Non-Native Invasive Earthworms as Agents of Change in ...

Non-Native Invasive Earthworms as Agents of Change in Northern Temperate Forests
Author(s): Patrick J. Bohlen, Stefan Scheu, Cindy M. Hale, Mary Ann McLean, Sonja Migge,
Peter M. Groffman, Dennis Parkinson
Source: Frontiers in Ecology and the Environment, Vol. 2, No. 8 (Oct., 2004), pp. 427-435
Published by: Ecological Society of America
Stable URL: http://www.jstor.org/stable/3868431
Accessed: 19/02/2010 09:54
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=esa.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

Ecological Society of America is collaborating with JSTOR to digitize, preserve and extend access to Frontiers
in Ecology and the Environment.

http://www.jstor.org

REVIEWS REVIEWS REVIEWS

Non-native invasive earthworms as agents I

of change in northern temperate forests

Patrick J Bohlen1, Stefan Scheu2, Cindy M Hale3, Mary Ann McLean4, Sonja Migge5, Peter M Groffman6, and
Dennis Parkinson7

Exotic earthworms from Europe and Asia are invading many northern forests in North America that cur-
rently lack native earthworms, providing an opportunity to assess the role of this important group of
invertebrates in forest ecosystems. Research on earthworm invasions has focused on changes in soil struc-
ture and carbon (C) and nitrogen (N) cycling that occur following invasion. These changes include the
mixing of organic and mineral soil horizons, decreases in soil C storage, and equivocal effects on N cycling.
Less well studied are changes in the soil foodwebs that accompany earthworm invasion. Soils of north tem-
perate forests harbor a tremendous diversity of microorganisms and invertebrates, whose distribution and
abundance can be substantially altered by earthworm invasion. Furthermore, invasive earthworms can
affect understory plant communities, raising concerns over the loss of rare native herbs in some areas. The
ecological consequences of earthworm invasion are mediated through physical, geochemical, and biologi-
cal effects. These effects vary with different earthworm species, as well as with the characteristics of the
site being invaded. Earthworm invasions may have important interactions with other rapid changes pre-
dicted for northern forests in the coming decades, including climate and land-use change, increased nutri-
ent deposition, and other biological invasions.

I

Front Ecol Environ2004; 2(8): 427-435

Northern forestsare experiencing multiple stressesand (Vitousek 1990). Understandably,much of the focus on

changes, including nutrient deposition and acidifica- exotic species invasions in forestsand other ecosystemshas

tion, diseaseand pest outbreaks,changing climate, and bio- been on abovegroundinvaders,which are the most appar-

logical invasions (Aber et al. 2001). Biological invasions in ent. However,belowgroundinvasionsmaybe equallywide-

northern forests threaten to alter ecosystem structureand spreadand may become better known as more ecologists

function, especiallywhen they change the habitat of other begin to recognizethe importanceof links between above-

species, alter the availabilityor transformationratesof key ground and belowground communities (Scheu 2001;

resources, or compete with or replace native species Wardle2002). Invasionof northernforestsby exotic earth-

worms,for example, is receiving increasingattention.

In a nutshell: Earthwormsare the best known of the large soil fauna,

* Manynortherntemperateforestsin the US have had no nat- but many northern forests in North America lacked

uralearthwormpopulationssince the lastglacialperiod,due to earthworm populations prior to European settlement,
slownorthwardexpansionof native species probably because of slow northward expansion of native
* Exotic earthwormspeciesfromEuropeand Asia are invading earthworm populations following the last glacial period
manyof these forests,substantiallyalteringforestsoils (James 1995). Foreign earthworm species, mainly of
* These invasionscan alternutrientstorageand availabilitya, nd European and Asian origin, are currently invading these
greatlyaffectpopulationsand communitiesof other floraand

faunathat inhabitforestsoils forests over a wide geographic area (eg Alban and Berry
1994; Scheu and Parkinson 1994; Bohlen et al. 2004a).
* Invasions by earthwormsare likely to increase in northern
forestswith increasedhuman activity and changing climate; The earthworm species involved belong to a group that
the consequences should be considered along with other have been termed the "peregrine"species, representing a
importantchangestakingplace in these ecosystems small portion of overall earthworm diversity, and are

characterized by their ability to colonize new habitats,

'ArchboldBiologicalStation, Lake Placid, FL (pbohlen@ spread rapidly, and tolerate a wide range of ecosystems

archbold-station.o2rDg)e;partmeonftBiologyT, echnischUeniversitdt and environmental conditions (James and Hendrix

DarmstadDt, armstadGt, ermany3;TheNaturalResourceRsesearch 2004). The introduction of these species into new habi-

Institute,Universityof MinnesotaD, uluth,MN; 4IndianaState tats is facilitated by human activity, such as the construc-

UniversityT, erreHaute,IN; 5DepartmeonftBiologyU, niversitoyf tion of logging roads,direct releases of unused fishing bait

GottingenG, ottingenG, ermany;6Insttuteof EcosystemStudies, by anglers,and the relocation of fill or horticulturalmate-

MillbrookNY; 7Departmenotf Biology,Universityof Calgary, rials (Hendrix and Bohlen 2002).

CalgaryC, anada. Earthworm invasions of northern forests can lead to

?C The EEccoollooggiiccaallSSoocciieettyy of America wwwwww..ffrroonnttiieerrssiinneeccoollooggyy..oorrgg

Invasiveearthwormsin temperateforests PJBohilen et al.

