus Morpariiis, an invasive shrub, and Australiannative shrubs. Plant Ecol. 144: 27-35.

Fox. M. D. and Fox. B, .1. 19S6. The susceptibility of naturalcommunities to invasion. - In: Groves. R. H. and Burdon.J. J. (eds). Ecology of biological invasions. CambridgeUniv. Press, pp. 57-66.

Gleason. H. A. and Cronquist. A. 1991. Manual of vascularplants of northeastern tJnited States and adjacent Canada.2nd ed. - New York Botanical Garden. Bronx,

642 liCOGRAPHY 24:(. (201)1)

Glenn-Lewin, D. C. and van der Maarel, 12. 1992. Patternsand processes of vegetation dynamics. In: Glenn-Lewin,D. C , Peet, R. K. and Veble'n, T. \. (eds). Plant succes-sion: theory and prediction. 1st ed. ('hapman and Hall, pp.11 -59.

Hager, H. A. and McCoy. K. D. 1998. The implications ofaccepting untested hypotheses: a review of the effects ofpurple loosestrife (Lythrum salicaria) in North America. --Biodiv. Conserv. 7: 1069-1079.

Hamilton, J. G., Holzapfel. C. and Mahall, B. F. 1999.Coexistence and interference between native perennialgrass and non-native annual grasses in C'alilbrnia. Oe-cologia 121: 518-526.

Harrison, S. 1999. Native and alien species diversity at theloeal and regional scales in a grazed California grassland.- Oecologia 121: 99-106.

Hobbs, R. J. 1989. The nature and effects of disturbancerelative to invasions. - In: Drake, J. A. et al. (eds),Biologieal invasions: a global perspective. Wiley, pp. 389-405.

Hutchinson, T. K. and Vankat, J. L. 1997. Invasibility andeffects of amur honeysuckle in southwestern Ohio forests.

Conserv. Biol, II: 1117-1124.Jackson, J. R. and Willemsen, R. W. 1976. Allelopathy in the

first stages of sccotidary succession on the piedmont ofNew .lersey. - Am. J, Bot. 63: 1015-1023.

Jackson, S. T., Futyma, R. P. and Wilcox, D. A. 1988. Apalcoeeologiea! test of a classical hydrosere in the LakeMichigan dunes. - Ecology 69: 928-936,

Knops, J. M. H. et al. 1999. Effects of plant species richnesson invasion dynamics, disease outbreaks, insect abun-dances and diversity. - Fcol. Lett. 2: 286-293.

Kotanen, P. M., Bergelson, J. and llazlett, D. L. 1998. Habi-tats of native and exotic plants in Colorado shortgrasssteppe: a comparative approach. - Can. J. Bot. 76: 664-672.

Levine, J. M. and D'Antonio, C. M. 1999. Elton revisited: areview of evidence linking diversity and invasibility. -Oikos K7: 15-26.

Lodge, D. M. 1993. Biological invasion: lessons for ecology. -Trends Ecol. Evol. 8: 133 137,

Lonsdale, W . M . I 999. Global patterns of plant invasions andthe concept of invasibility. - Ecology 80: 1522-1536.

Marks. P. L. 1983. On the origin or the field plants of thenortheastern United States. Am. Nat. 122: 210 228.

Marler, M. J., 7abinski, C. A. and Callaway, R. M. 1999.Mycorrhizae indirectly enhance eompetitive elTects of aninvasive forb on a native bunchgrass. Ecology 80: 1180-1186.

McCarthy. B. C. 1997. Response of a forest understory com-munity to experimental removal of an invasive nonindige-nous plant (Alliaria petlolata. Brassicaceae). - In: Luken,J. O. and Thieret, J. W. (eds). Assessment and managementof plant invasions. Springer, pp. 117-130.

Meiners, S. J. and Pickett, S. T. A. 1999. Changes in commu-nity and population responses across a forest-field gradi-ent. - Ecography 22: 261 267.

Milberg, P., Lamont, B. B. and Perez-Fernandez, M. A. 1999.Survival and growth of native and exotic composites inresponse to a nutrient yradlent. Plant Lcol. 145: 125-132.

Morgan, J. W. 1998. Patterns of invasion of an urban remnantof a species-rich grassland in southeastern Australia bynon-native plant species. - J. Veg. Sci. 9: 181 190.

Myster, R. W. and Piekett, S. T. A. 1990. Initial conditions,history, and successional pathways in ten contrasting oldfields. - Am. Midi. Nat. 124: 231-238.

Myster, R. W. and PickeU, S. T. A. 1992. Dynamics ofassociations between plants in ten old fields during 31 yearsof succession. - J. Fcol. 80: 291 302.

Palmer, J. H. and Sagar, G. R. 1963. Agropyron repens (L.)Beauv. (Tritieum repens L.; Elytrigia repens (L.) Nevski). -J. Ecol. 51: 783-794.

Parker. I. M. et al. 1999. Impact: toward a framework forunderstanding the ecological effects of invaders. - Biol.Inv. I: 3-19.

Peart, D. R. and Foin, T. C. 1985. Analysis and prediction ofpopulation and community change: a grassland case study.

In: White, J. (ed.). The population structure of vegeta-tion. Dr. W. Junk Publ.. Dordrecht, pp. 313 339.