I rapid and dramatic changes in the soil environment accomplished through the redistributionand transforma-
(Langmaid 1964; Alban and Berry 1994) due to the tion of soil organic matter as earthworms consume

important role of earthwormsin modifying soil structure, organic-rich forest floor material and incorporate it into

redistributingorganic matter, and altering the habitat of underlying mineral soil (Figure 1). The degree of mixing

other organisms living in or on the soil (Neilson and of soil layersdepends upon the life history traits of partic-

Hole 1964). This key role of earthwormsin forest humus ular earthworm species, which are often broadly catego-

formation has long been recognized and is capturedin the rizedas belonging to one of three main ecological groups:

historical mull, moder, and mor terminology used to epigeic, endogeic, and anecic (Edwardsand Bohlen 1996).

describe forest soils (Parkinson et al. 2004). Epigeic species reside mainly in the upper organic layer

Several excellent recent reviews are available on the and may cause limited mixing of mineral and organic soil

ecology of earthworms, their critical role in soil genesis, layers.Endogeic species reside in the mineralor mixed soil

their effects on nutrient cycling and organic matter layers and often enhance mixing of organic and mineral

breakdown, and their interactions with soil microbial soil layers. Anecic species (including Lumbricusterrestris,

communities (Lavelle et al. 1999; Edwards 2004). The the common nightcrawler) form nearly vertical perma-

goal of this paper is not to review this extensive litera- nent burrowsup to 1-2 m deep. These species incorporate

ture, but rather to point out that earthworm invasions in litter into the soil and bring mineral soil from different

northern forests provide an opportunity to examine the depths to the surface,resulting in soil mixing that is very

generality of current concepts in earthworm ecology and different from the mixing caused by epigeic or endogeic

to compare and contrast the importance of differentfunc- species (Figures 1 and 2). Not all species fall neatly into

tional groups of organisms in invasion biology. Because these standard categories, but classification is useful for

earthworms modify the surface soil, which is where the differentiating major ecological groups and their effects.

primary interactions between aboveground and below- The physical transformationof the soil is the most obvious

ground communities are mediated, these invasions have sign of earthworminvasion in northern forests.How these

important consequences for the soil foodweb, nutrient physical changes and the associated redistribution of

cycling, and plant communities. The accompanying dis- organic matter influence ecosystem level processessuch as

turbances may be an important aspect of the current and total C storage,N transformationrates, and loss of nutri-

future environmental change in these forests. ents via hydrologic and gaseous pathways, and how these

change over time, is not well understood.

* Earthworminvasions and nutrientcycling A shift towards a faster cycling system implies that
earthworm invasion could result in a net loss of C from

Much of the researchon earthworminvasions has focused the soil. Northern forestsare importantglobal C sinks, but

on the consequences for nutrient cycling and soil micro- environmental changes may turn them into C sources

bial processes(eg Scheu and Parkinson 1994; Bohlen et al. (McKane et al. 1997; Hobbie et al. 2002). Earthworm

2004b; Groffman et al. 2004). Earthwormsshift the soil invasion could enhance increased C flux from northern

system from a slower cycling, fungal-dominated system to forests due to global warming, especially as the ranges of

a faster cycling, bacterial-dominated system, or at least different ecological groups of earthworms expand into

one that is less fungus dominated (Wardle 2002). This is more of these forests. In fact, most studies have docu-

Figure 1. Effectsof invasiveearthwormosn forestsoilsin northernMinnesota.(a) Understoryvegetationandherblayerin areas
withoutearthwormsN. ote theabundanceof treeseedlingsandnativeherbsandtheundisturbesdoillayerwithcoverof leaflitter.(b)
Understoryvegetationandsoilsurfaceat a siteinvadedbyearthwormsN. ote thelackof herblayer,exposedsoilsurfacewithreduced
litterlayerandabundantearthwormcasts,andexposedrootsof canopytreesduetodisappearancoefforestfloor.

wwwwww..ffrroonnttiieerrssiinneeccoollooggyy..oorrgg ?? TThhee EcologicalSociety ooff AAmmeerriiccaa
EcologicalSociety

P1JBohlen et al. Invasiveearthwormsin temperateforests

mented an increase in soil C loss following earthworm Figure 2. Earthwormsrepresentativeof differentecological
invasion. Carbon loss of around 600 kg per hectare per groups. (a) Epigeicspecies, such as Dendrodrilusrubidus,
year for a period of 14 yearswas reportedfor mixed hard- inhabitorganic-richsurfacelayersandfeed mainlyon surface
wood forestsin Minnesota (Alban and Berry1994), and of organic matter. (b) Endogeicspecies, such as Octolasion
28% of total surface soil C for sugarmaple (Acer saccha- tyrtaeum,consumemoremineralsoil thanepigeicspecies,and
rum)forestsin the northeasternUS (Bohlen et al. 2004b). mixmineralandorganicsoillayerstogether(.c) Anecicspecies,
Such losses occur because recalcitrant C pools that accu- suchas Lumbricusterrestris,livein deepverticalburrowsf,eed
mulate at the soil surfacein the absence of earthwormsare mainlyon surfacelitter,andincorporatleitterintothesoilas well
as transportinmg ineralsoilto thesurfacefromdeepersoillayers.
exposed to greater mineralization rates after earthworms (Note thatonly the anteriorend of L terrestrisis shownhere;
invade. The increased mineralization is partly a result of theposteriorendremainsin itsburrow).
earthworm respiration, but is mostly due to the stimula-
tion of microbial activity in earthworm guts and casts.
Mineralization is also increased by the secretion of labile
C compounds in mucus and subsequent alteration of soil
structure (Edwardsand Bohlen 1996). In the long term,
however, earthworm activity may contribute to stabiliza-
tion of soil C in earthwormcasts and other stable aggre-
gates formed by earthworm activity (Scheu and Wolters
1991; Tiunov and Scheu 2000).

The effects of earthworm invasion on N cycling and
retention are more complex and less conclusive than the
effects on C transformations.Nitrogen is both a limiting
nutrient and an atmospheric pollutant, so the effects of
earthworms on N cycling potentially influence nutrient
uptake and the fate of atmospheric N deposited in forest
ecosystems. Earthwormshave been shown to increase the
N mineralization and leaching of N from forest soils in
both lab and microcosm studies (Haimi and Huhta 1990;
Scheu and Parkinson 1994; Burtelowet al. 1998), indicat-
ing that earthworminvasions have the potential to accel-
erate rates of N transformation. Earthwormactivity has
been shown to increase denitrification rates in some

instances, though not in others, and the question of
whether earthworminvasion influences overall gaseousN
flux from forest soils remains unanswered (Burtelow et al.

1998; Bohlen et al. 2004b). It is also unclear whether
earthworm invasion substantially alters N retention in
those systems. They did not lead to large increases in N
leaching from surface soil in undisturbed forest plots in
New York(Bohlen et al. 2004b).

Analysis of soil C and N content and stable isotope sig-
natures of invaded forest sites in New Yorksuggested that
total soil N remained unchanged following invasion,
despite the fact that soil C storage decreasedsubstantially
(Bohlen et al. 2004b). An increase in total microbial bio-

mass, which may act as a strong immobilization sink for
available N in C-rich soils (Groffman et al. 2004), is one
possible mechanism for N retention following earthworm
invasion. This is an interesting finding because some
studies have reported a decrease in microbial biomass in
response to earthwormactivity (Blair et al. 1997; Hendrix
et al. 1998; Wolters and Joergensen 1992). However, most
previous studies that reporteda decrease in microbial bio-
mass with earthworms involved agricultural or mixed
soils, not organic-rich forest soils where earthworms can

stimulate microbial biomassby mixing soil layerstogether
(Scheu and Parkinson 1994; Saetre 1998). In the New

?C ThheeEEccoollooggicicaalSSloocciieettyyoffAmmeerricicaa wwwwww..ffrroonnttiieerrssiinneeccoollooggyy..oorrgg

Invasiveearthwormsin tteemmppeerraatteeforests PJBohilen etal.

I Yorkstudy,the increasein microbialbiomasswas attributed Dendrobaenaoctaedrainto pine forest soil in Alberta sug-
to a morefavorableenvironment that increasedthe "carry- gest that in the long term, fungal community diversity
ing capacity"of the mixed surface soil relative to undis- and richness decreased and dominance among fungi

turbedforestfloor.Understandingor predictingthe effects increased, apparently due to the disruption of fungal

of earthworminvasion on N cycling in forestsoils depends hyphae, decreased resource availability, and reduced spa-

on the quantity and quality of soil organic matter,because tial heterogeneity as the organic layers became homoge-

of the tight coupling of soil C and N cycling. Earthworms nized (McLean and Parkinson 2000a). Competition

may increaseN flux more in systemsfertilizedwith N, such among fungi was reduced soon after invasion, possibly

as agricultural or pasture systems (Knight et al. 1992; through the addition of nutrients or as a result of distur-

Dominguezetal. in press);earthworminvasionmight there- bances due to burrowingand deposition of casts into the

fore be expected to increase N flux more in forests with organic layers (McLean and Parkinson 1998b).

high levels of N deposition than in moreN-limited systems, Given the differences between epigeic, endogeic, and

though this remainsto be investigated. anecic earthwormsin termsof feeding behaviorand effects

Earthworm invasion also affects phosphorus (P) on the soil profile, one might expect the response of soil

cycling in soil, which is strongly influenced by physical fungi to earthworm invasion to vary in accordance with

and chemical modifications. Trees in areaswith an intact the ecological groupof the invading species. In fact, results

forest floor concentrate a large proportion of their fine from a laboratorymesocosm experiment suggestthat some

roots in this layer, where up to 80% of their annual P species, including fast-growingspecies of Trichodermaa,re

requirement is met by the tight linking of organic P min- favored by the presence of the epigeic D octaedra,but are

eralization and uptake (Wood et al. 1984; Yanai 1992). By inhibited by the presence of anecic and/orendogeic earth-

eliminating the forest floor and mixing it with the under- worms, indicating that epigeic earthworms favor fungal

lying soil, earthworms can greatly alter P cycling by species that tolerate moderate disturbances.Other fungi,

increasing P fixation by soil minerals and by altering the including Mortierella sp (Zygomycetes), uniformly

mineralization of organic P. For example, a study of sugar decreased in the presence of epigeic, anecic, or endogeic

maple stands in Quebec showed that all stands that had earthworms, reflecting their inability to tolerate hyphal

forest floors mixed with mineral soil had lower concentra- disruption;this is due to the lack of septa to prevent cell

tions of available P than did stands with intact forest content leakage. With so little data available, conclusions

floors, undisturbed by earthworm activity (Pare and about the effects of earthwormson microfungalcommuni-

Bemier 1989 a, b). A comparison of plots with and with- ties must remainpreliminary.

out earthworms in New York suggested that invasive It is probable that earthworm invasions not only affect

earthworms influenced soil P cycling, but that the effects saprophytic but also mycorrhizalfungi, but this has not

depended on the species composition of the earthworm been well studied. Results from a laboratorystudy on the

community, possibly because various earthworm species impacts of earthworms on mycorrhizalfungi are congru-

differentially affect the degree of mixing of soil layersand ent with observations on saprophytic microfungi. The

the redistribution of organic and mineral components presence of earthworms of different ecological groups

within the soil profile (Suarez et al. 2004). The effect of decreasedcolonization rates and abundance of arbuscular

earthwormson P cycling is largelydependent on regional mycorrhizaeof sugar maple, most likely due to physical

soil mineralogy, the degree of mixing of soil layers, and disruption of fungal mycelia (Lawrence et al. 2003).

the timeframe of the invasion. An initial increase in Changes in the abundance and functioning of arbuscular

organic P mineralization in the early stages of invasion mycorrhizalfungi probably contributed to alterations in

may be followed by a decrease in available P as soil min- forest herb communities.

erals fix more P.