Pickart, A. J., Miller, L. M. and Duebcndorfer, T. E. 1998.Yellow bush lupine invasion in northern California coastaldunes 1. Ecological impacts and manual restoration tech-niques. - Rest. Ecol. 6: 59-68.

Pickett, S. T. A. 1982. Population patterns through twentyyears of oldfield succession. - Vegetatio 49: 45-59.

Piekett, S. T. A. 1989. Spaee-for-time substitutions as analternative to long-term studies. In: Likens, G. E. (ed.).Long-term studies in eeology. Springer, pp. 110 135.

Pimentel, D. et al. 2000. Environmental and economie costs ofnonindigenous species in the United States. BioSeience50: 53 65.

Rejmanek, M. 1989. Invasibility of plant communities. - In:Drake, J. A. et al. (eds). Biological invasions: a globalperspeetive. Wiley, pp. 369-388.

Rejmanek. M. 1999. Invasive plant species and invasiblee^cosystems. - In: Sandlund, O. Y., Schei, P. J. and Viken,A. (eds). Invasive species and biodiversity management.Kluwer, pp. 79 102,

Richardson, D. M,, Macdonald, I. A. W. and Lorsyth, Ci. G.1989. Reductions in plant species richness under .stands ofalien trees and shrubs in the Eynbos biome. - S. Afr. For.J. 149: 1 8.

Rosenzweig. M. L. and Abramsky, Z. 1993. How are diversityand productivity related? - In: Rieklefs, R. E, andSehluter, D. (eds). Species diversity in ecological communi-ties. Univ. of Chicago Press, pp. 52-65.

Schierenbeck, K. A., Mack, R. N, and Sharitz. R. R. 1994.Fffects of herbivory on growth and biomass allocation innative and introduced species of l.oniecra. - Ecoloey 75:1661-1672.

Schweitzer, J. A. and Larson, K. C. 1999. Greater morpholog-ieal plasticity of exotic honeysuckle species may make thembetter invaders than native species. J. Torrey Bot, Soc.126: 15-23.

Small. J. A.. Buell, M. F. and Siccama, T. G. 1971. Old-fieldsuccession on the New Jersey Piedmont-the first year. -Wm. L. Hutcheson Mem. For. Bull, 2: 26-30.

Smith, M. D, and Knapp, A. K. 1999. Fxotic plant speeies ina C4-dominated grassland: invasibility, disturbance, andcommunity structure. - Oecologia 120: 605-612.

Stapanian, M. A. et al, 1998. Alien plant species compositionand associations with anthropogenic disturhance in NorthAmerica, - Plant Eeol. 139: 49-62,

Stohlgren, T. J. et al. 1998. Riparian zones as havens forexotie plant species in the central grasslands. - Plant Ecol.138: 113 125.

Stohlgren, T. J. ct al. 1999. Exotic plant species invadehot spots of native plant diversity. - Ecol. Monogr. 69:25-46.

Thomas. 1-. K, 1980. The impact of three exotic plant speeieson a Potomac island. Nat. Park Serv. Sei. Monogr. Ser..no. 13. U.S Dcpt Interior.

Tilman, D. 1997. Community invasibility, recruitment limita-tion, and grassland biodiversity. Ecology 78: 81-92.

Tilman, D. and Piicala, S. 1993. The maintenance of speciesrichness in plant communities. In: Rieklefs, R. E. andSchiuter, D. (eds|, Speeies diversity in eeologieal communi-ties, Univ. of Chicago Press, pp. 13 25,

Treberg, M. A, and Husband, B. C. 1999. Relationship be-tween the abundance of Lythrum salicaria (purple looses-trife) and plant speeies richness along the Bar River,Canada. Wetlands 19: I IK 125.

ECOGRAPHY 24:() (20U1) 643

Vitousek, P, M, et al, 1987, Biological invasion by .W>'ffra/ij_ra Wiser, S. K, et al, 1998, Community structure and forestalters ecosystem development in Hawaii, - Science 238: invasion f'>y an exotic herb over 2.1 years. - Ecology 79:802-804, 2071 2081,

Walker, L, R, and Smith. S, D, 1997, Impacts of invasive Woods. K, D, 1993, Effects of invasion by Lonicera larlaricaplants on community and ecosystem properties, - In: L, on herbs and tree seetllings in four New England forests,Luken, J. O. and Thieret, J, w', (eds). Assessment and - Am. Midi, Nat, 130: 62 74,management i^( plant invasions. Springer, pp, 69 86. Woods, K, D, 1997, Community response to plant invasion, -

Weber, E, 1997, Morphological variation of the introduced In: Luken, J. O. and Thieret, ,1, W, (eds). Assessment andperennial Solidago eaiuidensis L, sensu lato (Asteraceae) in management of plant invasions. Springer, pp. 56-68,Europe, - Boi, J, Linn, Soc. 123: 197 210, Wyckoff. P, H, and Webb, S, L. 1996, Understory influence of

Williamson, M. and Fitter. A, 1996, The varying success of the invasive Norway maple (Aeer plaianoides). - Bull,invaders. Ecology 77: 1661-1666, Torrcy Bot, Club 123: 197-20.5,

i'.COGRAPHY 24:0 (2()