* Alterationof the soil foodweb Soil invertebratecommunityresponses

Soil microbialcommunityresponses Earthworm invasions can rapidly alter soil structure,
humus forms, and plant communities, and some of these
Organic layers in the soil provide microhabitats and changes can occur in as short a period as 2 to 5 years
resourcesthat support an abundant and diverse soil com- (Langmaid 1964). However, the influence of these
munity. The profound changes in the soil organic layers ecosystem engineers on other soil biota, eg microarthro-
resulting from earthworm invasion greatly alter micro- pods (mites, collembolans), enchytraeids (potworms), or
habitats and resources for microorganisms and inverte- nematodes is poorly understood, especially with repect to
brates (Figure3). However, with the exception of studies changes that occur following earthworminvasions in for-
in forests in Alberta, Canada, and in a maple forest in est ecosystems. Soil-inhabiting vertebrates and their
New York,data on the effects of earthworm invasions on responses to invading earthworms as new food resources
soil microbial communities in forests are limited. have only recently been examined in more detail (Maerz
et al. in press), although the importance of earthwormsas
Studies on the response of soil fungi to the invasion of a food source for a wide variety of vertebrates is well

wwwwww.f.frroonntiteiresrinsienceoclooglyo.goyrg.org ?? TThhee EEccoollooggiiccaallSSoocciieettyy ooff AAmmeerriiccaa

PJBohlen et alt. Invasiveearthwormsin temperateforests

I

Figure 3. The diversityof soil mixingeffectsof differenteco-
logicalgroupsof earthwormsin boxescontainingreconstructed
soil layersfroman aspenforestin southwestAlberta,Canada.
(a) Boxeswithno earthworms(;b) boxescontainingindividuals
of theepigeicearthwormspecies,D octaedra,whichhavemixed
theorganiclayerinto themineralsoil;and (c) boxescontaining
individualsof the endogeicspecies, 0 tyrtaeum, whichhave
thoroughlymixedthe mineraland organiclayerstogether.For
moredetailsseeScheu& Parkinson(1994).

established (Edwardsand Bohlen 1996). mineralsoil in much greaterquantitiesthan D octaedra,and
therefore compete vigorously with microarthropodsfor
The rapidly spreading earthworm D octaedra mainly microbial and organic food. They alter habitat and food
inhabits the forest floor, which is also the major habitat resourcesmorestronglyand createmechanical disturbances
for soil-inhabiting microarthropods.Studies in a pine for- which, as demonstratedin a 1-yearlaboratoryexperiment,
est in Alberta confirmed the expected, positive influence may result in lower microarthropodabundances than are
of earthwormactivity on soil invertebrates(Wickenbrock found in agricultural soils under conventional tillage
and Heisler 1997; Lorangeret al. 1998), but only either in (Migge 2001). The same processesoccur in the field; these
the short term or in unfavorablehabitats. A high biomass will be slower in dry continental climates such as in the
of D octaedraresulted in increased oribatid mite diversity; forestsof southernAlberta, but are likely to be much faster
in a laboratoryexperiment, intermediate levels of earth- in the milderclimates of the easternforests.To date, there
worm activity tended to increase microarthropod abun- are no comparabledata available from other regions, and
dance after 3 months, probably due to an increase in data on enchytreaids,nematodes, and macroarthropodsare
microhabitats and microbial food resources (McLean and
lacking.
Parkinson 1998a). Similarly, the dry and unfavourable
pine litter (L-layer) was modified by earthworm activity Soil inhabiting vertebrates
in a way that resulted in increased oribatid mite species
richness and diversity (McLean and Parkinson 2000b). Recent studies on the foraging behavior of salamanders
However, high earthworm biomass and activity in the (Plethodonsp) suggest that introduced earthworms may
longer term and in the lower organic layers, the main play an important role in salamanderdiets, especially in
habitats of microarthropodsin climates with hot and dry lowland forests and during rainy seasons (Maerz et al. in
summersand cold winters, resulted in strong decreases in press). Earthwormsapparently increase the fecundity of
diversity and abundance. These declines can be attrib- adult salamanders by providing a high protein food
uted to the restructuringof the organic layer into earth- source, but decrease the survival rates of young salaman-
worm casts and associated mechanical disturbance, as ders, presumably by reducing populations of smaller

well as the competition for microbial food resources.
More dramatic decreases in microarthropod diversity

and abundance were observed in aspen forest soil in
Alberta when L terrestris, an anecic species, and
0 tyrtaeumand Aporrectodeacaliginosa, both endogeic
species, invaded soils previously devoid of earthworms.
These large earthworm species mix organic materialand

? The EEccoollooggiiccaallSSoocciieettyy of America www.frontiersinecology.org
www.frontiersinecology.org

Invasiveearthwormsin tteemmppeerraatteeforests PJBohilen etal.

I invertebrate food sources, or by altering the habitat in diverse community of herbs was present even in some
other ways. Further investigations are needed to under- areasthat supporteda high biomass of earthwormspecies
stand how this influences salamanderpopulation dynam- other than L rubellus.However, where the L rubellusbio-
ics and what impacts the invading earthwormswill have massreached its maximum, the herbaceousplant commu-
on other predators,such as birds or small mammals. The nity was dominated by Carex pennsylvanicaand Arisaema
changes observed in decomposer and plant communities, triphyllum,with rare occurrences of other native plant
as well as vertebrateforagingstrategies, imply that below- species. The disappearanceof established populations of

ground invasions can also dramatically alter the above- the raregoblin fern (Botrychiummormo)were also associ-

ground food web. ated with the invasion of this earthworm(Gundale 2002).

Earthworminvasion of hardwood forests can have both

* Plant community responses to earthworm direct and indirect effects on the understoryplant com-
invasions munity. Direct mortality of herbaceous plants and small
tree seedlings rooted in the forest floor occurs when

The abundance and diversity of native plant species and earthworms eat the forest floor out from under them

tree seedlings can decline steeply following earthworm (James and Cunningham 1989; Hale 2004). Earthworms

invasion (Figure 1). Some hardwood forest stands in may reduce plant regeneration by ingesting and burying

northern Minnesota that had lush and diverse understory seeds and by reducing the shelter offered by the forest

plant communities only 40 years ago now have only one floor layer, thus exposing seeds and seedlings to desicca-
species of native herb and virtually no tree seedlings tion and predation by insects, small mammals, and other
remaining after earthworm invasion, although other fac- organisms(Leck 1989; Kostel-Hughes 1995). After initial

tors may also have contributed to these changes (Hale earthworm invasion, smaller understory plant popula-

2004). Concerns have therefore been raised about the tions become more vulnerable to the impacts of deer graz-

potential for widespreadloss of native forest plant species ing, which can lead to local extirpation (Augustine et al.

and the stability of their communities in hardwood forest 1998). Changes in soil nutrient dynamics, including

ecosystems following earthworm invasion. increases in N and P loss due to leaching and decreased

Earthworminvasions in sugarmaple-dominated forests availability, may also affect the growth and composition

in northern Minnesota were first noted 15 to 20 yearsago of herbaceous plant communities following earthworm

(Mortensen and Mortensen 1998). Many stands con- invasion (Bohlen et al. 2004b; Hale 2004).

tained discrete, visible leading edges of invasion where Individual responses of plant species to earthworm

forest floor thickness decreased from 10 cm to zero in dis- invasion will be important in determining the trajectory

tances as short as 75 meters. Associated with the loss of of compositional changes in hardwood forests following

the forest floor across the leading edges were rapid earthworm invasion (Scheu 2003; Hale 2004). In a

increases in earthworm biomass, the successive appear- greenhouse experiment, the addition of earthworms

ance of up to eight earthworm species, and substantial increasedplant mortality and altered root-to-shoot ratios,

changes in soil physical and chemical properties (Hale et but the magnitude and direction of those changes

al. in press). When not associated with some underlying depended on both the plant species and the particular

gradient in initial soil conditions, this leading edge of species of introduced earthworm. In field studies,

earthworm invasion provides an opportunity to assessthe Arisaematriphyllum(Jack in the pulpit) and Allium tricoc-

relationships of earthworm biomass and species assem- cum (wild leeks) were both positively associated with

blages to changes in the understory plant community, increasing earthworm biomass (Hale 2004). Changes in

while controlling for site-specific factors that often the soil fungal community following earthworm invasion

thwart field-based comparative studies. (Johnson et al. 1992; McLean and Parkinson 2000a;

In study plots in northern Minnesota, the abundance Lawrence et al. 2003), may have consequences for native

and diversity of the herbaceous plant species and tree understoryplants, the vast majorityof which are strongly

seedlings across leading edges of earthworm invasion mycorrhizalin northern forests (Brundrett and Kendrick
decreased with increasing total earthworm biomass, but 1988; Baskin and Baskin 1998). The dominance of earth-

the magnitude of change depended on the species assem- worm-invaded sites by non-mycorrhizal herbs such as

blage of the earthworm community (Hale 2004). C pennsylvanicain Minnesota and Allia petioloata(garlic

Herbaceous plant communities were more strongly mustard) throughout the eastern US suggests that earth-

affected when the epi-endogeic species Lumbricusrubellus wormsmay enhance the competitive success of non-myc-

was present. A diverse community of herbaceous plants, orrhizalherb species at the expense of mycorrhizalones

including species that are often used as indicators of rich (Francisand Read 1994; Brussaard1999; Hale 2004).

sugar maple-dominated hardwood forests in the Great In addition to effects on understory herbs, earthworm

Lakes region (such as CaulophyllumthalictroidesU, vularia invasion may influence overstory trees by influencing

grandiflora, Trillium spp, Osmorhiza claytonii, Asarum root distribution and function (Fisk et al. 2004). Invading

canadensis,and Polygonatumpubescens;Kotar2002), could earthworms can lower the soil surface by removing the

be found in areas where L rubelluswas not present. This forest floor, thus leaving tree roots exposed and changing

wwwwww..ffrroonnttiieerrssiinneeccoollooggyy..oorrgg ?? TThhee EEccoollooggiiccaallSSoocciieettyy ooff AAmmeerriiccaa

PPJ1Bohlen et al. Invasiveearthwormsin tteemmppeerraatteeforests

their distribution in the soil Dispersal Resource quantity Resource quality Soil factors I
* source population * plantprocductivity *vegetationtype * moisturehydrology
(Figure lb). Forest sites in New * naturael xpansion *soil orgaanicmatter * litterC:Nratio *texture,sand/silt/clay
York overrun by earthworms had *humanactivity *tannins,polyphenotics *acidity,base cations
fewer fine roots than sites without *streams,rivers `11

earthworms, while the higher N b` b
concentration in roots in invaded
a
sites indicated that earthworms
Physicaleffects Geochemicaleffects - Biologicaleffects
altered allocation and possibly burrowingandcasting ?change insoil habitat
also functioning of fine roots and *mixingsoil layers *fasternutrienctycling
the efficiency of N uptake. The ? litterremoval *adsorption/desorption<
long-term consequences of such ?soil aggregates * mineralweathering less fungally-dominated
changes for tree nutrition, ?fewermycorrhizae
seedling establishment, and seed porosity *change inminerology ?change inrootingzone
survival are unknown. ?alteredseedbed
hydrology
? erosion

* Conclusions

Earthworm invasion of northern Ecosystem properties < Ecologicalcommunities
*C loss (shortterm) *plantinvasions
forests is influenced by many fac- *C stabilization (long term) . * loss of native herbs
tors and has multiple ecological *N retention? ?soil invertebratceommunityshifts
effects mediated through the * P availability? * microbiacl ommunityshifts
physical, geochemical, and bio- *Treenutrition?
logical changes that occur follow-
ing invasion (Figure 4). The eco- Figure 4. This conceptualmodel illustrateshow the influenceof earthwormson the
logical consequences of these biologicalc, hemical,andphysicalcharacteristicosf thesoilecosysteminteractto determine
invasions depend upon which thenet influenceof earthworminvasionon ecosystemprocessesandecologicacl ommunities
(C = carbon,N = nitrogen,andP = phosphorus)S. ee textforexplanation(.Modifiedfrom

species or species complexes Bohlenet al. 2004a).
invade and the characteristics of

the site being invaded. Enough evidence exists to gener- sent an inadequate food source for earthwormswith rapid
alize about some of these consequences, such as carbon growth rates. The quality of soil organic matter or plant
loss and gross changes in microbial communities. Yet, litter, including such properties as the C:N ratio, or tan-
despite the vast literature on earthworm ecology, too lit- nin (polyphenol) content, is known to influence earth-
tle evidence is available to make general predictions worm feeding, growth, and fecundity. Forests with large
about their effects on nutrient availability or loss, soil amounts of soil organic matter but poor food quality may
erosion, and native microbial, plant, or invertebratecom- prevent establishment or limit growth and expansion of
munities. Examining the response of the soil ecosystem to earthworm populations (Edwards and Bohlen 1996).
earthworm invasion may help clarify current concepts in Finally, soil factors, such as texture, acidity, richness of
earthworm ecology. It may even lead to new concepts base cations, and moisture, influence the establishment
that will supplant long-held views about the role of earth-
worms in forest ecosystems. and size of earthwormpopulations. These various factors
may affect different species in dissimilar ways, so that
Most research on earthworm invasions has focused on conditions that prevent colonization by one species may
not prevent the establishment of another.
the consequences of invasion at particular forest sites,
with much less emphasis on the mechanisms of invasion The potential widespread loss of native understory
and the broaderlandscape- and regional-scale factorsthat plant species that could resultfrom expanding earthworm
contribute to, or limit, invasion. These factors include invasion of northern hardwood forests raisesthe question
of whether strategies could be developed to prevent fur-
agents of dispersal, the quantity and quality of resources ther invasions or provide for restoration following inva-
at invaded sites, the nature of the overstory vegetation, sion (Proulx 2003). Although local control of invasions
and soil characteristics (Figure 4). Human activity has may be possible in some situations, the magnitude and
been implicated as an important cause of earthworm dis- regional scale of invasion by non-native earthwormssug-
persal, including anglers dumping unused fishing bait, gest that in the next few decades many northern temper-
logging truckscarryingmud that contains earthwormsor ate forests that currently lack earthwormswill be invaded
cocoons into newly logged areas, and the spreading of to some degree, rendering regional control strategies
horticultural materials such as mulch or compost in gar- impractical.Reducing agents of dispersalby, for example,
dens or along trails. Factorsthat affect either the quantity educating fishermen about the effects of earthworm
or quality of soil organic matter are important because introductions, limiting the spreadof horticultural materi-
organic matter is the primaryresource base for invading als containing earthworms, or limiting new road con-
earthworms.Low quantities of organic matter may repre-

? The EEccoollooggiiccaallSSoocciieettyy of America www.froneiersinecology.org
www.frontiersinecology.org

Innvvaassiviveeearrtthhwwoorrmms sin tteemmppeerraateteforreessttss PPJJBBoohhileenn et all..

I struction into forest tracts, may help slow dispersal of pools in response to earthwormpopulation manipulationsin
exotic earthworm species. The decline in some under- agroecosystemswith differentN sources.SoilBiolBiochem29:

361-67.

story plant species following invasion is due to the co- Bohlen PJ,GroffmanPM, FaheyTJ,et al. 2004a. Ecosystemconse-

occurrence of other factors, such as high deer densities quences of exotic earthworm invasion of north temperate
(Augustine et al. 1998). In this situation, control of earth- forests.Ecosystems7: 1-12.
worms, even if it were possible, may not be sufficient to Bohlen PJ, Groffman PM, Fahey TJ, et al. 2004b. Influence of
enhance recolonization of understory herbs without earthworminvasionon redistributionand retention of soil car-
efforts to control deer densities.
bon and nitrogen in northerntemperateforests.Ecosystems7:
13-27.

Northern temperate forests are facing rapid environ- BrundrettMC and KendrickB. 1988. The mycorrhizalstatus,root
mental transformations that include climate change, anatomyand phenologyof plants in a sugarmapleforest.CanJ
species invasion, changes in land-use patterns, and con- Botany66: 1153-73.
tinued atmospheric deposition. Earthworminvasion is an
additional factor that may interact with, and contribute BrussaardL. 1999. On the mechanisms of interactions between

earthwormsand plants.Pedobiologi4a3: 880-85.
BurtelowAE, Bohlen PJ, and GroffmanPM. 1998. Influence of

to, these changes. Positive or negative interactions may exotic earthworminvasion on soil organic matter, microbial
occur, such as an acceleration of C loss (climate warming biomass and denitrification potential in forest soils of the
and earthworm invasion), increased gaseous or hydro- northeasternUnited States. ApplSoilEcol9: 197-202.
logic flux of N (atmospheric deposition and earthworm DominguezJ, Bohlen PJ,and ParmeleeRW.Effectsof earthworms
invasion), or nutrient stress (drought or pollution and on movement of nitrogen into groundwaterin agroecosystems.
earthworm invasion). The increasing length of growing EcosystemsI.n press.
seasons and enhanced productivity currentlyoccurring at EdwardsCA (Ed). 2004. Earthwormecology,2nd edn. BocaRaton:
northern latitudes (Papadopol 2000; Zhou et al. 2001) are CRC Press.
likely to contribute to a more rapidnorthwardexpansion
of earthworm populations and introductions of exotic EdwardsCA and Bohlen PJ. 1996. Biology and ecology of earth-
species formerly limited by colder temperatures. worms,3rded. London,UK: Chapmanand Hall.
Earthworminvasions are likely to interact with other bio-
logical invasions occurring concurrently or predicted to FiskMC, FaheyTJ,GroffmanPM,etal. 2004. Earthworminvasion,
occur in forest ecosystems (Hansen et al. 2001). These fine root distributionand soil respirationin hardwoodforests.
invasions will have important consequences, not only for Ecosystems7: 55-62.

FrancisR and Read DJ. 1994. The contributions of mycorrhizal
fungito the determinationof plant communitystructure.Plant
Soil 159: 11-25.

GroffmanPM, Bohlen PJ,FiskMC, et al. 2004. Exotic earthworm

invasion and microbial biomass in temperate forest soils.

biodiversity and conservation, but also for nutrient Ecosystems7: 45-54.
cycling and ecosystem processesin northern forest ecosys- Gundale MJ. 2002. The influence of exotic earthwormson soil
tems.
organichorizon and the rarefern Botrychiummormo.Conserv
Biol 16: 1555-73.

HaimiJ and Huhta V. 1990. Effectsof earthwormson decomposi-

* Acknowledgements tion processesin rawhumusforestsoil: a microcosmstudy.Biol
FertSoils10: 178-183.

Some of the research reported here was supported by Hale CM. 2004. Ecologicalconsequencesof exotic invaders:inter-
grants from the National Science Foundation (DEB- actions involving Europeanearthwormsand native plant com-
9726869 and DEB-0075236). The ideas presented in this munitiesin hardwoodforests(PhD dissertation).St Paul,MN:
paper benefited from discussions at a workshop on earth-
Universityof Minnesota.
Hale CM, FrelichLE,and Reich PB. Exotic Europeanearthworm

worm invasions sponsored by the National Science invasiondynamicsin northernhardwoodforestsof Minnesota,
Foundation and the Institute of Ecology at the University USA. EcolAppl.In press.
of Georgia and organizedby Paul Hendrix. We thank Tim Hansen AJ, Neilson RP, Dale VH, et al. 2001. Global change in
Fahey, Melany Fisk, Lee Frelich, Derek Pelletier, Peter
forests: responses of species, communities, and biomes.
Bioscience51: 765-79.

Reich, and Esteban Suarezfor their contributions to the Hendrix PF and Bohlen PJ. 2002. Exotic earthworminvasions in

work presented here. North America:ecological and policy implications.Bioscience
52: 801-11.

* References Hendrix PF, Peterson AC, Beare MH, and Coleman DC. 1998.

AberJD,NeilsonRP,McNultyS, etal.2001.Forestprocesseasnd Long-termeffectsof earthwormson microbialbiomassnitrogen
globalenvironmentaclhange;predictingtheeffectsof individ- in coarseand fine texturedsoils. ApplSoilEcol9: 375-80.
ualandmultiplestressorsB.ioscienc5e1: 735-51. Hobbie SE, NadelhofferKJ,and HogbergP.2002. A synthesis:the
role of nutrients as constraints on carbon balances in boreal
AlbanDH andBerryEC.1994.Effectsof earthworminvasionon
morphologyc,arbonandnitrogenof a forestsoil.ApplSoilEcol and arcticregions.PlantSoil242: 163-70.
1:243-49. JamesSW. 1995. Systematics,biogeography,and ecology of nearc-

AugustineDJ,FrelichLE,andJordonPA. 1998.Evidencefortwo tic earthwormsfromeastern,central, southernand southwest-
alternatestablestatesin anungulategrazingsystemE. colAppl ern United States. In: Hendrix PF (Ed). Earthwormecology
8:1260-69. and biogeographyin North America. Boca Raton, FL:Lewis
Publishers.
BaskinCCandBaskinJM.1998.Seeds:ecologyb, iogeographaynd
evolutionof dormancyand germinationS. an Diego, CA: JamesSW and CunninghamMR. 1989. Feedingecology of some
Academic Press. earthwormsin Kansas tallgrass prairie. Am Midl Nat 121:

BlairJM, ParmeleeRW,Allen MF et al. 1997. Changes in soil N 78-83.

JamesSW and Hendrix PF 2004. Invasion of exotic earthworms

wwwwww.f.frroonntiteiresrisnienceocloogloy.goyr.gorg ?? TThheeEEccoollooggicicaalSSloocciieettyyooff AAmmeerricicaa

P1JBoohhileenn et al. Innvvaassiviveeeaarrtthhwwoorrmms sinn tteemmppeerraatetefoorreessttss

I

into North America and other regions. In: EdwardsCA (Ed). PapadopolCS. 2000. Impacts of climate warming on forests in I
Earthwormecology,2nd ed. Boca Raton, FL:CRC Press. Ontario: options for adaptationand mitigation. ForestChron
JohnsonNC, TilmanDG, and Wedin D. 1992. Plant and soil con- 76: 139-49.
trolson mycorrhizaflungalcommunities.Ecology73: 2034-42.
Knight D, Elliot PW,AndersonJM, and Scholefield D. 1992. The Pare D and Berier B. 1989a. Origin of phosphorusdeficiency
role of earthwormsin permanentmanagedpasturesin Devon observed in declining sugar maple stands in the Quebec
England.SoilBiolBiochem24: 1511-17. Appalachians.Can] ForestRes 19: 24-34.
Kostel-HughesE 1995. The role of soil seed banksand leaf litter in
the regeneration of native and exotic tree species in urban PareD and Bemier B. 1989b. Changes in phosphorusnutrition of
forests(PhD dissertation).Bronx,NY:FordhamUniversity. sugar maple along a topographic gradient in the Quebec
KotarJ, KovachJA, and BurgerTL. 2002. A guide to forestcom- Appalachians.CanJ ForestRes 19: 132-34.
munities and habitat types of northern Wisconsin, 2nd edn.
Madison, WI: University of Wisconsin Departmentof Forest ParkinsonD, McLeanMA, and Scheu S. 2004. Impactsof earth-
Ecologyand Management. wormson other biota in forest soils, with some emphasison
LangmaidKK. 1964. Some effectsof earthworminvasion in virgin cool temperate montane forests. In: Edwards CA (Ed).
podsols.CanJ SoilSci44: 34-37. Earthwormecology,2nd edn. Boca Raton, FL:CRC Press.
Lavelle P, BrussaardL, and Hendrix P (Eds.). 1999. Earthworm
Proulx N. 2003. Ecological risk assessment of non-indigenous
management in tropical agroecosystems.New York: CABI earthwormspecies.St Paul,MN: US FishandWildlifeService,
International. InternationalAffairs,Division of Scientific Authority.

LawrenceB, FiskMC, FaheyTJ,et al. 2003. Influenceof nonnative Saetre P. 1998. Decomposition, microbial community structure,
earthwormson mycorrhizalcolonization of sugarmaple (Acer and earthworm effects along a birch-spruce soil gradient.
saccharum)N. ew Phytol157: 145-53. Ecology79: 834-46.

Leck M. 1989. Ecology of soil seed banks. San Diego, CA: Scheu S. 2001. Plantsandgeneralistpredatorsas linksbetween the
Academic Press. below-groundand above-groundsystem. Basic Appl Ecol 2:
3-13.
LorangerG, PongeJF,BlanchartE, and Lavelle P. 1998. Impactof
earthwormson the diversity of microarthropodsin a vertisol Scheu S. 2003. Effectsof earthwormson plantgrowth:patternsand
(Martinique).BiolFertSoils27: 21-26. perspectives.Pedobiologi4a7: 1-11.

MaerzJC, KaruzasJM, Madison DM, and Blossey B. Introduced Scheu S and ParkinsonD. 1994. Effectsof earthwormson nutrient
invertebrates are important prey for a generalist predator.
DiversDistrib.In press. dynamics,carbon turnoverand microorganismsin soils from
cool temperateforestsof the CanadianRockyMountains- lab-
McKaneRB, RastetterEB,ShaverGR, et al. 1997. Reconstruction oratorystudies.ApplSoilEcol 1: 113-25.
Scheu S and WoltersV. 1991. Influenceof fragmentationand bio-
and analysisof historical changes in carbon storage in arctic turbationon the decomposition of carbon-14-labelledbeech
tundra.Ecology78: 1188-98. leaf litter.SoilBiolBiochem23: 1029-34.
McLeanMA and ParkinsonD. 1998a. Impactsof the epigeicearth-
wormDendrobaenaoctaedraon oribatidmite communitydiver- SuarezER, FaheyTJ, GroffmanPM, et al. 2004. Effectsof exotic
sity and microarthropodabundances in pine forest floor: a earthwormson soil phosphoruscycling in two broadleaftem-
mesocosmstudy.ApplSoilEcol7: 125-36. perateforests.Ecosystems7: 29-44.
McLean MA and Parkinson D. 1998b. Impacts of the epigeic
earthwormDendrobaenaoctaedraon microfungalcommunity Tiunov AV and Scheu S. 2000. Microbialbiomass,biovolume and
structurein pine forestfloor- a mesocosmstudy.ApplSoilEcol
8: 61-75. respirationin LumbricusterrestrisL. cast materialof different
age. SoilBiolBiochem32: 265-75.
McLeanMA and ParkinsonD. 2000a. Fieldevidence of the effects VitousekPM. 1990. Biological invasionsand ecosystemprocesses:
toward an integration of population biology and ecosystem
of the epigeic earthwormDendrobaenaoctaedraon the micro- studies.Oikos57: 7-13.
fungal community in pine forest floor. Soil Biol Biochem32:
351-60. WardleD. 2002. Communitiesand ecosystems:linking the above-
ground and belowground components. Princeton, NJ:
McLeanMA and ParkinsonD. 2000b. Introductionof the epigeic PrincetonUniversity Press.
earthwormDendrobaenaoctaedrachanges the oribatidcommu-
nity and microarthropodabundancesin a pine forest.SoilBiol WickenbrockL andHeislerC. 1997. Influenceof earthwormactiv-
Biochem32: 1671-81.
ity on the abundance of Collembola in soil. Soil Biol and
Migge S. 2001. The effect of earthworm invasion on nutrient Biochem29: 517-21.
turnover, microorganismsand microarthropodsin Canadian
aspen forest soil (PhD dissertation). Darmstadt, Germany: WoltersV andJoergensenRG. 1992. Microbialcarbonturnoverin
Technische Universitat Darmstadt. beech forest soils workedby Aporrectodeacaliginosa(Savigny)
(Oligochaeta:Lumbricidae).SoilBiolBiochem24: 171-77.
Mortensen S and Mortensen CE. 1998. A new angle on earth-
worms.The Minnesota ConservationVolunteer,July-August: Wood T, BormannFH, andVoigtGK. 1984. Phosphoruscycling in
20-29. a northern hardwoodforest:biological and chemical control.
Science23: 391-93.
Neilson GE and Hole FE. 1964. Earthwormsand the development
of coprogenousAl horizonsin forestsoils of Wisconsin.SoilSci Yanai RD. 1992. Phosphorusbudget of a 70-year-old norther
SocAm Pro28: 426-30. hardwoodforest.Biogeochemistr1y7: 1-22.

Zhou L, Shabanov NV, Myneni RB, et al. 2001. Variations in
northernvegetation activity inferredfromsatellite dataof veg-
etation index during1981 to 1999.J GeophysRes- Atmos106:
20069-83.

? TheEEccoollooggiiccaaSSlloocciieettyyofAmerica wwwwww..ffrroonntiteiresrisnienceocloogloy.goyr.gorg